MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

J.Haible's Living VCO for Eurorack [Build Thread]
MUFF WIGGLER Forum Index -> Music Tech DIY  
Author J.Haible's Living VCO for Eurorack [Build Thread]
logicgate
Yesterday, after 3 months after ordering, I've finally received the PCBs. As I promised, here is the build thread. Very simple, nothing fancy.


Start with the SMD capacitors cause it would be almost impossible later so, don't forget! A good technique is to put some solder in one of the pads first, only a tiny bit so you can fix the capacitor using tweezers. Then, solder the other side and add a bit more to the other (side).




Then you can move to the diodes, resistors, beads and IC sockets (if using them). You can see that I forgot to solder one diode lol . IT's VERY IMPORTANT to leave the TEMPCO resistor for last, after you finished everything!!!





After that, you can pretty much solder all the rest, besides the pots. I would start with the small capacitors, then the trimmers/pin headers for power and panel wiring/ big caps. You can see I used two film caps for C7 and C8, but you can also use Non-Polar Elecs. Actually, the original BOM of the Saw/Tri converter calls for them. I tested with both and I couldn't perceive any difference.





Now it's time for a good wash, I always clean the solder side with isoprop alcohol or acetone, and afterwards scrub with warm water and detergent, then we get a nice, clean and shiny solder side like in the pic below. It's a very good time now to pick up a magnifying glass and check for cold solder joints or shorts.





For the panel wiring, I like to use pin headers and those 24AWG dupont cables you can find on ebay for cheap, it's easier to disassemble if needed:





The wiring and calibration are pretty much self-explanatory, just wire all grounds together and the signals outs and ins from the PCB to jack signal tips. If you want you can wire the FM/PWM input as a passive attenuators.


The calibration can be found here:

http://www.jhaible.com/living_vcos/jh_living_vcos.html


The v/oct trimming is done by ear, everything actually is down to taste, besides the pulse width:

"Pulse width adjust. With the front panel Pulse Width knob in 12 o'clock position, adjust the trimmer for 50% pulse width."


Also, attached to this post is the BOM made by Barcode which is very good. And of course, feel free to post here your build pics too!!! Would be awesome. I still have to wire mine and build a couple more so I can post some videos or soundcloud tracks.


I hope you all enjoy it! nanners
Randy
Fantastic, thanks logicgate! I'm looking forward to starting this one.

Randy
logicgate
Randy wrote:
Fantastic, thanks logicgate! I'm looking forward to starting this one.

Randy



You're welcome!


I would recommend going slowly with the build. It's very dense/crowded so it's easier to make a solder bridge by mistake.

Diptrace component's footprints are misleading if you never used it before! Only after receiving the PCBs I can see how close the transistor pads are! Very close, be extra careful and soldering the transistors.

The resistors are very close together too, but it was necessary to make the PCB fit the cheaper size option. Anyways, as you can see, that wasn't a problem.
Maco
applause
logicgate
Finally got some time this morning to drill my panels, I hope that tonight I'll have them wired and running!

Just wanted to show you guys how I'm making my standalone LVCOs:






I only soldered the detune pot and used it to fix the PCB to panel.
Skaput
Awesome stuff Logicgate, love the Pittsburgh look SlayerBadger!
mush
About the transistors when the pads are tight. Always start with the middle leg, then you can put the solder iron on the outside of the other legs to minimize the risk of solder-bridges.
diablojoy
Quote:
IT's VERY IMPORTANT to leave the TEMPCO resistor for last, after you finished everything!!!

Hmm my preferred method on tempco's is to mount them under the chip
just cut 2 lengths of SIL pins and instal a standard DIL socket in them to get the spacing right then solder the SIL pins in and remove the socket. Tempco runs between the SIL pins as drawn on silk screen and is soldered in so that the top of it sits in line with the top of the SIL pins
this way it sits below the chip itself once the chip is installed hard up against the bottom of the chip.

couple of benefits to this
the tempco is kept closer to the temperature of the chip as it's less affected by droughts and replacing the chip itself if necessary is easier as there is nothing to desolder .
if you look at logicgate's 'picture it should look similar to how he has done all of his 8 pin DIL packages , must have been out of 8 pin DIL sockets I guess
logicgate
diablojoy wrote:
Quote:
IT's VERY IMPORTANT to leave the TEMPCO resistor for last, after you finished everything!!!

Hmm my preferred method on tempco's is to mount them under the chip
just cut 2 lengths of SIL pins and instal a standard DIL socket in them to get the spacing right then solder the SIL pins in and remove the socket. Tempco runs between the SIL pins as drawn on silk screen and is soldered in so that the top of it sits in line with the top of the SIL pins
this way it sits below the chip itself once the chip is installed hard up against the bottom of the chip.

couple of benefits to this
the tempco is kept closer to the temperature of the chip as it's less affected by droughts and replacing the chip itself if necessary is easier as there is nothing to desolder .
if you look at logicgate's 'picture it should look similar to how he has done all of his 8 pin DIL packages , must have been out of 8 pin DIL sockets I guess



Interesting method, never thought of that. Haha and yes, I'm out of 8 pin sockets, but I got loads of socket strips.

Actually there's no need to desolder anything, you just need to slide the tempco resistor to one side and replace the IC and slide back on top of it again, I had done it before. You just need to solder it "not tight" against the IC, leave some room so you can press it down. Do you apply thermal paste under the IC?
logicgate
@skaput

Haha thanks!! Ghetto pittsburgh look lol


@mush

Good technique! It's also good to spread the outer legs.
diablojoy
Quote:
Do you apply thermal paste under the IC?

No I stopped using it a while ago ,most are conductive and they tend to creep over time which can cause really difficult to trace faults. I just try to make sure there is some physical contact .
Having the SIL pins surrounding the tempco pretty much keeps the chip and tempco at the same ambient temperature better which is really all you want.
raisinbag
looks great! BOM is totally painless. Will the Anekom tempcos work in this bad boy?
logicgate
@raisingbag

Yep they will work just fine w00t
logicgate
Finally all wired up and calibrated!

Gonna try to record something and post here!




raisinbag
Looks great me big knob! I'm so excited to build mine.
sduck
It looks like we have 2 threads that are build threads for these - why don't we consolidate? repost anything that's recently been posted here into the other thread, and I'll lock up this one (or vice versa - you tell me).
logicgate
sduck wrote:
It looks like we have 2 threads that are build threads for these - why don't we consolidate? repost anything that's recently been posted here into the other thread, and I'll lock up this one (or vice versa - you tell me).



Good Idea.

I just think that the other thread has so many pages and people can get lost and don't find what they are looking for.

Anyways, I don't think it's up to me to decide, maybe more people could chime in.
sammy123
As long as the other is still available I am fine with closing it to new posts.
sammy123
So I have a bit of a problem. I have no 22uf electros for c9 and c10 of the vco boards. Can I use a 47uf or 33uf? Thanks.
sduck
Ok, a little later I'll lock up the old one and provide a pointer to this one - unless anyone has objections. One of them should be shut, having two threads going with build information is bad form. Personally I'd choose closing this one, as there's very little in it, but whatever.

Or I could just leave them both open, and you guys could try to direct build queries here, and board sourcing and such questions to the other thread. After all, there are numerous TTSH threads, which works fine as long as they have different content. Whatever...
logicgate
sammy123 wrote:
So I have a bit of a problem. I have no 22uf electros for c9 and c10 of the vco boards. Can I use a 47uf or 33uf? Thanks.



You can use anything between 22uf and 100uf, it's just power filtering.



@sduck

That sounds like a good plan hihi
sammy123
thanks logicgate
sammy123
So I have this up and running and have a couple questions.

-My detune pot seems backwards as well on all 3 VCOs. It is indeed very subtle.

-The triangle seems to be affected by the PWM cv. Does anyone else have this issue?
logicgate
sammy123 wrote:
So I have this up and running and have a couple questions.

-My detune pot seems backwards as well on all 3 VCOs. It is indeed very subtle.

-The triangle seems to be affected by the PWM cv. Does anyone else have this issue?



Hi sammy,

If your boards are the barcode edition then you have to apply the fix for the detune pots. From what I gather it's indeed subtle, it's not intended to work as a "fine tune" feature, but only to, well, detune them a little bit for some pleasant frequency beating (some lush "phasiness" by the way).



I have just tested here and the triangle and saw are not affected by PWM cv nor PW pot, you must have a solder bridge somewhere. I checked the schematics and there's no interation between the two circuits, the only thing that goes to the SAW to TRI circuit is a copy of the SAW signal.

Look around these components, they are the SAW to TRI circuit:

R52, R53, C7, TL074, R54, R55, R56, D5, D6, R57, R58, C8, R59, R60


Good luck, and if you have any questions just ask! Guinness ftw!
sammy123
I knew there was a coarse tune fix for the driver, but I dont recall seeing any detune fix for the cores. I'll do some more research.

The PWM just bends the tops of my triangles. It affects all 3 VCOs so I would thinks its more than a solder bridge. ..but of course I'll check to be safe.

Thanks logicgate.

In other news. I only had 2k tempcos so I just socketed some standard resistors and it has been pretty stable. Also for those doing a setup similar to the Barcode panel, buy side adjust trimmers for scale and offset...it will make life a bit easier...especially if you try and tune it while panelled. Thedriver is pretty much impossible to tune while panelled. I really wish tayda sold the side adjust. Otherwise tuning was easy...but I've done it a bunch before.
logicgate
@sammy

Oh is it only for the driver? TBH I just remember that there was a fix for the detune pot, just not sure if for driver only...


Well, I have just tested here and the detune pot works as it should. Tuning the oscillators with detune pot FCCW and then slowly turning it CW, to me , , , sounds as expected, very subtle detuning.


Do you have diptrace? I could send you the schematic and PCB files, it makes the debug easier.


I have attached to this post an image with the silkscreen, might be of help.
tojpeters
I think it comes down to the definition of detune.
I think of detuning as moving away from pitch,sharp or flat.
In that definition the pots are backwards.
The way you describe detune as being in the flat direction only they are correct.
logicgate
Hey guys,

About the R39 fix:

It seems that the best way to find the correct value for your build is by using sockets at R39 and keep swapping resistors (between 820K and 1M) until you find the right for you, since the resistances of each build will be a bit different and the PW is very picky.


I say that because I used 820K on my other two LVCOs and found out that the square wave still disappears when PW pot is FCCW, so I guess I will need a higher value for the other two, but not much.
sammy123
Logicgate, even if you adjust the trimmers? I put in 820Ks and then adjusted r40 I think. It's not perfect though. I didn't even try that before I out in the 820Ks.
logicgate
sammy123 wrote:
Logicgate, even if you adjust the trimmers? I put in 820Ks and then adjusted r40 I think. It's not perfect though. I didn't even try that before I out in the 820Ks.



Yep even after trimming...
logicgate
Just so you guys know, what I end up using at R39 on my other two LVCOs is a 1M5 resistor, now I get the full range of the square wave on them too and PW pot at FCCW position doesn't make the wave disappear.
sammy123
Cool. Glad you got it.
logicgate
Hey guys! Today I had some free time to fool around with my modular and recorded some audio. This is two LVCOs going into the yusynth minimoog filter clone. From the modular output into the crappy notebook mic in, but it's not that bad, actually.

Frequency of the LVCOs is being modulated by the turing machine, quantized by the Intellijel uScale. Filter cutoff is modulated by the 4ms PEG.

[soundcloud url="https://api.soundcloud.com/tracks/167177852" params="auto_play=false&hide_related=false&show_comments=true&show_use r=true&show_reposts=false&visual=true" width="100%" height="450" iframe="true" /]
LED-man
i want build the LVCO in MOTM and would start in order a MOTM panel from schaeffer.
But i have seen on many eurorack panels in the LVCO thread most panels have SIN out.
I´m wondering from where they have the signal - i dont find on the pcb pictures a SIN OUT connector.

further i miss here the schematics and BOM for the driver, one place for all documentation files - maybe on first page would be very helpful.
logicgate
Barcode, you are summoned to post here the driver related files lol


@LED-man

I know that you can check the driver schematics at Haible's website:

http://www.jhaible.com/living_vcos/jh_living_vcos.html


The tri to sin converter I don't remember from where he got the circuit.
tojpeters
MFOS
Barcode
Attached is the tri to sine I used. When I was working on the layout I did a quick search and found that one. It was simple enough.

I will work on getting the driver schematic exported today.
bsmith
Is there a bom for the driver board somewhere?
edit:
found on barcode's site....
http://www.baud-rate.com/blog/wp-content/uploads/2014/05/Living-VCO-Dr iver-BoM-v1.1.pdf
Barcode
Here are the BoM's for the Barcode Edition of the VCO Core and the Driver.
LED-man
@Barcode:

have seen you found 2 failures on the pcb
shown on your website http://www.baud-rate.com/blog/

are this Isues on production pcbs too ?
Barcode
All of the issue I reported have been fixed on the production PCB's. thumbs up
LektroiD
Boards arrived today and look awesome. Anyone planning to do Eurorack panels for the full PCB set?
bsmith
so.. going off the bom - those 100n 0504 smt caps - having difficulty finding them on mouser - there don't seem to be any .1uf 0504 - any clues out there?
LED-man
bsmith wrote:
so.. going off the bom - those 100n 0504 smt caps - having difficulty finding them on mouser - there don't seem to be any .1uf 0504 - any clues out there?


Its a typo. Pls. Use 805
cleaninglady
Just looking at the components for the LVCO

I was wondering about a source for the SMT Tempco ?
LED-man
cleaninglady wrote:
Just looking at the components for the LVCO

I was wondering about a source for the SMT Tempco ?


Mouser sells this.
Panasonic
diablojoy
I am just using thru hole ones. I have a bunch of them
but I will run them under the chip just replace the socket with SIL strip socket
instead once the chip is inserted the bottom sits hard against the tempco
no thermal grease needed either.
sceledra
So I mistakenly made a post like this in the wrong place on the forum so I'm going to put it in the right place this time, fingers crossed. Would someone who has completed this build, be willing to share a Mouser cart and/or other order instructions for purchasing all of the needed components for this build? I do not have the background knowledge of parts to know for fact what I'm ordering is what is necessary. The last thing I want to do is order parts and then find out that all the resistors that I've ordered are 1% and they are supposed to be 5% and I didn't order the right ones because you're supposed to "know" that for this type of build that's what you need. Alternatively if someone would like to assuage my fears of this by saying that other than what the BOM actually states for specs on a part you don't have to be concerned with other specs for the parts. I'd be more than willing to compensate you for the time you spent compiling the info and responding if you care to share with me. Thanks
cleaninglady
diablojoy wrote:
I am just using thru hole ones. I have a bunch of them
but I will run them under the chip just replace the socket with SIL strip socket
instead once the chip is inserted the bottom sits hard against the tempco
no thermal grease needed either.


Cheers man , sounds good. SlayerBadger!

Do you have a certain technique for achieving optimum height of the Tempco for contact with the Chip ?
raisinbag
If I'm reading you right, the last few times I put a tempco over a chip I soldered it in after, plopped a glop of white stuff on chip, and soldered in place touching the top of IC. You can push the tempco off to the side ifyou ever need to remove chip. But if you solder it in after, it is easy to get height. I you do it under id use two strips of sil machied sockets and solder tempco just a wee bit above the top of those so chip will push against it when you pop the chip in sockets.
diablojoy
Quote:
I you do it under id use two strips of sil machied sockets and solder tempco just a wee bit above the top of those so chip will push against it when you pop the chip in sockets.


correct
also I found if you are using the small tempco's available from thonk you can just use a normal socket in which case install the socket then the tempco
the tempco leads will straddle the ends of the socket which are lower and places the tempco at almost the perfect height to push against the underside of the chip, I don't use thermal paste but if you do make sure it specifies it is not electically conductive,- most are and they also tend to creep over time that may cause a world of pain trying to track down a fault months later.
bsmith
So for that driver lm13600 - where are y'all sourcing this? Found this on mouser - pinout is the same, should it work fine?
http://www.mouser.com/ProductDetail/NJR/NJM13600D/?qs=sGAEpiMZZMuUbyQT l9BuV%252bWkQ1cVapN4VDou46f5Gn8%3d
sammy123
I ordered a bunch from ebay.
LektroiD
bsmith wrote:
So for that driver lm13600 - where are y'all sourcing this? Found this on mouser - pinout is the same, should it work fine?
http://www.mouser.com/ProductDetail/NJR/NJM13600D/?qs=sGAEpiMZZMuUbyQT l9BuV%252bWkQ1cVapN4VDou46f5Gn8%3d


Use LM13700, basically the same OTA
fuzzbass
Just have to say, Dymo Labelmate applause ! It reaffirms my faith in humanity that P-Touch did not put them out of business. I'm sure its toxic as hell, but the adhesive on these things is magical. You see panels made in 1960s with these things still clinging like mad. Wouldn't that be great if there were some on the Voyager spacecraft? They would still be stuck on after drifting through the cosmos for a billion years, and the ETs would no doubt be impressed.
olga42
bsmith wrote:
So for that driver lm13600 - where are y'all sourcing this? Found this on mouser - pinout is the same, should it work fine?
http://www.mouser.com/ProductDetail/NJR/NJM13600D/?qs=sGAEpiMZZMuUbyQT l9BuV%252bWkQ1cVapN4VDou46f5Gn8%3d


Some of the NJM-manufactured ICs were 2nd or 3rd sources of more or less standard analog components so these are the same as LM13600. NJR also make their 13700 variant, or at least used to.
qfactor
So LM13600 can really be swapped with the LM13700? hmmm.....
Coz I have a bunch of 13700s in hand nanners
LED-man
i want order in next weeks from Schaeffer a LVCO MOTM Panel.
pictures here:
https://www.dsl-man.de/display/MAIN/Juergen+Haible+Living+VCO+in+MOTM

1-4 users: each 245€
5 -10 Users each 220€
10 users each 196€
20-29 users each 171€
bsmith
thanks for the pointers on the 13600 being more or less the same as the 13700, have those, great.
So - trimmers and board mounting...
How are y'all going to go about this? There are a fair amount of trimmers on these dudes and the orientation on some is such that even side mounted adjustments will not really be reachable if the boards are mounted off the panel by the pots. So seems there are a few options.
a) just calibrate as they are built/mounted, and if future adjustments are needed un panel and deal with it..
or
b) Don't panel mount the boards but sandwich them in two pairs and use wires for the panel pots and come up with some bracketology - this allows board mounting the trimmers for easy access when needed...
or....
c) Fly trimmers out to a piece of stripboard and mount that off the back and mount the panel off of pots anyhow.

i'm leaning towards 'b'.
what have y'all who have completed the build done?
tojpeters
I used plan b
qfactor
Hi,

I sourced the CA3046 chip to Futurelec, so just wanted to be sure that this should be able to work in the LVCO, right?
I mean you won't need a specific type of 3046 for it to work as planned, right? hmmm.....
raisinbag
I don't think you need anything special in a suffix. I don't think I have ever seen that noted in any 3046 builds. I actually just found a bunch of 3046's that aren't even CA, I think they are TL? But I plan on using them and doubt there is any notable difference. But I have been wrong once or twice in my life. hihi
nigel
qfactor wrote:
Hi,

I sourced the CA4036 chip to Futurelec, so just wanted to be sure that this should be able to work in the LVCO, right?


Just a heads up - I cancelled my last Futurlec order after about three months, when they still hadn't sourced most of it. (And had constantly ... let's say, made inaccurate estimates of when things would be available.)
raisinbag
Bride chamber has them if you are in a pinch.
qfactor
nigel wrote:
qfactor wrote:
Hi,

I sourced the CA4036 chip to Futurelec, so just wanted to be sure that this should be able to work in the LVCO, right?


Just a heads up - I cancelled my last Futurlec order after about three months, when they still hadn't sourced most of it. (And had constantly ... let's say, made inaccurate estimates of when things would be available.)


I know what you mean as far as Futurelec goes! sad banana
This would be my 3rd time using them, the 1st two times, it was "nail bitingly" long for my order to arrive.
Only reason is I can't find this at my "usual" online places (Tayda, Small Bear) but at least now I can also try Bridechamber's! Rockin' Banana! Rockin' Banana!
Thanks!
khakifridge
qfactor wrote:
I cancelled my last Futurlec order after about three months . . .


I get the feeling that they're called Futurelec because that's when your stuff arrives. And of course tomorrow never comes... confused
LeftyLogic
So while it unfortunately isn't of a whole lot of use with this project, I was looking around on Mouser and it turns out that TI is now making 3046's again! Unfortunately, it looks like they're only making them in surface mount as opposed to through hole, but here's the part number anyway: 926-LM3046M/NOPB. I suppose that if worst comes to worst and you couldn't find any anywhere else you might be able to scrounge up an SOIC to DIP adapter, but I wouldn't really call that an ideal solution.
It really stinks that Tayda's run out. sad banana

Seth
sduck
utsource - I've used them for a lot of stuff and everything's worked.
diablojoy
Quote:
utsource - I've used them for a lot of stuff and everything's worked.

same here
fuzzbass
re ca3046 ... I have twice obtained them from ebay seller iseetheworld and in both cases received good OEM Harris chips. They are working in my TTSH. As with all ebay transactions, you take a risk. But this was my experience.

Treat these chips with care, they are fragile and especially heat sensitive. Socket them.
CJ Miller
I eagerly look forward to starting on mine, but I need to wait to order the SMT caps. My 1206s are too large to fit, and I am too desperately broke to order parts anything soon. It drives me nuts to hold off building for a dollars worth of parts.
qfactor
Bah!! seriously, i just don't get it
Futurelec just got back to me that they've run out of CA3046 and offered a refund as they are expecting difficulty in re-stocking these chips anytime soon! sad banana

So I'm going with this guy, iseetheworld on eBay, as mentioned by fuzzbass here. He's selling 10 chips for about US$17, including shipping.
Now I wait..... sad banana


fuzzbass wrote:
re ca3046 ... I have twice obtained them from ebay seller iseetheworld and in both cases received good OEM Harris chips. They are working in my TTSH. As with all ebay transactions, you take a risk. But this was my experience.

Treat these chips with care, they are fragile and especially heat sensitive. Socket them.
bsmith
I've got the 3046's squared away (if that planets align correctly I hope to finally hear my build this weekend, halleluja) but just curious, would a ca3086 work in lieu of them? Seems like I saw Yves or someone state as much over on e-m....
lysander
Might work but the differential pair in the 3086 is not matched so the V/Oct performance might be negatively affected.
sammy123
So has anyone been able to squeeze precision pots for the frequency? I have been looking and the smallest i see are about 3/4" at the base. Probably too tight to wire in place of a pcb mounted 9mm alpha with Barcodes panel. It would be nice though because the tuning on this is tough...not that it needs to be perfect, but small adjustments are difficult.
LeftyLogic
sammy123 wrote:
So has anyone been able to squeeze precision pots for the frequency? I have been looking and the smallest i see are about 3/4" at the base. Probably too tight to wire in place of a pcb mounted 9mm alpha with Barcodes panel. It would be nice though because the tuning on this is tough...not that it needs to be perfect, but small adjustments are difficult.

I'm not sure about Barcode's panel, but I did some measurements on my design, and it looks like there should be room for 10-turn pots that are up to 7/8" in diameter! (assuming that the other pots are 16mm instead of 9mm)

Seth
thetwlo
CJ i'll sort that for ya. you solder my smt caps i buy you caps? silly less than a dime. fuck it, happy to send caps to cj anyway, pm me your addy.
sammy123
LeftyLogic do you have source for 10 turns that size?
LeftyLogic
sammy123 wrote:
LeftyLogic do you have source for 10 turns that size?

7/8" seems to be a pretty standard size for 10-turn pots, and it looks like jameco carries them for pretty cheap: http://www.jameco.com/1/1/50319-3540s-1-103-vp-3540-wirewound-potentio meter-ohms-10-k.html

I'm thinking I might have to grab a few at some point for one of my builds! It's peanut butter jelly time!
Zaibach
After weeks of sourcing parts and building, my twin Living VCOs are finally complete! It's just beautiful... I owe Jurgen Haible and everyone involved big time! thumbs up I have to give Logicgate credit for my panel design and for the choice of knobs, which I more or less 'copied' because I liked it. I was not tight on space, since I started a new rack, so I just made use of two 42HP Doepfer blind panels. It's not perfect (the writing and symbols are not quite aligned right) but anyway, sounds great. I even 'crafted' a mixer (switchable between 8 inputs/2 outputs or 4 inputs/1 output each) especially for these.

I used single-turn pots for frequency, and I've got to say that it is quite possible to tune the three oscilltors to unison without too much trouble, but it is a bit unnerving sometimes, because you touch one knob accidently and you have to tune again. I was thinking about multiturns but I was tight on money. If I would build it again, I would definitely use multiturns...

qfactor
fuzzbass wrote:
re ca3046 ... I have twice obtained them from ebay seller iseetheworld and in both cases received good OEM Harris chips. They are working in my TTSH. As with all ebay transactions, you take a risk. But this was my experience.

Treat these chips with care, they are fragile and especially heat sensitive. Socket them.


Just got a bunch of these (10, in fact) from this ebayer, as you've suggested!
Got them in record time, for US$17
Now, hopefully they all turn out real!! eek!
But since you've gotten them twice from him before, should be "safe", eh?! d'oh!
Mindarp
I have a pile of bc558 and bc548. Will these work ok with this build?
LED-man
Zaibach wrote:
After weeks of sourcing parts and building, my twin Living VCOs are finally complete! It's just beautiful... I owe Jurgen Haible and everyone involved big time! thumbs up I have to give Logicgate credit for my panel design and for the choice of knobs, which I more or less 'copied' because I liked it. I was not tight on space, since I started a new rack, so I just made use of two 42HP Doepfer blind panels. It's not perfect (the writing and symbols are not quite aligned right) but anyway, sounds great. I even 'crafted' a mixer (switchable between 8 inputs/2 outputs or 4 inputs/1 output each) especially for these.

I used single-turn pots for frequency, and I've got to say that it is quite possible to tune the three oscilltors to unison without too much trouble, but it is a bit unnerving sometimes, because you touch one knob accidently and you have to tune again. I was thinking about multiturns but I was tight on money. If I would build it again, I would definitely use multiturns...




from where you get the TRI/SIN signal ?
Zaibach
Quote:

from where you get the TRI/SIN signal ?


The Tri to Sine converter is on the driver pcb. So you have to wire a triangle out from the LVCO to the panel, as well as to the converter on the driver pcb, and then the resulting sine back to the panel.
Randy
I'm having a bit of trouble finding the 100nf SMT 0504 caps. There seems to be all sorts of 100nf caps and different sizes but that combination seems to be tricky. And I'm not that familiar with SMT/SMD stuff, which doesn't help.

Anyone aware of a seller on eBay that might have these, or which part number on Mouser would work?

Thanks

Randy
bsmith
typo - use 0805... I ran into this earlier in the thread too..
https://www.muffwiggler.com/forum/viewtopic.php?p=1698136#1698136
Randy
bsmith wrote:
typo - use 0805... I ran into this earlier in the thread too..
https://www.muffwiggler.com/forum/viewtopic.php?p=1698136#1698136


I went through the thread and of course missed that! Thanks.

Randy
bsmith
and if you are wanting to go the smt route for the tempco....
http://www.mouser.com/ProductDetail/Panasonic/ERA-S33J102V/?qs=%2fha2p yFaduiamcDwRD0o%252bQMOomQv6GK8d7nJwl%252bk7XjidcJ1MaHlKg%3d%3d
qfactor
Tayda sells the 100nF SMDs too at 1 cent a piece! thumbs up
Part #: A-3511
Benjamin AM
Sorry to bring this up so late in the game. I found that the Barcode BOM v1.3 had some values that did not work my build. I feel I should share my experience in case anyone out there has had the same issue. In regards to the PWM section of the circuit, resistors R39 and R36 were troublesome for me. I came to this conclusion because I felt that the PWM wasn't working as advertised because the pulse wave was cutting out even before full CW and CCW. This circuit is not supposed to be like this. The diodes are supposed to protect the pulse from completely shutting off. This means to me that the gain and bias on the transistor at the summing stage of the PWM circuit is to high. So all that I did to change this was return R36's value to the suggested value by Haible(750k opposed to the Barcode BOM 1.5M). Then the PWM mixer resistor(R39) back to the original value 1M, instead of Barcode BOM value(510k). I also found it nice to return one of the biasing resistors(R33) back to original value 20k. Now all is good for me.
R33=20k
R39=1M
R36=750k
Benjamin AM
I noticed an error on the Barcode PCBs. The FM and PWM inputs are connected incorrectly. Currently, the inputs are routed to the middle leg of the panel pot. It should be set up so the inputs are connected to the CW leg of the pots, the summing resistor is on the middle leg and ground is on the CCW leg. This way the pot functions as an attenuator rather than gain control(current setup).
Benjamin AM

This I s how I reworked the PCB without cutting any traces. For the PWM circuit, I unsoldered one side of R38(connected to CW leg of PWMPOT) and added a length of wire to route the loose side of the resistor to the pad labeled as "PWM". You will notice in the picture that I did this with green wire and used some white heat shrink for a little protection. Now R38 is connected to the middle leg of PWMPOT. The PWM input then has to move to the CW leg of the pot. I did this by wiring the PWM input jack directly to the board. Fortunately there are two holes for 9mm and 16mm accomidations. Simply use the free hole and wire directly to the input rather than using the normal connection(MTA header in photo). In the photo you might notice that I made this connection with yellow wire. As for the FM circuit, I used the same sequence. The only difference is that the FM summing resistor is R2. It should be re-routed to the "FM" pad and the input should be hard wired to the CW of FMPOT.
logicgate
@Benjamin

My fix was:

"Hey guys,

About the R39 fix:

It seems that the best way to find the correct value for your build is by using sockets at R39 and keep swapping resistors (between 820K and 1M) until you find the right for you, since the resistances of each build will be a bit different and the PW is very picky.


I say that because I used 820K on my other two LVCOs and found out that the square wave still disappears when PW pot is FCCW, so I guess I will need a higher value for the other two, but not much."
Benjamin AM
Yes, it was good advise. My value was for R39 would have been 2M without changing the overall gain of the summing circuit. I could have left it like that but I wanted the PWM input to be not as hot so I also changed the overall gain. I could have just adjusted R38 instead of R36 but there are more resistor value options when the overall summing gain is lowered. This worked well in all three of my VCOs.
Siri
For those having got the barcode pcbs, do we have to do the modifications benjamin just described?

And the "R39 fix"?

thanks,
Mindarp
I bought what I think are the right tempcos for this project, but mine are 1/4W. Is that all right? They aren't as beefy looking as some of the other builds I've seen. I'm assuming that others have used 1/2W.
sammy123
1/4 are what I'm using.
Mindarp
Thanks sammy123
sammy123
Actually I lied....mine are 1/8th. They are the kind Synthcube and Thonk sell.
sduck
There are several varieties of 1/4w tempcos - some look like big fat black things, some look like regular sized resistors. Either will work.
Siri
Siri wrote:
For those having got the barcode pcbs, do we have to do the modifications benjamin just described?

And the "R39 fix"?

thanks,


And also, what are the smt pads close to the R11 labelling for? I thought they would be for the smt version of the tempco but they don't look to be connected at the same place than the big R11 drawing?



EDIT : d'oh! Yes they are connected, it's for the tempco. w00t
qfactor
So is the SMT an alternative to using the through hole version (of the tempco)?
negativspace
Certainly appears that way. thumbs up
Randy
"After that, you can pretty much solder all the rest, besides the pots. I would start with the small capacitors, then the trimmers/pin headers for power and panel wiring/ big caps. You can see I used two film caps for C7 and C8, but you can also use Non-Polar Elecs. Actually, the original BOM of the Saw/Tri converter calls for them. I tested with both and I couldn't perceive any difference. "

The BOM doesn't mention the film/NP electrolytics here and the PCB markings hint the polystyrene looks like a film cap. Argh! I used film there and two normal MLCC caps for C7 and C8. Is it worth desoldering them?

Randy
Siri
correct me if i'm wrong... If I use the surface mount tempco, I should not use headers for the CA3046?
LeftyLogic
That's how I did it. SMT Tempco and no socket on the 3046. Although if you want better temperature stability you will probably want to put some kind of thermal goop in between the tempco and the IC, as they don't seem to make direct contact by themselves. (I'm not doing that on my builds though. In my opinion a bit of temperature drift is part of the charm.) hihi

Seth
latigid on
Many thanks for the BOMs:


J. HAIBLE'S OSCILLATOR DRIVER Barcode Edition v1.1
J. HAIBLE'S LIVING VCO CORE Barcode Edition v1.3


Do these take into account the fixes described above?
Phetus
Do the 550s and the 560s need to be matched?
Siri
LeftyLogic wrote:
That's how I did it. SMT Tempco and no socket on the 3046. Although if you want better temperature stability you will probably want to put some kind of thermal goop in between the tempco and the IC, as they don't seem to make direct contact by themselves. (I'm not doing that on my builds though. In my opinion a bit of temperature drift is part of the charm.) hihi

Seth


Thanks Seth!!

Phetus wrote:
Do the 550s and the 560s need to be matched?


I don't think so. Looking at these BOMs and J. Haible website and his own BOM, no mention of it, only mention of the matched resistors!
Phetus
Thanks! Just wanted to make sure i hadnt missed anything...
qfactor
In the version 1.3 Driver BOM, there's a 1M pot that was labelled as Log-based.
Can I use a Linear type instead? Any major issues if that's used? hmmm.....
Thanks
Randy
latigid on wrote:
Many thanks for the BOMs:


J. HAIBLE'S OSCILLATOR DRIVER Barcode Edition v1.1
J. HAIBLE'S LIVING VCO CORE Barcode Edition v1.3


Do these take into account the fixes described above?


Not from the versions I have, same as yours.

qfactor, where did you get a version 1.3 Driver BOM? On the version 1.1 I have, the 1M pot is linear.

Randy
Siri
@Randy

on this same thread, right here, where the latest BOMs are : https://www.muffwiggler.com/forum/viewtopic.php?p=1695437#1695437

When you download the driver BOM the title says V1.1 but if you go down down down you see "BOM revision 1.3" and if you go up again a little bit, on the 1M pot it says "1M Log Pot".

razz
Randy
Argh! I downloaded the originals and then didn't look around too much because I started sourcing components. Figured I'd go back through the thread but I wouldn't have re-downloaded the files because the titles where the same.

Suppose I better compare them, since I have already populated the boards. I wonder why the portamento pot needs to be log? I already ordered linear so I'll most likely just use that.

Thanks Siri.

Randy
qfactor
Yeah, I believe the differentiation would be that the later (updated?) BOM for the Driver has the "Total" component count column added to the list. The older one doesn't.
And yes, I'm confused too as far as which version this BOM really is! d'oh! because it's stated "1.1" at the top and "1.3" in finer print, at the bottom!! eek! . So just decided to call it the later version, like the BOM for the LVCOs seriously, i just don't get it

But I guess it wouldn't do too much "damage" to swap a Lin with a Log pot, eh?
confused
Randy
I'm thinking maybe the log pot makes it easier to dial in small amounts of glide.

Randy
Siri
For those building in 4U system, here is my take on the pcb placement behind the panel.

so excited Rockin' Banana! It's motherfucking bacon yo Rockin' Banana! It's motherfucking bacon yo



LeftyLogic
Siri wrote:
For those building in 4U system, here is my take on the pcb placement behind the panel.

so excited Rockin' Banana! It's motherfucking bacon yo Rockin' Banana! It's motherfucking bacon yo


Oh that's clever!
Siri
thanks lefty logic!
for information :
long standoffs are 1" so it gives you some room.
small ones are 3/8".
total depth is under 2.5"

EDIT : Ahhhh, just realised i'm 15V and this is is a 12V version. Which resistors should I change, for which values?
haceka
i need information too for 15V build !
Thanks !
Stides
haceka wrote:
i need information too for 15V build !
Thanks !


There are a couple of substitutions on the Driver BOM for 15v, I don't see any listed on the VCO boards.
LeftyLogic
I just finished a 15v build, and aside from the changes listed on the BOM, I have a couple of suggestions: firstly, I'd recommend using 500k trimmers for the +- octave switch. I'm using 100k trimmers, and it doesn't seem to be possible to trim either one to the correct value. My other recommendation is to use some kind of socket for r36 on the VCO boards. The BOM lists a 1.5 meg resistor for r36, but I had to go all the way down to 300k to get the PWM to work properly.

Good luck!
Seth
Siri
Thanks!

From all i read in the thread, to sum up :
- sockets for R39, and swap resistors
- sockets for R36 (if 15V version), and swap resistors

Any tips on what I'm suppose to look for on the oscilloscope by swapping those resistors?

Benjamin AM wrote:
I noticed an error on the Barcode PCBs. The FM and PWM inputs are connected incorrectly. Currently, the inputs are routed to the middle leg of the panel pot. It should be set up so the inputs are connected to the CW leg of the pots, the summing resistor is on the middle leg and ground is on the CCW leg. This way the pot functions as an attenuator rather than gain control(current setup).


Any thoughts about this?

Going to pick up the 3046s at the shop this week and start calibrate Guinness ftw!
Siri
Got some help from logicgate:

Quote:
If you can get a 50% square wave with PW pot at noon, a 10% FCCW and 90% FCW, then you are good.
Remember this: you also need to adjust the trimpot after swapping the resistors, at the same time test the PW pot, it's a combination of all three.

Actually, the instructions for the PW calibration can be found at Haible's website:
http://www.jhaible.com/living_vcos/jh_living_vcos.html
You just need to add the resistors swapping part! hihi This is fun!
Ah, and remember that we can never get 100% accuracy in the analog realm, we have digital for that.
So, don't sweat too much trying to get everything "perfect" thumbs up


Also, might look obvious, but if you use the surface mount CA3046, you can only use the through hole tempco, but not the SMT. Learned it the hard way Dead Banana
Phetus
Siri wrote:
Got some help from logicgate:

Quote:
If you can get a 50% square wave with PW pot at noon, a 10% FCCW and 90% FCW, then you are good.
Remember this: you also need to adjust the trimpot after swapping the resistors, at the same time test the PW pot, it's a combination of all three.

Actually, the instructions for the PW calibration can be found at Haible's website:
http://www.jhaible.com/living_vcos/jh_living_vcos.html
You just need to add the resistors swapping part! hihi This is fun!
Ah, and remember that we can never get 100% accuracy in the analog realm, we have digital for that.
So, don't sweat too much trying to get everything "perfect" thumbs up


Also, might look obvious, but if you use the surface mount CA3046, you can only use the through hole tempco, but not the SMT. Learned it the hard way Dead Banana


Hahahah! How did you give the 3046 legs? I realised after finally succeeding using to terrible methods, that the easiest way is probably to stick a bunch of pins into the holes, solder them in and then bend them into position, plonk the 3046 on top and bobs your uncle.
Siri
using this guy :


http://www.dipmicro.com/store/PCB-SOIC-SSOP16

then the chip is too high to use the smt tempco so i tried to "througholify" the tempco by soldering legs but when i bent the legs it snapped off the tempco's extremities.

It's easier to use the through hole CA3046. hihi
raisinbag
Hey so I'm grabbing all the stuff I need that I don't have on hand, which mainly seems to be SMD stuff these days. Sooooo I'm making a digikey order for other stuff and looking for 0504 ceramic 100n caps, but they dont seem to exist there. I am a looser when it comes to SMD specs, so will any other sizes of SMD work in this case? I see 0508 (1220 metric) ones, 0402 (1005 metric), 0402 (1005 metric)....... Man I really know shit about smd sizes, someone hold me..... hmmm.....
duff
raisinbag wrote:
Hey so I'm grabbing all the stuff I need that I don't have on hand, which mainly seems to be SMD stuff these days. Sooooo I'm making a digikey order for other stuff and looking for 0504 ceramic 100n caps, but they dont seem to exist there. I am a looser when it comes to SMD specs, so will any other sizes of SMD work in this case? I see 0508 (1220 metric) ones, 0402 (1005 metric), 0402 (1005 metric)....... Man I really know shit about smd sizes, someone hold me..... hmmm.....


Pretty sure that is a typo and it should be 0805 - I think there is a post to that effect in one if the LVCO threads.
raisinbag
wicked! thanks Duff, i was feeling totally confused. d'oh!
jules
Does anyone know how deep down these are going?
Close to LFO speed... or not?
FingerTappin
I was wondering if there would be any noticeable difference between using either KRL/Bantry through hole tempcos, or the Ankaeohm through hole tempcos?

I see off the bat that the Ankaeohms are 1/6w 3300PPM, where as the KRL/Bantry are 1/4w 3500PPM.

I don't know too much about Tempcos admittedly.
qfactor
Siri wrote:
Thanks!
Benjamin AM wrote:
I noticed an error on the Barcode PCBs. The FM and PWM inputs are connected incorrectly. Currently, the inputs are routed to the middle leg of the panel pot. It should be set up so the inputs are connected to the CW leg of the pots, the summing resistor is on the middle leg and ground is on the CCW leg. This way the pot functions as an attenuator rather than gain control(current setup).


Any thoughts about this?


Indeed, any thoughts on this?!! Those who have gotten your LVCO up and running, did you have to correct this error?? seriously, i just don't get it
Seems like quite a MAJOR issue, wouldn't it? hmmm.....
Benjamin AM
Remember I have the Barcode edition PCBs. I wouldn't consider this a big issue because it is easily fixed. Essentially, It appears legs 2 and 3 of both the FM and PWM pots are swapped. Here is a picture of how I fixed this issue without cutting any traces. The PWM and FM inputs are connected via a yellow wire. These wires connect directly from the input jack to the extra CW leg of associated potentiometer. The middle leg of the pots are tapped at the nearby holes labelled PWM and FM. I solder a green wire at the middle leg tap and jump it over to the resistor leg that has been lifted off the board(R2-FM R38-PWM).
Siri
Quote:
Essentially, It appears legs 2 and 3 of both the FM and PWM pots are swapped.


So for those who are not soldering the pots straight in the pads, just need to swap the 2 and 3 connections on FM and PWM?
raisinbag
Benjamin AM wrote:
Remember I have the Barcode edition PCBs. I wouldn't consider this a big issue because it is easily fixed. Essentially, It appears legs 2 and 3 of both the FM and PWM pots are swapped. Here is a picture of how I fixed this issue without cutting any traces. The PWM and FM inputs are connected via a yellow wire. These wires connect directly from the input jack to the extra CW leg of associated potentiometer. The middle leg of the pots are tapped at the nearby holes labelled PWM and FM. I solder a green wire at the middle leg tap and jump it over to the resistor leg that has been lifted off the board(R2-FM R38-PWM).


SHITTY, I am just starting to build my VCO's and have a few from Lefty but a whack from original Group buy. So is there a trace kludge documented somewhere, if we plan on pcb mpunted pots?
Skaput
raisinbag wrote:
Benjamin AM wrote:
Remember I have the Barcode edition PCBs. I wouldn't consider this a big issue because it is easily fixed. Essentially, It appears legs 2 and 3 of both the FM and PWM pots are swapped. Here is a picture of how I fixed this issue without cutting any traces. The PWM and FM inputs are connected via a yellow wire. These wires connect directly from the input jack to the extra CW leg of associated potentiometer. The middle leg of the pots are tapped at the nearby holes labelled PWM and FM. I solder a green wire at the middle leg tap and jump it over to the resistor leg that has been lifted off the board(R2-FM R38-PWM).


SHITTY, I am just starting to build my VCO's and have a few from Lefty but a whack from original Group buy. So is there a trace kludge documented somewhere, if we plan on pcb mpunted pots?


Here is the page of the original thread original thread link

Though if you ordered from infraXpert,you wont need to do this as he waited till the Pcbs
had been confirmed and the gerbers updated, as far as i can recall.

Not sure about you folks on the other side of the atlantic though
raisinbag
Thanks man. Appreciate that. It's not so bad a fix. Just a minor pain in the ass when I have to do it 15 times!!!!! Dead Banana
idealer
Oh crap . . . now i have some resistors to desolder
buildt the barcode board v 1.1 driver last night and in the end the resistors did not sum up very well . . an example is r11 that is a trimmer on the bom but a resistor on my board

there must be a newer bom somewhere but i am not able to find it anywhere
not even on barcodes own page even if it is liked there in some threads. .

any one have a correct bom to share with me to help me out?
LeftyLogic
Hey guys! Just so you all know, I didn't make any changes whatsoever to the PCB layout for my run, so any weirdness present in Barcode's run will still be there in mine. However, I did build a set using Barcode's PCBs, and all the controls seem to work fine without any kludges.
I think really the only concern is that with the current layout it's quite easy to overload the PWM circuit and force the pulse wave into nonexistence. I'm pretty sure Benjamin AM's fix should correct this though. The FM circuit doesn't seem to be adversely affected at all, so being the lazy bum that I am, I'd say just leave those be. razz (BTW, these oscillators sound really badass when cross-modulated!)

Oh also, I can confirm that the issue with the tune pot on the driver board has been fixed, so if you ordered from Barcode or myself you shouldn't need to do any trace-cutting on that one. thumbs up

Seth
LeftyLogic
Oh, and idealer, here are the current BOMs!
raisinbag
Well that's good news dude! I have half barcodes stuff half yours (well one set from logicgate) so I'll have to really study wtf I have going on, but if no traces need cutting I'm a happy camper. Parts are on the way, so hopefully I'll have all these rocking by spring time.
LektroiD
Anyone got a wiring diagram for the 4 boards?
logicgate
LektroiD wrote:
Anyone got a wiring diagram for the 4 boards?



Here you can find 3 ways to wire up the boards:

http://www.jhaible.com/living_vcos/jh_living_vcos.html
idealer
Thanks alot Seth!!

It was not to bad... fixed it without major pain to the pcb 😀
And i figured out why i did not find the correct bom too.. i searched for it on the forum without being logged in d'oh!
LektroiD
logicgate wrote:
LektroiD wrote:
Anyone got a wiring diagram for the 4 boards?



Here you can find 3 ways to wire up the boards:

http://www.jhaible.com/living_vcos/jh_living_vcos.html


Which is the right diagram for the panel offered?

I was looking more for a panel diagram... Also, many of the abbreviations don't seem to tie in with what is written on the PCBs. Very confusing! I know if I wire it from that sketch I'll probably smoke something expensive.

Anyone got a panel wiring diagram for the actual panel offered? More in line with this sort of style...

qfactor
Must BOTH the 550 and 560 trannies be of the "C" variant for this to work? hmmm.....
Jop
Quote:
I was looking more for a panel diagram... Also, many of the abbreviations don't seem to tie in with what is written on the PCBs. Very confusing! I know if I wire it from that sketch I'll probably smoke something expensive.

Anyone got a panel wiring diagram for the actual panel offered? More in line with this sort of style...


No diagram sadly, but everything is pretty much self-explanatory, with a few comments:

* Driver board: VD = Vibrato depth = Depth on Seth's panel

And comment from Barcode for wiring the drivers and tri->sine converters:

Quote:
Wiring up the sine converter is super easy. Just run one of the Triangle outs from the Core to the Triangle In on the Driver. The the Sine Out on the Driver to your panel. Repeat 2 more times.

For the Driver, just connect one of the Driver Outs to the center of the lugs on your switch. Then connect one of the other lugs to the Driver In on the Core. Repeat 2 more times, separate switches per Core of course.


* One driveout goes also to the driver jack on the panel.

Quote:
Must BOTH the 550 and 560 trannies be of the "C" variant for this to work? hmmm.....

Not sure but better play safe, I've used for both C variants and are working fine.

Just finished one VCO core:
* Implemented Benjamin AM fixes for the PWM & FM pots by just using the resistor leads itself with some heat shrink.
* Socketed R39 and placed a 2M resistor (no 1,5M resistor on hand). 510K is not going to work on my build.
* My goal is to install 10 turn pots for the frequency. tight fit and some small adjustments needs to be done:
- Enlarge the panel holes
- Remove the opamp near the frenquency pot to stripboard.

Everything seems to work well except the scaling is not good at all. Pitch is stable but the tracking is poor, no matter what I do with R9. It keeps in tune maybe for one octave at most???? I have made some pictures, to show how the 10-turn pots fits and hopefully somebody have some advice for my poor tracking.

Thanks.






Benjamin AM
Quote:
I have made some pictures, to show how the 10-turn pots fits and hopefully somebody have some advice for my poor tracking

In this picture, I notice that you haven't brought the v/oct input to the front panel. I then might question how you then tested your module. Was the voltage source and the Osc sharing a common ground? If not, then that could be the problem. Second question. Did you install the SMT tempco in parallel with that though hole?? If you did, remove one. Sorry if the questions are basic but they have to be asked.
Jop
Hi Benjamin,

For testing I have used some alligator clips tapped from the wires down at the bottom, including ground for all jacks.

I didn't install the SMT tempco, only the thru hole KRL part.

It should be able to track easily 5 octave's. or?
Benjamin AM
Jop wrote:


It should be able to track easily 5 octave's. or?

Definitely. You should have no problems "tuning" the thing up. My Living VCOs track good enough for 6 or 7 oct., which is relative because the circuit was designed to drift a bit. Fortunately, this problem that your experiencing will be easy to solve.The problem is either with the expo converter"circuitry around CA3046" or your equipment or how you are using it. I would check solder connections around the expo converter. Look for cold solder connections and solder bridges. Did you make any attempt to wash away excess flux after completing the project? Are you using BC560C transistors as suggested?
How are you tuning the oscillators? Frequency counter?? If so, I might suggest to use the saw wave output to tune. It is a 0-10V output which your multimeter might like better. Good luck
Jop
Just had a good look at my board, checked for continuity, bridges, resistor codes / cap values but no oddities found... All components are tested before they are being soldered.

Flux was washed away and board looks clean (besides the thru-hole bypass cap instead of SMT C14..., lifted the pad a bit here.... d'oh!

I have used C variants for all transistors.

For now I just tuned them by ear with my kenton pro 2000 connected to a digital synth.

What would be the best and most efficient way to check further?
My guess is:

* Look at Q1 & Q2, these are now just thrown in randomly and not matched. Maybe I have bad luck and picked two that are not in spec?

* Re-place CA3046 with a new one, don't hope this is needed because of the tempco glued on.

Thanks!


LektroiD
Anyone got a problem with the DC offset on the saw? Mine starts at 0V and goes into the +V, whereas the Triangle and Pulse are perfectly centred on my scope.

I've only wired one oscillator so far. As I add more to the panel, it is impossible to access the trimmers of the previos, so I'm doing them one at a time.

Also, does anyone know what each trimmer does? All I have is designators (both on the BOM and the PCB), but no guide to suggest what they are? Jurgen's designators and placents are completely different from these boards, so I can't even use that to relate...
LektroiD
Benjamin AM wrote:

This I s how I reworked the PCB without cutting any traces. For the PWM circuit, I unsoldered one side of R38(connected to CW leg of PWMPOT) and added a length of wire to route the loose side of the resistor to the pad labeled as "PWM". You will notice in the picture that I did this with green wire and used some white heat shrink for a little protection. Now R38 is connected to the middle leg of PWMPOT. The PWM input then has to move to the CW leg of the pot. I did this by wiring the PWM input jack directly to the board. Fortunately there are two holes for 9mm and 16mm accomidations. Simply use the free hole and wire directly to the input rather than using the normal connection(MTA header in photo). In the photo you might notice that I made this connection with yellow wire. As for the FM circuit, I used the same sequence. The only difference is that the FM summing resistor is R2. It should be re-routed to the "FM" pad and the input should be hard wired to the CW of FMPOT.


Where's the picture?
Jop
Quote:
Anyone got a problem with the DC offset on the saw? Mine starts at 0V and goes into the +V, whereas the Triangle and Pulse are perfectly centred on my scope.


* Not that I remember, but will check this is in more detail tonight.

Quote:
Also, does anyone know what each trimmer does? All I have is designators (both on the BOM and the PCB), but no guide to suggest what they are? Jurgen's designators and placents are completely different from these boards, so I can't even use that to relate...


* Designators from Jurgen's shematics page 1 are mostly the same:
R8 = Frequency
R9 = Scale, 1 v/oct tracking
R18 = High frequency tracking
R40= PW adjust

* How does yours track? Did you match Q1 & Q2?
Benjamin AM
Quote:
Anyone got a problem with the DC offset on the saw? Mine starts at 0V and goes into the +V, whereas the Triangle and Pulse are perfectly centred on my scope.

I've certainly got a problem with this but that is the circuits normal function. Haible choose to leave it that way so it would be easier to use with wave shapers. Honestly I wish that it would have been biased to +/- 5.
Quote:
I've only wired one oscillator so far. As I add more to the panel, it is impossible to access the trimmers of the previos, so I'm doing them one at a time.

I ended up swaping out some of the trimmers for right angle versions so I could access them from the bottom. I would highly suggest doing this before you get much further.
Quote:
Where's the picture?

It was lost when the webpage went down a month ago. Reference the picture that I posted on the previous page. You may need to be logged in to see it.
LektroiD
Another problem I have is that my PWM CV is barely usable, I have to highly attenuate it to get any sort of result. What may I have done wrong to cause this? I've triple checked all components against the latest BOM.
Jop
Quote:
Another problem I have is that my PWM CV is barely usable, I have to highly attenuate it to get any sort of result. What may I have done wrong to cause this? I've triple checked all components against the latest BOM.


My first suspect would be R39, what did you use here? Looks like most builders need to change this resistor to around 1.5M.

Did you calibrate R40 correctly :

Pulse width adjust. With the front panel Pulse Width knob in 12 o'clock position, adjust the trimmer for 50% pulse width.

* does yours track correctly?

Thanks.
Benjamin AM
LektroiD wrote:
Another problem I have is that my PWM CV is barely usable, I have to highly attenuate it to get any sort of result. What may I have done wrong to cause this? I've triple checked all components against the latest BOM.

I addressed this issue back on page 4 of this thread. There is a quick fix.
logicgate
Hey guys, hope everyone is having fun building those nanners


One thing I did to make sure all oscillators were calibrated the same was:

1- Leave them running for 10 - 20 minutes, to warm up
2 - Appply 1V at the 1v/oct input
3 - Monitor pin 1 of CA3046 with those tiny grabbers (using your DMM, one clip to pin 1 the other to ground)
4 - Adjust the freq pot to, let's say, C2 and write down the readings from pin 1
5 - repeat the steps for all oscillators, you gonna have to adjust the 1v/oct trimmer to be able fine tune them so they all have the exact same readings


At least this worked for me, the tracking is not THAT great, but they work... I don't know if it's possible to achieve "perfect" tracking with those oscillators cause they were meant to drift a bit so... One thing I do when a riff has some slightly off notes (and this is not only with the LVCOs, other VCOS do drift too) is to use Melodyne to correct some notes, usually semi tones.

Good luck


EDIT:

Someone asked in a post about the need to use the C kind of transistors! This is CRITICAL for you LVCO to work, if you use another kind they WON'T WORK!!
qfactor
And by this do you mean BOTH the 550 and 560 trannies?
hmmm.....
spacedog
LM311P or LM311N ? hmmm.....
the bad producer
spacedog wrote:
LM311P or LM311N ? hmmm.....


Doesn't matter, also

Quote:
And by this do you mean BOTH the 550 and 560 trannies?


Yes...

thumbs up
qfactor
Just out of curiosity, anyone know why ONLY the "C" variant trannies work for the Osc? What's the difference between "C" and the "B" that makes it a go/no-go situation? hmmm.....
logicgate
qfactor wrote:
Just out of curiosity, anyone know why ONLY the "C" variant trannies work for the Osc? What's the difference between "C" and the "B" that makes it a go/no-go situation? hmmm.....



If you check the datasheets you can see that the "C" kind has higher Hfe (which I believe is the gain). I'm no EE so I have no clue why one works and the other doesn't... I only know that this gave me a major headache cause when I had the circuit on breadboard it wouldn't work (I was using B and "plain" kind), when I switched to C kind it worked.
elmegil
logicgate wrote:
qfactor wrote:
Just out of curiosity, anyone know why ONLY the "C" variant trannies work for the Osc? What's the difference between "C" and the "B" that makes it a go/no-go situation? hmmm.....

If you check the datasheets you can see that the "C" kind has higher Hfe (which I believe is the gain).


Yup smile
pantalones
LeftyLogic wrote:
Oh, and idealer, here are the current BOMs!


I admit being over my head here — can you clarify what tolerance to use on the resistors, any specific kinds of pots, etc?
LeftyLogic
pantalones wrote:
I admit being over my head here — can you clarify what tolerance to use on the resistors, any specific kinds of pots, etc?

Personally, I used almost entirely 1% resistors, but I think 5%'s should probably work just as well. (Although 1% resistors aren't really any more expensive than 5%'s, so I'd recommend just getting those anyway.) For the 3 0.1% resistors I just matched a few 1%'s as closely as I could with my multimeter.

For pots, you could theoretically cram a 10-turn in for the frequency, but a standard 1-turn will work perfectly fine as well. The only downside is that it'll be a bit more finicky getting the VCOs in tune, but IMO that really isn't a huge deal.

If you have any more questions, feel free to ask! thumbs up
Seth
Royalston
Hi,
Ive got a couple of questions about some BOM components:

for the LM13600/NS on the Driver, could I use something like this?:

http://www.ebay.com.au/itm/1-piece-LM13600-LM13600AN-/140750760796?pt= LH_DefaultDomain_15&hash=item20c5662b5c


its an LM13600AN. Im having trouble finding one with "/NS". I don't know what that signifies either...

Also,
The 2.7n Polystyrene in the VCO core...is quite hard to find. Is some other material ok?


Thanks
Royalston
Also: Just wondering for this and most other eurorack DIY projects - what is the minimum power rating for resistors I should use?

Thanks!
the bad producer
pantalones wrote:
LeftyLogic wrote:
Oh, and idealer, here are the current BOMs!


I admit being over my head here — can you clarify what tolerance to use on the resistors, any specific kinds of pots, etc?


1% is pretty much the standard, but you could use 5% if you liked, the important ones are the 100K matched 0.1% for the CV ins.

Any pot that you like the feel of/fits behind a panel/isn't too expensive/fits the knobs you want to use/mounts to PCB if you want it to and has the correct value is the correct pot.
the bad producer
Royalston wrote:
Also: Just wondering for this and most other eurorack DIY projects - what is the minimum power rating for resistors I should use?

Thanks!


I use 1/4W for everything unless higher is specced for.
the bad producer
Royalston wrote:
Hi,
Ive got a couple of questions about some BOM components:

for the LM13600/NS on the Driver, could I use something like this?:

http://www.ebay.com.au/itm/1-piece-LM13600-LM13600AN-/140750760796?pt= LH_DefaultDomain_15&hash=item20c5662b5c


its an LM13600AN. Im having trouble finding one with "/NS". I don't know what that signifies either...

Also,
The 2.7n Polystyrene in the VCO core...is quite hard to find. Is some other material ok?


Thanks


I think NS may stand for National Semiconductor, the mfr... Generally the suffixes for IC's denote grade or package and other things like that, reading the items datasheet should clear up any questions, but in this case any LM13600 will suffice, as long as it fits the footprint (ie is through-hole DIP)

The 2N7 could be bought from Rush On-Line, or you may try a C0G ceramic instead of polystyrene...

https://www.rush-on-line.co.uk/cap_index.php

or Mouser 810-FK18C0G1H272J which is 2700pf C0G ceramic

Another thing, if I'm searching US sites for caps, I'll use 2700pf for 2N7, I don't think they use 'nano' in the states...
Royalston
The bad Producer- THANK YOU! smile
qfactor
I've been looking around eBay for the 10k, 10-turn pot and came across this. Am wondering if the shaft diameter would fit the panel?
The data sheet says the shaft diameter is 6.3mm while the "mounting hole" is 10mm.
Another thing is can anyone shed light as to the pinout of this pot? Which is CW and which, CCW?
Thankie! thumbs up


EDIT: Hang on! Looking a little more d'oh! I found this:


Now are the 2-1-3 the same as these 1-2-3? seriously, i just don't get it
the bad producer
Yeah I got that wrong on my first Living VCO, the pots caught fire!
(This was a long time ago when I was first starting out BTW!) I think you'll find that the pin numbers are written on the body of the pot too...

Someone posted a picture of a build using these further back in the thread, I think they had to move an IC out the way on the pcb to get them to fit...
Randy
If there was one thing I would change for further runs of this PCB, it would be the pads for the trannies. I've just removed all of the 550s because they were not 'C' types, but the 560s were fine. I'm not looking forward to trying to solder those 550s back on again.

Randy
logicgate
Randy wrote:
If there was one thing I would change for further runs of this PCB, it would be the pads for the trannies. I've just removed all of the 550s because they were not 'C' types, but the 560s were fine. I'm not looking forward to trying to solder those 550s back on again.

Randy



Hi Randy,

One thing I recommend before starting soldering like no tomorrow is checking the original thread, given that it is the source of the PCBs... If you read through it, you will find that I clearly state the importance of using the "C" kind of transistors, because the circuit wouldn't work with another type.
Randy
logicgate, I take full responsibility for my screw-up, there was certainly no blame intended. I found soldering the transistors the first time the only tricky part of the build, at least so far. I can only imagine how tricky this is going to be the second time, but again, my fault entirely.

Randy
Royalston
Can anyone show me a mouser number/ name of -
- the 10 -16 pin power cables that eurorack generally uses
- the 10 pin power connection thing for euro

Ive got everything else on the BOM but I can never work out what these are actually called - so I can never found them.

Thanks!
medbot
I'm not sure where you're located, but you can get the power cables from either Erthenvar in the US or Thonk in the UK. The power connectors are called pin headers, and you can get them very cheaply from Tayda here. They come in 40 pin long sticks and you just break off however many you need for a module. If you have to get them from Mouser, I think this should work, but that's just for one.
spacedog
logicgate wrote:


....checking the original thread, given that it is the source of the PCBs... .


Witch thread is the original? hmmm.....
logicgate
spacedog wrote:
logicgate wrote:


....checking the original thread, given that it is the source of the PCBs... .


Witch thread is the original? hmmm.....


https://www.muffwiggler.com/forum/viewtopic.php?t=93892&postdays=0&pos torder=asc&start=0
spacedog
Thank you!
Guinness ftw!
fluxivity
I there someone who could share the mouser cart? I promise to post audio of my girl using these to modulate her vocoder. thanks

p
qfactor
the bad producer wrote:

Someone posted a picture of a build using these further back in the thread, I think they had to move an IC out the way on the pcb to get them to fit...


So....I guess the 10turner pot definitely won't fit in with that IC there! eek!
Besides Seth saying that it doesn't really matter and that one can still get by with a "normal" pot (can I use a "normal" 100k pot instead?? seriously, i just don't get it ), anyone else have tried the Frequency knob NOT be the 10-turn pot?? hmmm..... hmmm.....
logicgate
qfactor wrote:
the bad producer wrote:

Someone posted a picture of a build using these further back in the thread, I think they had to move an IC out the way on the pcb to get them to fit...


So....I guess the 10turner pot definitely won't fit in with that IC there! eek!
Besides Seth saying that it doesn't really matter and that one can still get by with a "normal" pot (can I use a "normal" 100k pot instead?? seriously, i just don't get it ), anyone else have tried the Frequency knob NOT be the 10-turn pot?? hmmm..... hmmm.....


I used normal 100k pots on mine, but I think I´m gonna swap for the 10k 10 turn ones, cause I´m finding imposible to keep them in tune at different octaves. They ALMOST keep in tune with each other, depending of the notes...
qfactor
logicgate wrote:

I used normal 100k pots on mine, but I think I´m gonna swap for the 10k 10 turn ones, cause I´m finding imposible to keep them in tune at different octaves. They ALMOST keep in tune with each other, depending of the notes...


I don't understand...I assume when you can't keep them in tune over different octaves, you're using a keyboard to control it?
If so, how does the 10-turn 10k pot help to keep it in tune? hmmm.....
qfactor
Sorry duplicate posting because the server timed out on me!! d'oh!
qfactor
Triplicate posting!! sad banana
logicgate
@qfactor

Yes with keyboard playing.

I may be shooting in the dark here but I believe that the precision pots will "lock" them in the same frequency more "firmly", less drift from each other.
qfactor
logicgate wrote:
@qfactor

Yes with keyboard playing.

I may be shooting in the dark here but I believe that the precision pots will "lock" them in the same frequency more "firmly", less drift from each other.


sad banana
the bad producer
Personally I'd be surprised if the pot type affected the tracking - I can see that it may help to adjust the frequency of each VCO with finer precision, enabling closer matching when played... FWIW wire-wound pots are apparently less stable (noisy) than cermet, certainly Ian Fritz recommends cermet pots for stability on his Teezer VCO...

I intend to build these PCB's with fine and coarse tune, all my old LVCO's (original Haible PCB's) had 10-turn pots but I think I prefer two single-turn, not least for the cost saving!

Also, if using a single-turn why use 100K rather than 10K?
logicgate
the bad producer wrote:
Personally I'd be surprised if the pot type affected the tracking - I can see that it may help to adjust the frequency of each VCO with finer precision, enabling closer matching when played... FWIW wire-wound pots are apparently less stable (noisy) than cermet, certainly Ian Fritz recommends cermet pots for stability on his Teezer VCO...

I intend to build these PCB's with fine and coarse tune, all my old LVCO's (original Haible PCB's) had 10-turn pots but I think I prefer two single-turn, not least for the cost saving!

Also, if using a single-turn why use 100K rather than 10K?


Hummm now that you´ve mentioned it, I really don´t know why I´ve used 100K pots for the coarse frequency hmmm..... lol

I´m gonna replace with 10K and see if it makes them drift less from each other when playing with the keyboard or sequencer before going for the multiturn
qfactor
Do keep us informed! smile
raisinbag
Cool re all the pot stuff, Ill wait to put mine till you give us a heads up...


Ok here's my questions;

1) does ferrite bead size make a difference? I can get ones that are 3.5mmx4.4mm OR 3.5mx5.9mm (but there are only a few of the bigger ones left at store, so I'm hoping the small ones will be cool)

2) anyone have a good pic of how they hooked up power and piggy backed from one board to the next? all headers are lined up, so not sure how I extned the pins from the first into he next..... Unless there is another single row for stacking headers. Of course I only thought of this as I'm tying this at work, and have no boards to look at.... I' probably do this d'oh! when I get home and look closely.....
elmegil
I can't speak for this board in particular, but what I've done if the power headers align sufficiently is to create a single long ribbon cable with multiple connectors and enough slack between each to just plug them in.

Here's an example from the Dual Mirror Core....
the bad producer
As elmegil says like that!

The size of the beads shouldn't matter at all, in fact smaller would likely be better in this instance.

You could do something like this for power (not LVCO's but the same idea):

Royalston
Sorry to be a pain - what are the parts you use for the double power connector thing? Obviously you use plain old ribbon cable and make your own cables here (I've never done this before). I was just going to buy 4 power power cables and plug them into my rack from each core and driver separately! Is this wrong?
Royalston
Ive found the basic 16 pin connectors, just looking for the piggy back solution to get one cable going to the next
elmegil
Royalston wrote:
Sorry to be a pain - what are the parts you use for the double power connector thing? Obviously you use plain old ribbon cable and make your own cables here (I've never done this before). I was just going to buy 4 power power cables and plug them into my rack from each core and driver separately! Is this wrong?


It's not *wrong*, but it ties up a lot of your power bus. It works just fine, and is effectively the same thing.

So those headers actually have multiple parts. Not only the socket and it's "top" that crimp together on the ribbon cable, there is also a strain relief piece which is what I used when looping it around as you see in my photo.

The Living VCO PCBs I have from barcode use the 10-pin IDC header rather than the 16.

So this is the header on the PCB itself:

http://www.mouser.com/Search/ProductDetail.aspx?R=66506-066LFvirtualke y64910000virtualkey649-66506-066LF

(it's keyed, make sure pin one, with the triangle, goes to the -12V side where your red stripe will go)

This is the mating socket:

http://www.mouser.com/ProductDetail/FCI/71600-610LF/?qs=sGAEpiMZZMs%25 2bGHln7q6pmxAVkKtOEC39qUmxYyyW36g%3d



Or with strain relief:

http://www.mouser.com/ProductDetail/FCI/66900-210/?qs=sGAEpiMZZMs%252b GHln7q6pmxDhjuO85y2pxNQfmgCgEgY%3d

I have NOT THOROUGHLY CHECKED the data sheets on this one with the strain relief to be 100% certain they are exactly what you're looking for. I've purchased this kind of socket in the past, but Mouser doesn't make it terribly easy to sift through old orders looking for "strain relief".

But if they are not exactly right, they are very close and should give you a starting point to look from. Read the data sheets! smile
LeftyLogic
Something I noticed on my prototype is that the LFO from the driver seems to bleed into the oscillator frequency when the power is daisy-chained, even when the oscillators were not set to track the driver. It doesn't seem to be audible, but it makes calibration a pain in the rear, so I'd recommend just using a 3-way chain for the VCOs and an individual power cable for the driver.

Also, the power connector on the driver is inverted with respect to that of the VCOs, and since none of these boards are reverse protected, I'd call that another good reason to give the driver its own power cable. Dead Banana FUUUCCKKKK!!!

Seth
elmegil
Can't argue with that logic, LeftyLogic smile
Royalston
Thanks for the info!
LeftyLogic
So now that spring break's started, I finally have enough time to get my next set of living VCOs built! It's peanut butter jelly time!
So far I've only gotten one of the oscillators finished, but I have already had a couple of important insights regarding 10-turn pots:

First, they do not have an intuitive pinout! The ones I bought had a diagram on the side, but it is pretty well impossible to decipher. Thus, I ended up burning out one of my pots due to hooking it up wrong. very frustrating Check the pinout with a multimeter or find a datasheet before you hook them up!

Second, don't cheap out on 10-turns! I bought the cheapest ones I could find from Jameco, and they suck. They have a seriously bad crackle problem, so I think I'm going to replace them all with these from mouser. (Although if anyone has a part number for 10-turns that are known to be good, I'm all ears!)

Seth
LeftyLogic
Oh, also, I've been taking step-by-step pictures as I build these, so once I'm finished I'm going to get a nice photographic build guide put together for you all! This is fun!
Altitude909
oh thank you, just got done reading through all 9 pages of this thread which seems to cover 3 different builds and has a mess of info. Please start with errata on the boards you just shipped (i.e. bad pinouts for some pots). I haven't ordered anything yet so I'll post my cart if no one does it before me
qfactor
^^ yeah, I second this!
Would be good to "reset" all that's said for this project and start afresh based on this latest version of PCB. thumbs up
the bad producer
I'll be adding a Living VCO page to my website with wiring diagrams and errata (build notes) I'm working on it now, and even though it will reference my 4U panel, hopefully the info will be useful to others building the VCO's in other formats!

The page is here, but there's not much yet!

http://www.loudestwarning.co.uk/portfolio/living-vco/
Altitude909
Ok, here is my cart for one core. Bourns Model 91 cermet pots, no 3046s (get those at Jameco)

*edit pulled. Normal 1/4 watt resistors do NOT fit the PCB

I have no idea what an "0504" size capacitor even is so all the bypass caps are 0603 which is what the footprint is for
Stides
LeftyLogic wrote:
Oh, also, I've been taking step-by-step pictures as I build these, so once I'm finished I'm going to get a nice photographic build guide put together for you all! This is fun!


This would be welcome. I've been debating whether to take this on, or sell the boards on. I recently finished a couple modules and I just have a very tough time with the calibration. X3 and with the driver board seems daunting. If you can include a calibration section it would be helpful too.
qfactor
Altitude909 wrote:

I have no idea what an "0504" size capacitor even is so all the bypass caps are 0603 which is what the footprint is for


I think it was mentioned in Pg2 of this thread, that 0504 was an error! It's supposed to be 0805! thumbs up
Randy
the bad producer's page for the Ultrafade is amazing, looking forward to seeing the page for the LVCO. Unfortunately, I think even the best instructions may not help me.

Randy
Altitude909
qfactor wrote:
Altitude909 wrote:

I have no idea what an "0504" size capacitor even is so all the bypass caps are 0603 which is what the footprint is for


I think it was mentioned in Pg2 of this thread, that 0504 was an error! It's supposed to be 0805! thumbs up


That's not an 0805 footprint, it's smaller. Compare it to the tempco resistor footprint. You can probably fit an 805 part on there but why not just get the right part
qp
Altitude909 wrote:
Ok, here is my cart for one core. Bourns Model 93 cermet pots, no 3046s (get those at Jameco)

http://www.mouser.com/ProjectManager/ProjectDetail.aspx?AccessID=4f8d2 2ba71

I have no idea what an "0504" size capacitor even is so all the bypass caps are 0603 which is what the footprint is for


Thanks for this! Thems are some posh pots for sure.
Altitude909
I figured this thing deserves them and they have everything that's needed in one place. As far as posh pots go, those are pretty reasonable. I'll get the other board BOM done tonight
LeftyLogic
Hey, just so you know, I did some measuring and unfortunately those pots aren't going to work on the driver section if you're using my panel. Due to the design of the driver PCB, using anything smaller than 16mm alpha type pots will plant the PCB smack dab in the middle of the oscillator 2 tracking switch. very frustrating

The bourns pots should work great for the oscillators themselves though! nanners

Seth
Altitude909
good catch. I wasnt planning on using those for the driver but not for that reason, so 16mm alphas it is there.
the bad producer
Just noticed - whilst putting some info together for building - that Dave Brown used 100K pots on his build:

http://modularsynthesis.com/jhaible/lvcos/jhlvcos.htm

Which got me thinking, do you mean drift between VCO's when they're all being driven by the driver?

Jurgen talked about this briefly in the orginal electro-music thread - highlighting the need for achieving unity gain through the driver, here page 16:

http://electro-music.com/forum/topic-30749-375.html

Dave Brown also mentioned this and says:

Quote:
Since there is no gain adjustment for U11B and U12 I used 0.1% resistors for R137, R141, R144, and R145.


logicgate wrote:
the bad producer wrote:
Personally I'd be surprised if the pot type affected the tracking - I can see that it may help to adjust the frequency of each VCO with finer precision, enabling closer matching when played... FWIW wire-wound pots are apparently less stable (noisy) than cermet, certainly Ian Fritz recommends cermet pots for stability on his Teezer VCO...

I intend to build these PCB's with fine and coarse tune, all my old LVCO's (original Haible PCB's) had 10-turn pots but I think I prefer two single-turn, not least for the cost saving!

Also, if using a single-turn why use 100K rather than 10K?


Hummm now that you´ve mentioned it, I really don´t know why I´ve used 100K pots for the coarse frequency hmmm..... lol

I´m gonna replace with 10K and see if it makes them drift less from each other when playing with the keyboard or sequencer before going for the multiturn
logicgate
@the bad producer

Hi mate,

No, I didn't build the driver section, just the VCOs... Using a buffered multiple I send the voltage to all three
livefreela
the bad producer wrote:
Yeah I got that wrong on my first Living VCO, the pots caught fire!
(This was a long time ago when I was first starting out BTW!) I think you'll find that the pin numbers are written on the body of the pot too...

Someone posted a picture of a build using these further back in the thread, I think they had to move an IC out the way on the pcb to get them to fit...


panel run on "jh flaming vco" charlie? hihi

additionally, for those wondering about transistor variants, i have one core board functionally running 2n5089 at the moment if any of you guys happen to have any of those on hand and would care to experiment. no spectacular ill effects thus far, though ive yet to build a set with the spec npns for more subtle comparison of noise and such. i just received the bc npns i had ordered ages ago and intend on building those into the remaining boards. i'll report any differences as soon as i'm able.
Royalston
I'd love a photo walk through.

I ordered barcode's PCBs and panels and am quite lost as to which updates/mods I need to do, and the pot wiring...but I'm not at that stage yet.

I also just found out I ordered the wrong size trimmers for 2 complete sets. And a whole bunch of the wrong through hole resistors that don't fit through the tiny resistor holes! very frustrating

Are the trimmers on the driver board 1/4" or 3/8" or should it not matter?
Altitude909
Royalston wrote:
..1/4" or 3/8" or should it not matter?
shouldn't matter, they both have the same pinout spacing, just make sure you get the right pin arraignment (straight or the V)
livefreela
another heads up - the holes on the resistor pads are extremely tight on these new boards (the black ones). i was unable to fit some brands with more robust leads (yageo from mouser as well as the vishay mil spec i bought for r1-3). xicon will fit with a bit of molestation. shitty tayda ones work fine. i didn't run into this on the green barcode boards so perhaps the new plating may be the culprit here...
logicgate
@bad producer

Interesting. I only matched the R1, R2 and R3 resistors... Gonna try matching all the rest, perhaps it will make a difference, let's see. (By "all the rest" I mean the resistors from CV input til signal reaches CA3046, on all 3 osc, basically r1, r2, r3, r4, r10, r13, r12)
Altitude909
livefreela wrote:
another heads up - the holes on the resistor pads are extremely tight on these new boards (the black ones). i was unable to fit some brands with more robust leads (yageo from mouser as well as the vishay mil spec i bought for r1-3). xicon will fit with a bit of molestation. shitty tayda ones work fine. i didn't run into this on the green barcode boards so perhaps the new plating may be the culprit here...


They do seem pretty tight, xicon and KOA fit snuggly. The vishay mil 0.1% weren't going to fit length wise anyway so I used the KOA 0.1% in my bom which are the standard size
qfactor
So.....what's gonna happen now? seriously, i just don't get it
https://www.muffwiggler.com/forum/viewtopic.php?t=134788
elmegil
Seems logical to discuss it with Ralf. Given that he's actively looking for new PCB designs (the old Eagle files are gone, apparently), this seems like a no brainer to jumpstart with, the question is how exactly that will work.
Royalston
Can anyone tell me what are the two small holes in between the three main 'holes' where the each of the pots sit on the core pcbs?
Altitude909
Royalston wrote:
Can anyone tell me what are the two small holes in between the three main 'holes' where the each of the pots sit on the core pcbs?


Dual footprints for 9mm or 16mm pots
Siri
@royalston

Look at the traces.

They are here to allow different types of pots to be mounted.

Those :



and those :



don't have the same pin spacing.
Altitude909
These tiny ass holes for the resistors are a real problem. I can get the xicon parts through with a struggle but everything else does NOT fit at all. This needs to be fixed on any further revision.

Furthermore, the spacing is wrong for standard 1/4 watt parts, they dont fit vertically and I am using Xicon parts which are small to begin with. Is this laid out for 1/8 watt parts? I can get Stackpole resistors (I think panasonic make them too) which are 1/8 watt size but rate for 1/4 watt but they are 5x more expensive
the bad producer
I had similar problems, thought about going for 0.4W R's but in the end just stacked them up and squeezed them in - I did dremel out a couple of holes with a tiny bit, but I'm not sure I'd recommend that as it destroys the thru-plating! I'm quite pleased with the results of the build. So far I've just double checked fit to my panel design with the driver and one VCO. I've got 5 octaves without trying very hard, so I'm happy - I also tweaked the PWM part of the circuit, may still do a bit more there tho....



duff
I just sat down to start stuffing these boards and hit the same problem. The biggest issue being the hole sizes. If opening then up isn't recommended I guess I need to work out what make the thin legged resistors I have are and get some more.
the bad producer
duff wrote:
I just sat down to start stuffing these boards and hit the same problem. The biggest issue being the hole sizes. If opening then up isn't recommended I guess I need to work out what make the thin legged resistors I have are and get some more.


If you've got some PCB drill bits it's worth a try, but you have to ensure you solder the pads on each side of the board to ensure continuity... It' surely best not to have to!
Altitude909
Digikey stocks the stackpole and panasonic parts, ill do up my bom with those and post here..
logicgate
I didn't have this issue with any of my boards...

I buy resistors from all over the place, but mostly from Tayda, the leads are thin enough
Altitude909
logicgate wrote:
I didn't have this issue with any of my boards...

I buy resistors from all over the place, but mostly from Tayda, the leads are thin enough


Those must be tiny resistors, I have Xicom, KOA, Vishay, and Yageo 1/4 watt here on hand and only the xicoms fit the holes (barely). None of them fit vertically in a row without overlapping. Those dont look like the right footprints for standard 1/4 watt resistors. I'm looking at the Stackpoles at digikey and they make 0.1% 100K and stock a ton of values so those may be the way to go for this since they are 1/4 watt parts but in a much smaller package (like an 1/8 watt part)
thetwlo
Altitude909 wrote:
Those must be tiny resistors, I have Xicom, KOA, Vishay, and Yageo and only the xicoms fit the holes (barely). None of them fit vertically in a row without overlapping.


yeah, the Tayda have very thin leads. And I believe some have posted they often aren't very accurate, which is usually fine. Not sure here.

The Meng Qi Rollz board has the same issue.

Perhaps 1/8th watt resistors would also be an option?
Altitude909
Alright, good news, Digikey has Stackpole parts in all the required values (5% for the 1M5, everything else 1% or 0.1%)
logicgate
I always measure the resistors with DMM before soldering, each one of them... Even the pharmacy priced ones from Mouser, Newark... Don't think that only because you buy there that the values will be accurate, they're not.


TBH I have never had any problems with Tayda resistors being that off, some are spot on, some are 0.5 - 1 ohm +/-


I had to make the footprints like that (and place the resistors real close to each other) so the circuit would fit on the board (any bigger and it would get more expensive), and it needed to be small, for Euro. I didn't mind some resistors overlapping here and there, it works the same, in the end... TBH this is the first PCB I've ever designed so, wasn't aware of standards, just "went with my gut" on this one.. lol
j9k
I'm having troubles with also with what I think are fairly standard resistors and the holes on the latest run of boards. The resistors I'm using are .55mm and it's a struggle to get them through the holes.
livefreela
logicgate, like i mentioned earlier, your earlier green boards worked just fine - even with the yageos that i favor (snug, but that's not a bad thing). the tolerances on the layout may well be very tight, but i'm inclined to chalk this one up to discrepancies at the fab. tayda resistors are certainly workable with the new boards - i usually hate them not due to tolerances but because of their mechanical fragility compounded by my being something of a ham-fist. hihi

btw re: xicons i only used these for a few values, but i did have to select from the big bags individual resistors that would fit. a few holes on the new boards were really uncooperative and i had to whittle the leads down a bit with wire strippers. (not the fixed gauge sort, but the kind where cutting diameter can be "dialed" in)
ZibraZibraZibra
time to buy some diddy legged resistors!...

i got my boards a couple of days ago woo

luckily pay day is stopping me from jumping in and starting right this second, got a week then building 6 oscillators..

so hopefully all the niggles i gotta watch out for will be sorted by then razz

im putting them in a 4u format. 3 in each module with a driver. but also toying with the idea of it having a built in mixer. with a multi waveform select 4 way switch.... best mixer circuit for this? I'm going for a tl072 one at the minute.
ThecureForSin
Has anyone actually built LeftyLogic's boards yet? I don't quite understand the wiring. The individual vco wiring makes sense, but how does the driver factor into it and the free/track switches?
Royalston
Anyone using barcode's panels - 16mm pots fit the board perfectly but then you can't mount the PCB and jacks behind the panel (There's no room for the jacks). Is anyone selling 2 eurorack panels of a different design?

Thanks
Royalston
I just finished my first core....Im finding the pitch is wobbling everywhere (constantly, even with a flat input voltage) - swinging several semitones. I used a 1k through hole tempco under the chip (almost touching). This kind of wobbling shouldn't be there Im guessing. Should I use some heat "goo" to make the chip and tempco touch? I haven't calibrated yet...still trying to work out what each trimmer refers to.
Royalston
I can even make it wobble by waving my hand near it!
sammy123
The chip not making contact with the tempco would not cause wobbliness. I had my Living VCOs for months without tempcos without any issues. Ot must be something else. Could the vibrato be enabled?
sammy123
Royalston wrote:
I can even make it wobble by waving my hand near it!


That happened to me once with my resonant EQ. I would resolder all your joints.
Royalston
resolder! ok...I'll try. No vibrato yet - its just a single core on its own
Royalston
It tracks over octaves (apart from its inherent wobbliness)
Royalston
I removed the "FM/ PWM fix" that someone suggested earlier in this thread and my pitch seems much more stable, however my problem now (one of many obviously) is that when the fm and pwm knobs are moved from fully CCW position (with some cv plugged in), there is a pitch jump in the oscillator as the knob turns (on all the different wave outputs)...which is odd.

Any ideas?
Benjamin AM
Royalston wrote:
I removed the "FM/ PWM fix" that someone suggested earlier in this thread and my pitch seems much more stable, however my problem now (one of many obviously) is that when the fm and pwm knobs are moved from fully CCW position (with some cv plugged in), there is a pitch jump in the oscillator as the knob turns (on all the different wave outputs)...which is odd.

Any ideas?

When the mod was installed, where the FM and PWM inputs normalled to ground? If not, then that could explain the weirdness. While your at it don't forget to normal the 1v/Oct jack.
Royalston
BenjaminAM - thank you for the help. Any chance you could explain normalling a little better for a noob like me. Would this be the "switch" part of the jack needing to be connected to the ground? At the moment I have the switch connection unconnected (on all jacks). I have tried connecting all these switch connections to the ground but it doesn't seem to help. Should they be connected to 0V?

The main weirdness I'm getting now it whenever I touch a pot, or any part of the PCB (just lying on the desk at the moment in testing phase), the pitch changes. When I move pWM, and PW, the pitch of the oscillator steps up (even on the saw wave). The FM seems to be working correctly now though. I have removed all mods after speaking to Barcode.


I'd love to get this one solid before I make 5 more and two drivers!
Benjamin AM
Royalston wrote:
BenjaminAM - thank you for the help. Any chance you could explain normalling a little better for a noob like me. Would this be the "switch" part of the jack needing to be connected to the ground? At the moment I have the switch connection unconnected (on all jacks). I have tried connecting all these switch connections to the ground but it doesn't seem to help. Should they be connected to 0V?

Yes exactly. I imagine that you have switching jacks that are "Normally Connected" or "NC". This is easily tested with your multimeter. If no patchcords are inserted into the jack then the switching lug is connected to the tip. In the case of the Living VCO, input jacks that carry voltages to the frequency section should be normalled ground or 0V. The 1V/Oct is normalled to ground and connected to the frequency mixer through a 100k resistor the FM input is also normalled to ground and connected to this section through a 100k resistor(and the pot). Same thing with the PWM circuit(different summing cicuit). Anyhow the purpose is to add stability to your circuit via the steady reference to ground. This is more important in the frequency mixer because the human ear is so sensitive to pitch change. Check out Haible's schematics, there is no resistor in the mixer section to ground, therefor he relies on the switching jack to serve this function. In the wiring diagram you ill notice that the three connections are represented by an arrow pointing away from the jack sleeve toward the jacks tip.
Quote:

The main weirdness I'm getting now it whenever I touch a pot, or any part of the PCB (just lying on the desk at the moment in testing phase), the pitch changes. When I move PWM, and PW, the pitch of the oscillator steps up (even on the saw wave). The FM seems to be working correctly now though. I have removed all mods after speaking to Barcode.

Yes because there is no consistent reference to ground. I could explain why it is acting different between your previous modded version and the way it is wired on Barcode edition PCB. This would include a discussion on how an attenuation is wired. Lets get the core working properly first.
Siri
How should you do with bananas then?
Royalston
Thank you so much for the help Benjamin! I would be completely lost.
I have tried connecting all the switch connections of FM, PWM & V/OCT to ground. I also tried 0V (are these the same?). Ground seems to work better but maybe Im imagining it. My pitch still fluctuates whenever my hands touch the circuit but it seems more stable otherwise -when I'm not touching it. Pitch still jumps when I change the PW & FM knob in the last 1/8th of the turn (it steps up suddenly).

Thank you again! Its been a bit of a disastrous build so far. I've taken it slowly and carefully, triple checking my mouser BOMs as ordering mouser from Australia is expensive...Then I discovered my resistors were too large to fit through the holes so I rebought them, and then last night I discovered all my pots were too big for the panels too...so I rebought them too. Dead Banana

I guess I can probably offer someone 2 cheap sets of 16mm pots
Benjamin AM
Hmmmm... Now that variable is out of the equation. What power supply are you using??? When you do a continuity test between 0v and ground does your multimeter report no resistance(beep)? Are your jacks sleeves connected to ground?Does the jump in pitch occur when a knob is attached?
logicgate
@Royalston

Maybe this project is a bit advanced for your skills? You seem confused about very basic things.

I would suggest practicing and studying just a bit more before you move on to the other boards...

Perhaps take a week to learn how to properly read schematics so you can rely on them to debug.

This circuit is kind of simple to lay on breadboard, you could do that to practice the schematic reading, and then apply the skills on your build.

Debugging is hard in situ already... Over internet forum posts, it will be even harder

When I first got into DIY, I read a book called "Getting Started on Electronics" by Forrest Mims. It's a wonderful book for beginners, I highly recommend it! Gives you a clear description of the electronic components and some basic circuits. Believe, it makes a lot easier to understand the schematics.


Now, about your build, I would suggest double, triple check the wiring, sometimes it's a simple wiring error. Have you checked the solder joints of the transistors to see if there's any bridges or cold joints? I use a magnifying glass to do that, for all joints, good practice. If you have a micro blob of solder shorting something somewhere, it's enough to cause something similar to what you describe, or even worst (magic smoke).

Good luck
Royalston
Hi,
@logicgate. This project probably is a bit beyond my skills but I have built several other projects on here successfully so I thought I'd step up and give this one a try. I din't realise it would be especially difficult. I've had a pretty good look for solder bridges with a magnifying glassand I think my soldering is pretty decent. I've done lots of continuity tests between all the transistor pads and other components and comparing with the schematic.

2 things I noticed in my elementary debugging:
there is no continuity between +1vb as measured from the detune +-vb pad and the -1vb pad of r40. Connecting these however, resulted in the module not powering up.

no continuity between pin 2 of the CA3046 and R13. pin 1 has continuity.

I don't know much about debugging circuits or modifying circuits. I can read schematics and identify components - I could lay this project out on a breadboard -but I don't really understand how the circuits work and what each component is doing within the circuit - I have big holes in my knowledge of electronics that I have tried to fill by reading up on the internet, books, and by asking questions (probably very annoying on this forum but I'm not sure there is a better place). For example I don't know what components to modify so that my PWM is more attenuated (I have to knock the CV signal way down to make it useful) I think I have that book you mentioned and probably need to read it again.

@benjaminAM. There is continuity between 0V and sleeves on all the jacks.

As stated above:
there is no continuity between +1vb as measured from the detune +-vb pad and the -1vb pad of r40. Connecting these however, resulted in the module not powering up.

And no continuity between pin 2 of the CA3046 and R13. pin 1 has continuity to r13.

I've put some knobs on and the pitch is much less sensitive to me touching it - this is good! It seems pretty stable.

The jacks are still held together with banana clips so if the cables wobble there is a pitch bend...which seems normal to me and would stop if everything was soldered together.

The behaviour of the other pots continues -
Pitch - Good. works well
Detune - doesn't seem to do anything (I know it is supposed to be subtle). Also hard to tell with the inherent pitch wobbling.
PW - pitch jump when knob is turned, at the start of rotation. Functions properly though.
PWM - small pitch bend near full CW rotation. Functions properly (needs lots of attenuation though)
FM - seems good. no jumps. frequency modulation works properly.

Power source is a doepfer LC6 case. I just noticed when I plug a cable into any other module in the case (the other end can be unplugged or patched anywhere) , the LVCO pitch changes...so I guess this is something to do with grounding?

Thank you very much for your help!
synchromesh
Siri wrote:
How should you do with bananas then?


I'm wondering about this as well... Is it a case of using the right size resistor to ground (big enough to not compete with the output impedance of what's connected, small enough to not magnify leakage currents into noticeable voltages)? Can anyone offer any suggestions?
the bad producer
I've been using the Living VCO's for years with bananas - in fact my whole system is bananas. I've never had any problems at all. I know of this 'NC switch to GND' thing and I have occasionally just jacked ground in there using a cable from a GND jack to see what happens = not a lot. But that's the easiest way if you're worried and you're not actually using your 1V/oct input.
elmegil
the bad producer wrote:
in fact my whole system is bananas.


hihi
synchromesh
the bad producer wrote:
I've been using the Living VCO's for years with bananas - in fact my whole system is bananas. I've never had any problems at all.


That's what I wanted to hear, thanks the bad producer! It's peanut butter jelly time! I'd been anticipating maybe having to install small SPST switches on panels, but it would be nice not to have to bother...
Siri
Problem with the switch is if you switch to ground while you plug a signal in, the switch starts to smoke... Dead Banana And it happens when you're tired or when a friend tries it. hihi If prefer the idea of the ground jack, more safe (and one less hole in the panel) thumbs up
Benjamin AM
the bad producer wrote:
I've been using the Living VCO's for years with bananas - in fact my whole system is bananas. I've never had any problems at all. I know of this 'NC switch to GND' thing and I have occasionally just jacked ground in there using a cable from a GND jack to see what happens = not a lot. But that's the easiest way if you're worried and you're not actually using your 1V/oct input.

I wouldn't imagine that it does much. Maybe it offers a bit of stability. When I pointed this out we were trying to identify instability in Royalstons circuit. Taking a floating input out of the equation was my main concern.
the bad producer
That's cool Benjamin AM, I totally understand - I just wanted to assuage any fears that the other guys might have regarding banana usage! thumbs up

Benjamin AM wrote:
the bad producer wrote:
I've been using the Living VCO's for years with bananas - in fact my whole system is bananas. I've never had any problems at all. I know of this 'NC switch to GND' thing and I have occasionally just jacked ground in there using a cable from a GND jack to see what happens = not a lot. But that's the easiest way if you're worried and you're not actually using your 1V/oct input.

I wouldn't imagine that it does much. Maybe it offers a bit of stability. When I pointed this out we were trying to identify instability in Royalstons circuit. Taking a floating input out of the equation was my main concern.
synchromesh
Siri wrote:
Problem with the switch is if you switch to ground while you plug a signal in, the switch starts to smoke... Dead Banana And it happens when you're tired or when a friend tries it. hihi If prefer the idea of the ground jack, more safe (and one less hole in the panel) thumbs up


I guess an SPDT on-on switch would be a better choice, then... but I see your point about fewer panel holes. And it's great to hear people's experiences - this has all been theoretical for me thus far, but now I'm finally building stuff these questions are becoming more urgent... This is fun!
Siri
If i switch to ground if there is a signal in the jack, it should not smoke... Forget what i said, i have to check that wiring meh
elmegil
Siri wrote:
If i switch to ground if there is a signal in the jack, it should not smoke... Forget what i said, i have to check that wiring meh


The way to do it would be to connect the center of the switch to the input, one side of the switch to the banana, the other side of the switch to ground. No chance of shorting your control signal.
Altitude909
Here's what the board looks like with the Stackpole resistors and 0603 SMD caps, everything fits like it should. Problem is that Digikey is the only place that stocks Stackpole stuff in any serious quantities and the RNMF series parts are what I consider exotic (1/4 watt the size of 1/8 watt). I suppose you could go all 1/8 watt but someone more knowledgeable with the circuit should confirm that. Standard 1/4 watt parts dont fit the holes and dont fit vertically, you could hammer them in, but you can also hammer in a nail with a grapefruit if you try hard enough, doing it right is better solution. Same goes for the SMD bypass caps, those are 0603 footprints, not 0805




Randy
Tayda-supplied resistors (or almost anything from the eBay China, Hong Kong, etc suppliers) fit fine as well.

That's what I used, no problems fitting.

Randy
weedywhizz
Some people don't trust Tayda's quality for all parts.

@Altitude909

Those are the right ones ?
http://www.digikey.de/product-detail/de/RNMF14FTC10K0/S10KCACT-ND/2617 809
Randy
weedywhizz wrote:
Some people don't trust Tayda's quality for all parts


Why? I use them quite a bit, no problems yet. Does anyone have any actual (and recent) proof there are problems with Tayda quality?

Doesn't really matter to me what people use, I just don't see any reason to struggle through this stuff just because someone thinks resistors with thicker gauge leads are somehow better than thinner leads.

Randy
weedywhizz
Randy wrote:
weedywhizz wrote:
Some people don't trust Tayda's quality for all parts


Why? I use them quite a bit, no problems yet. Does anyone have any actual (and recent) proof there are problems with Tayda quality?

Doesn't really matter to me what people use, I just don't see any reason to struggle through this stuff just because someone thinks resistors with thicker gauge leads are somehow better than thinner leads.

Randy


Never really used Tayda's resistors or capacitors but filmsy thin legs is just not for me. Thicker legs gives me the feeling for better quality which I definitely personally prefer and also pay for.

I'm sure Doepfer 3.5mm cables are totally fine but I prefer TipTop cables - just because they feel better razz
Altitude909
Randy wrote:
weedywhizz wrote:
Some people don't trust Tayda's quality for all parts


Why? I use them quite a bit, no problems yet. Does anyone have any actual (and recent) proof there are problems with Tayda quality?

Doesn't really matter to me what people use, I just don't see any reason to struggle through this stuff just because someone thinks resistors with thicker gauge leads are somehow better than thinner leads.

Randy


Has nothing to do what someone thinks. If you design a board with footprint for a part only available from a discount shop in Thailand, you document that and make it perfectly clear that only that part fits.
Altitude909
weedywhizz wrote:
Some people don't trust Tayda's quality for all parts.

@Altitude909

Those are the right ones ?
http://www.digikey.de/product-detail/de/RNMF14FTC10K0/S10KCACT-ND/2617 809


Yes. Sackpole RNMF series (only, RNF are standard size)
weedywhizz
Altitude909 wrote:
Randy wrote:
weedywhizz wrote:
Some people don't trust Tayda's quality for all parts


Why? I use them quite a bit, no problems yet. Does anyone have any actual (and recent) proof there are problems with Tayda quality?

Doesn't really matter to me what people use, I just don't see any reason to struggle through this stuff just because someone thinks resistors with thicker gauge leads are somehow better than thinner leads.

Randy


Has nothing to do what someone thinks. If you design a board with footprint for a part only available from a discount shop in Thailand, you document that and make it perfectly clear that only that part fits.


+1000
Another type of part flooding my racks...
Royalston
Sorry to ask again - Does anyone have any pointers on where to start debugging my unstable LVCO build. I have re-done the wiring, tested for continuity throughout the circuit, referencing the schematic. I have changed power supplies. That all seems solid, however my oscillator pitch fluctuates when I move my hand anywhere near the module (within 10-20 cm).
I have now built and tested two different cores - both have the exact same problem, so I think that rules out bad soldering/ wiring and leads more to bad components/ wrong component values.

Would my replacement of the polystyrene cap with a film cap potentially cause this problem? Or a possibly fake Ca3046 chip? Ive used 1% resistors through pretty much.


Thank you.
ThecureForSin
Royalston wrote:
Sorry to ask again - Does anyone have any pointers on where to start debugging my unstable LVCO build. I have re-done the wiring, tested for continuity throughout the circuit, referencing the schematic. I have changed power supplies. That all seems solid, however my oscillator pitch fluctuates when I move my hand anywhere near the module (within 10-20 cm).
I have now built and tested two different cores - both have the exact same problem, so I think that rules out bad soldering/ wiring and leads more to bad components/ wrong component values.

Would my replacement of the polystyrene cap with a film cap potentially cause this problem? Or a possibly fake Ca3046 chip? Ive used 1% resistors through pretty much.


Thank you.


Theremin? That definitely some sort of interference going on. Very odd.
duff
Royalston wrote:
Sorry to ask again - Does anyone have any pointers on where to start debugging my unstable LVCO build. I have re-done the wiring, tested for continuity throughout the circuit, referencing the schematic. I have changed power supplies. That all seems solid, however my oscillator pitch fluctuates when I move my hand anywhere near the module (within 10-20 cm).


Have you tried swapping out the frequency pot and/or the trimmers used to tune it up?
Royalston
duff wrote:
Royalston wrote:
Sorry to ask again - Does anyone have any pointers on where to start debugging my unstable LVCO build. I have re-done the wiring, tested for continuity throughout the circuit, referencing the schematic. I have changed power supplies. That all seems solid, however my oscillator pitch fluctuates when I move my hand anywhere near the module (within 10-20 cm).


Have you tried swapping out the frequency pot and/or the trimmers used to tune it up?


Thanks - I'll try this smile
(do you mean try a different value component? I don't think its the pot as Ive tried two different types of pots and have 2 builds both acting like this. I haven't changed the trimmers though.

makeshift theremin! sad banana
duff
Royalston wrote:

Thanks - I'll try this smile
(do you mean try a different value component? I don't think its the pot as Ive tried two different types of pots and have 2 builds both acting like this. I haven't changed the trimmers though.

makeshift theremin! sad banana


Not a different value, I was just thinking you could rule out a dodgy pot or trimmer but if you have 2 doing the same thing then I doubt this is the case.
Benjamin AM
Royalston, how about posting some pictures?
Royalston
BenjaminAM - ok, will do tomorrow.
Royalston
yes, my soldering isn't great smile Not sure if this will help...the best photos I could get with my phone.
For the wiring I'm following what I understand of the wiring diagram:
V/oct tip is wired to V/oct, switch to 0V pcb connection , sleeve to ground.
Wave outputs are wired - tip to respective PCB connection, sleeve to ground. Switch is not connected.
PWM + PW wired: tip to respective pcb connection, sleeve to ground. switch - not sure about this one. I tried a few ways. How should this be wired?







Thanks
sixty_n
is there a short on the transistor nearest to the middle pot?

There's always going to be a big chance of wires touching when you mount everything away from the panel, you should pull the leads through from the back so the components are flush to the board. Also right angle pots with nuts would have worked out better. I like the pink resistors
Royalston
sixty_n wrote:
is there a short on the transistor nearest to the middle pot?

There's always going to be a big chance of wires touching when you mount everything away from the panel, you should pull the leads through from the back so the components are flush to the board. Also right angle pots with nuts would have worked out better. I like the pink resistors


It does look like there is a short but Ive done lots of probing and magnifying glass-ing. no short.

Unfortunately when the first set of my resistors and pots arrived I discovered that many of the resistors didn't fit in the holes. The ones that did had to have bent leads like that. I bought a whole batch of new ones that are the smaller size. For the pots, I bought two whole sets of 16mm right angle pots which fit the PCB nicely...unfortunately they didn't fit my panels d'oh! . so I then bought new ones. There were no right angled ones available from tayda in 9mm so I bought these cheap plastic ones. I have since realised they don't work with nuts and washers so thats a whole new problem to deal with later. very frustrating

My main focus atm it to get one core working properly - to stop it acting like a theremin/ having the pitch wobble when I plug a cable into another module in the same rack.

And the pink resistors - they're brown!
sixty_n
It is indeed a strange one. I think you can rule out bad ICs. This happens with two so you can rule out bad solder connections. As it's so close to working properly you can more or less rule out bad / wrong values of components as well. And you tried it on two power supplies so that's basically everything ruled out.

I would check the positive, ground and negative voltages of your power rails without anything plugged in then plug in the module and test them again. If it's changing ground to something other than 0V or offsetting the other voltages it would cause strangeness. A bit of a longshot though

That's a shame about the resistors being brown, I was already thinking of getting some for red circuitboards
Royalston
Thanks sixty n.

I was thinking you were trying to restart the whole gold/ blue dress thing but with pink resistors!

Its weird that the other modules int he case are working fine...its definitely something to do with grounding/ power....I plug a cable in on the other side of the rack (doepfer Low cost case) and the pitch wobbles. the cable isn't even plugged into anything else. Other modules work as normal.
Royalston
Also, my wiring - does this seem right?

V/oct tip is wired to V/oct, switch to 0V pcb connection , sleeve to ground.
Wave outputs are wired - tip to respective PCB connection, sleeve to ground. Switch is not connected.
PWM + PW wired: tip to respective pcb connection, sleeve to ground. switch - not sure about this one. How should these be wired?

Thanks
sixty_n
that sounds right but I haven't built this and don't have boards so I don't about the switches. I was just looking over the thread in case I can get a spare set from Seth. Wiring switches and jack sockets wrongly wouldn't explain the pitch fluctuations though.

I hope you get to the bottom of this
Royalston
Thanks Sixty n.

Not sure where to go from here. I've been double checking the resistor values/ measuring them off unsoldered cores. Tried swapping out the tempco too. I measure the power supply voltage with the module in and out - no difference. I can try getting some proper polystyrene caps...not sure if that will make much difference though.

Im still not 100% my wiring is right but I don't have any way to confirm.
ThecureForSin
Royalston wrote:
Thanks Sixty n.

Not sure where to go from here. I've been double checking the resistor values/ measuring them off unsoldered cores. Tried swapping out the tempco too. I measure the power supply voltage with the module in and out - no difference. I can try getting some proper polystyrene caps...not sure if that will make much difference though.

Im still not 100% my wiring is right but I don't have any way to confirm.


Wish someone would do a proper wiring diagram.
logicgate
Guys, the wiring on this thing is very simple and self-explanatory...

There's a 0V pad on the PCB, you just need to run a wire from it to all ground lugs of your jacks..

Then there's pads for the inputs and outputs, you only need to wire them to the respective signal lug of your jacks.. I can't see how could someone screw this up, it's the easiest part of the build. I will say it again: check the wiring diagrams at http://www.jhaible.com/living_vcos/jh_living_vcos.html



@Royalston


You seem to have a inverted transistor (Q5) right next to the LM311, probably it's the thing that's causing your trouble... confused
ThecureForSin
logicgate wrote:
Guys, the wiring on this thing is very simple and self-explanatory...

There's a 0V pad on the PCB, you just need to run a wire from it to all ground lugs of your jacks..

Then there's pads for the inputs and outputs, you only need to wire them to the respective signal lug of your jacks.. I can't see how could someone screw this up, it's the easiest part of the build. I will say it again: check the wiring diagrams at http://www.jhaible.com/living_vcos/jh_living_vcos.html



@Royalston


You seem to have a inverted transistor (Q5) right next to the LM311, probably it's the thing that's causing your trouble... confused


Im only stuck on the driver integration. I keep thinking I get it then don't.
Royalston
Thanks Logicgate...I think the reverse transistor might be an optical illusion - but I REALLY hope that is the problem! I'm close to paying someone to look at it and try to spot my error.

As for the wiring...its not that self explanatory for all of us smile Thanks for the explanation.
logicgate
Royalston wrote:
Thanks Logicgate...I think the reverse transistor might be an optical illusion - but I REALLY hope that is the problem! I'm close to paying someone to look at it and try to spot my error.

As for the wiring...its not that self explanatory for all of us smile Thanks for the explanation.



Cool, check Q5... It was the first thing I noticed on your pic. The flat side is facing the round part... Always double check the orientation of the components before soldering thumbs up
Benjamin AM
Royalston. This is my final bit of advise. You need to isolate the VCO to be sure that this is indeed a problem with this circuit. It is always nice to have a spare power supply when troubleshooting. If you don't, unplug all of your other modules and make sure that the VCO is not close to the power supply. Do not plug any other modules into the CV input, that is, use the tune knob to set Frequency. Does it now fluctuate? Does is act like an antennae? There should be small and gradual fluctuations but this is due to the expo converter design, giving the character of a more organic sounding oscillator. This purposeful design by Haible.
Troubleshooting becomes more problematic when you decide to place your components in non conventional ways. You may what to try to get the components more flush the board to eliminate variables such as unwanted component interaction.
Also, I am curious what Tempco you used? Is it PTC or NTC? I only bring this up because I think that there is a small possibility that a hand gesture that is moving around air over the temperature sensitive transistors(CA3046 expo converter) on a board that is not at all enclosed could be confused for the theremin effect that you mentioned earlier. Just a thought.
I'm sure you already know this but the transistors need to be BC550C and BC560C. The "C" is important.
Did you clean the flux from the board before installing the pots and ICs?
As you may know, flux can cause some irritating problems especially on boards where the leads are so close together.
And finally, I cannot overemphasize looking at Haible's original documentation. Go to his website, I also placed links to the schematics and wiring diagram earlier on this tread.
ThecureForSin
ThecureForSin wrote:
logicgate wrote:
Guys, the wiring on this thing is very simple and self-explanatory...

There's a 0V pad on the PCB, you just need to run a wire from it to all ground lugs of your jacks..

Then there's pads for the inputs and outputs, you only need to wire them to the respective signal lug of your jacks.. I can't see how could someone screw this up, it's the easiest part of the build. I will say it again: check the wiring diagrams at http://www.jhaible.com/living_vcos/jh_living_vcos.html



@Royalston


You seem to have a inverted transistor (Q5) right next to the LM311, probably it's the thing that's causing your trouble... confused


Im only stuck on the driver integration. I keep thinking I get it then don't.


Let me be more specific with what I'm stuck on. Was typing on my phone and get annoyed easily.

On the driver board: There are 3 jumpers for Sine Out/Tri In; What is the purpose of these? I have 4 pads for "Drive out" So I'm assuming one to the jack, other 3 to each core or.... the 3 switches?

On the core: Are the extra jumpers necessary at the bottom of the board? (GND, V/O, FM, PWM, PLS, Saw, TRI)

Hopefully that helps. I'm not confused about each section individually, it's just understanding how the switches and drive tie into eachother. I do follow Jurgen's diagram, just don't get it on these pcbs in specific. very frustrating
mush
Tri in from osc-core - Sine out. The triangle-sine converters are added and not part of JH's design - and these are on the driver PCB.

The double "jumpers" or connections are due to the two people who designed the PCB had different ideals on how it should be done, iirc.

Edit: The switch is for either driver or 1v/oct jack on each core - so driver out to a switch.
ThecureForSin
mush wrote:
Tri in from osc-core - Sine out. The triangle-sine converters are added and not part of JH's design - and these are on the driver PCB.

The double "jumpers" or connections are due to the two people who designed the PCB had different ideals on how it should be done, iirc.

Edit: The switch is for either driver or 1v/oct jack on each core - so driver out to a switch.


Thanks man! Just want to make sure I understand the driver... So it would be this x3:
Driver Out (on driver pcb) -> 1st switch pin
Core Driver Out -> 2nd switch pin
Jop
I'm also struggling with my build. I have just completed a second core and experience the same issue of poor tracking like I did on the first core very frustrating

Quote:
Everything seems to work well except the scaling is not good at all. Pitch is stable but the tracking is poor, no matter what I do with R9. It keeps in tune maybe for one octave at most???? I have made some pictures, to show how the 10-turn pots fits and hopefully somebody have some advice for my poor tracking.


Quote:
Just had a good look at my board, checked for continuity, bridges, resistor codes / cap values but no oddities found... All components are tested before they are being soldered.

Flux was washed away and board looks clean

I have used C variants for all transistors.


CA3046 is from Intersil , Tempco is from KRL. Changes done:
- removed the TL072 near the frequency pot to stripboard, this due to the fact I would like to instal 10-turn pots.
- Performed the mods described by Benjamin AM for the FM & PWM inputs.
-For the second core I matched Q1/Q2 & Q4/Q5

What would be the best option now the check first? My guess is to undo the FM input mod and see if this makes any change?

Thanks.
ThecureForSin
Jop wrote:
I'm also struggling with my build. I have just completed a second core and experience the same issue of poor tracking like I did on the first core very frustrating

Quote:
Everything seems to work well except the scaling is not good at all. Pitch is stable but the tracking is poor, no matter what I do with R9. It keeps in tune maybe for one octave at most???? I have made some pictures, to show how the 10-turn pots fits and hopefully somebody have some advice for my poor tracking.


Quote:
Just had a good look at my board, checked for continuity, bridges, resistor codes / cap values but no oddities found... All components are tested before they are being soldered.

Flux was washed away and board looks clean

I have used C variants for all transistors.


CA3046 is from Intersil , Tempco is from KRL. Changes done:
- removed the TL072 near the frequency pot to stripboard, this due to the fact I would like to instal 10-turn pots.
- Performed the mods described by Benjamin AM for the FM & PWM inputs.
-For the second core I matched Q1/Q2 & Q4/Q5

What would be the best option now the check first? My guess is to undo the FM input mod and see if this makes any change?

Thanks.


Poor tracking... Are you sure you have the correct value tempco?
logicgate
@Jop

What do you mean by poor tracking? My LVCOs play at least 6 octaves, maybe more.

But they don't stay in tune with each other for long when playing together, depending of the note, the drift apart by some cents.

What I'm gonna try to do is replacing and matching the resistors for all 3 oscillators so they all have the exact same values. Not ALL resistors but the ones in the CV chain.
Royalston
Benjamin AM wrote:
Royalston. This is my final bit of advise. You need to isolate the VCO to be sure that this is indeed a problem with this circuit. It is always nice to have a spare power supply when troubleshooting. If you don't, unplug all of your other modules and make sure that the VCO is not close to the power supply. Do not plug any other modules into the CV input, that is, use the tune knob to set Frequency. Does it now fluctuate? Does is act like an antennae? There should be small and gradual fluctuations but this is due to the expo converter design, giving the character of a more organic sounding oscillator. This purposeful design by Haible.
Troubleshooting becomes more problematic when you decide to place your components in non conventional ways. You may what to try to get the components more flush the board to eliminate variables such as unwanted component interaction.
Also, I am curious what Tempco you used? Is it PTC or NTC? I only bring this up because I think that there is a small possibility that a hand gesture that is moving around air over the temperature sensitive transistors(CA3046 expo converter) on a board that is not at all enclosed could be confused for the theremin effect that you mentioned earlier. Just a thought.
I'm sure you already know this but the transistors need to be BC550C and BC560C. The "C" is important.
Did you clean the flux from the board before installing the pots and ICs?
As you may know, flux can cause some irritating problems especially on boards where the leads are so close together.
And finally, I cannot overemphasize looking at Haible's original documentation. Go to his website, I also placed links to the schematics and wiring diagram earlier on this tread.


@benjaminAM thanks for your response:
Today I also soldered all the wiring together rather than using banana jacks. This has increased the stability of the pitch quite a bit, unless I touch the cables going into the module - then the pitch wobbles. It does not seem to be acting like an antennae anymore. There is also still the pitch change when I move the fm and pwm pots.

In answer to your questions:
I did have one transistor backwards on one core, the other was correct already- resoldered- no change! (this is weird in itself!)
Transistors: BC560CTA and BC550CBU
Tyco PTC tempco resistors 1k - I also tried the circuit with a normal resistor, it seemed to behave much the same.
INTERSIL/HARRIS DIP-14 CA3046 CA3046E - I bought these off Ebay. Maybe they are no good?

I have tried different isolated power supplies, also - tried with only the saw output attached (no other jacks).

Apart from my lack of knowledge in electronics I find the Jurgens schematics hard to compare with the barcode pcbs as the components are all numbered differently...drove me a bit mad until I figured this out.

Anyone interested in debugging a core of this for me for $? I will pay all the costs!
Jop
@ ThecureForSin => Yes, the TC is the correct value of 1K

@ logicgate => On both my builds i cannot achieve proper tracking for several octaves, to be more precise not even a few notes apart ?
It stay's nicely in tune on one frequency without wobbles, but drives out of tune when going higher or lower (even one note). You can even hear it without reference that the scaling is off when you play a sequence of sequential notes. Sure I tried to compensate it with R9 without success.

As mentioned before I have only changed some minor things on my build:
Quote:
- removed the TL072 near the frequency pot to stripboard, this due to the fact I would like to instal 10-turn pots.
- Performed the mods described by Benjamin AM for the FM & PWM inputs.


Components used:
* BC560C & BC550C trannies
* 1% resistors & 0,1% resistors for R1/R2/R3
* Polystyrene cap for C3
* All TDK caps for the MLCC's
* CA3046 from Intersil
* 1K TC from KRL
* Texas Instr. LM311P / TL072CP's / TL074CN
* Board is washed, no flux residue

What I have done so far is:

* Build a second core, same result as #1
* Use two other CA3046 IC (from same brand) without a positive effect
* Use two matched 2n3904's in place of the 3046, no change.

I'm not sure now how to continue with debugging.

Hope you have some tips for me thumbs up
Benjamin AM
@Jop
I guess my "mod", which is more of a correction than a modification to the circuit has the potential for messing up tracking if installed incorrectly. Do you have a pic??
Jop
Hi Benjamin, thanks for your reply.

Pictures from my first build are on page 6. I simple use the resistor lead itself with some heat schrink to go directly to the fm & pwm inputs. If needed I can make some better pictures. Do you think it would be wise to change / remove your correction and see if that makes a change to the tracking?
Royalston
I spoke to Barcode (after I had done that mod) and he said his board design didn't need that fix. I took it out and my LVCO worked (but not perfectly).
Benjamin AM
@Jop - Well I think it is always wise to build a module to spec and then change things.As far as I can tell from the photos the mod seems right. Lifted resistor legs go into the center pin of pots. The MTA connector is not used for FM and PMW, instead you solder to CCW of pot. This corrects the the circuit board to form a simple input attenuator. The switching jacks are connected to the sleeve(ground).

I can't stop looking at R13 on your board. From the pictures it appears to be 6.8K. Could be an optical illusion.
jules
Do you remember that mouser order made on a "one beer too many" evening... And the curses you spew when you received all these damned useless 1/8W resistors with annoyingly wobbly legs..
This build is when this order is finally useful.

Otherwise, I hope you still have yoda's phone number, 'cause you'll have to use the force (I feel the humour flowing down my veins on the 1st of april).


Edot: it's not only the holes which are small, but the pads look dangerously close and prone to shorting mistake. Got to be careful.
qfactor
jules wrote:
Do you remember that mouser order made on a "one beer too many" evening... And the curses you spew when you received all these damned useless 1/8W resistors with annoyingly wobbly legs..
This build is when this order is finally useful.

Otherwise, I hope you still have yoda's phone number, 'cause you'll have to use the force (I feel the humour flowing down my veins on the 1st of april).


Lucky you! I'm still waiting for the day I can use my 50 pcs of, wrongly ordered, 1W 1K resistors!! d'oh!
Benjamin AM
I sort of wish that the saw output would have Been biased on these boards. I was in the process of making a small daughterboard to make the saw +/- 5V but then I thought why not just add a cap that spans across the switching jack to the output lug and move the saw out wire to the switching lug and call it good. I used a 1 uF film cap that seems to work well with all of my mixers. I feel that this mod make the saw output more useable especially with DC mixers. The downside is that the purity of the wave might be effected if used as an LFO.
Siri
Benjamin AM wrote:
I noticed an error on the Barcode PCBs. The FM and PWM inputs are connected incorrectly. Currently, the inputs are routed to the middle leg of the panel pot. It should be set up so the inputs are connected to the CW leg of the pots, the summing resistor is on the middle leg and ground is on the CCW leg. This way the pot functions as an attenuator rather than gain control(current setup).


Royalston wrote:
spoke to Barcode (after I had done that mod) and he said his board design didn't need that fix.


eek! i'm lost here
qfactor
Could I get your opinions as to whether using these pots would work and if there should be anything else I may have missed when thinking how "great" an idea this would be (great as in $aving$ for me because I already have all these already!! d'oh! )

I have 2 LVCO (that would be 2 driver boards and 6 VCO boards + 2 panels)


The pot on the lower right is how the pins are oriented. My question is, if I were to bend the pins up, like the one on the top left hand image and mount these to the PCBs, would this then give me more space (since the pins are longer for these "upright" pin type) to mount the 10-turn pot without having to move the TL072 chip that would get in the way of the 10-turn pot?

I have 1 of each version of the LVCO PCBs - 1 of Barcode's version and the other, Seth's
Altitude909
Siri wrote:
Benjamin AM wrote:
I noticed an error on the Barcode PCBs. The FM and PWM inputs are connected incorrectly. Currently, the inputs are routed to the middle leg of the panel pot. It should be set up so the inputs are connected to the CW leg of the pots, the summing resistor is on the middle leg and ground is on the CCW leg. This way the pot functions as an attenuator rather than gain control(current setup).


Royalston wrote:
spoke to Barcode (after I had done that mod) and he said his board design didn't need that fix.


eek! i'm lost here


seriously. The complete lack of documentation has me ready to bin this project
Grumskiz
Yeah, I haven't even started on my LMVCOs Set and I am reluctant to do so as I am very confused about errors that have been reported by some and the black magic that seems to be involved in getting these to work at all :(

Maybe it's just me, but is there anyone who has successfully built the VCO set with the driver circuit based on the new boards (I am aware, that there are at least 3 different versions) in euro and is happy with the result?
If so, I would be very grateful to hear from you. Please share your experiences. I am especially interested in what components should be matched and whether to perform the mods that have been suggested.
I personally own a set of InfraXpert PCBs. From what I gather these are supposed to be free of errors. Correct me if I am wrong.

I'd hate to build this according to the BOM and then realize that it barely works. I would prefer getting it (mostly) right on the first try without having to rework/botch it multiple times.

Here is a list of what I think is very important when building it:
- Only use 550C and 560C variants, because the high hfe is key.(don't know why)
- Use closely matched resistors on the CV inputs on all VCOs to keep them in tune.


Please tell me if I forgot something.
Jop
Quote:
@Jop - Well I think it is always wise to build a module to spec and then change things.As far as I can tell from the photos the mod seems right. Lifted resistor legs go into the center pin of pots. The MTA connector is not used for FM and PMW, instead you solder to CCW of pot. This corrects the the circuit board to form a simple input attenuator. The switching jacks are connected to the sleeve(ground).

I can't stop looking at R13 on your board. From the pictures it appears to be 6.8K. Could be an optical illusion.


Finally got time to experiment with my boards. Re-checked R13, that's 68K so good. Undo your fix on one board and noticed that the tracking improved quite a bit. So I did remove it on the second board as well but noticed only a small improvement in tracking .... seriously, i just don't get it
After a good sleep I redid the calibration, started from the beginning and now the second boards tracks good! SlayerBadger!

I'm not sure what was causing my issue at the start, poor calibration, the fix or both? I'm glad it is working, and they sound very nice and lush..

Some pics from my work in progress.





pulse_divider
Is there any way to adjust the driver out CV?

I got my first core up and running and it's scaling perfectly over 5 octaves via it's own 1v/oct input. When I hook it up to the driver, it goes flat as i increase the input voltage.

The other driver functions (octave, glide, vibrato) seem to be working fine but the driver itself is not putting out unity voltage.
Has anyone else had this problem?
Benjamin AM
Jop wrote:


I'm not sure what was causing my issue at the start, poor calibration, the fix or both? I'm glad it is working, and they sound very nice and lush..

Some pics from my work in progress.

Yes it is always a smart idea to make a solid build before modifying. Otherwise one can easily come to the conclusion that the modification is the problem. I can guarantee that this is not the case with my FM and PWM input modification. Without the modification(Barcode rev1.3) the inputs from the PWM and FM are wired directly to the center lug of the corresponding potentiometer. The CW lug the leads the mixer resistor. CCW to ground. These are improperly wired input attenuators. If you don't believe me, take a look at Haible's wiring diagram on his website. If left PWM and FM inputs are left modified, the external modulation source will be completely off at full CCW and once the knob turned there will be large jump in modulation. If you like this behavior I say leave the circuit board alone. If not, I offer a solution to fix this without having to cut any traces. The CW lug and center lug connections just need to be reversed. Not a big deal. I do stress to only do the mod once you have verified that you circuit is working "properly".
finnurbjarna
Well, I've just finished building this from the pcb and panels I ordered from LeftyLogic. It's easily the most complex build I've tried but, to my surprise, it actually works. Not only that - this thing sounds amazing.

I ordered most of the components from Tayda and what they didn't have I got from Bitsbox and Banzai (plus 3046's off eBay).

I didn't match any resistors but took the advice to socket R39 and have actually ended up with 1.2M in there...

If I were to do it again (and I would) I would perhaps do the following:

- Plan the spacing for the resistors; there is only just enough room for them all to lie flat and I got caught out a couple of times by not doing them methodically.

- Try not to have massive ferrite beads; mine are too big and I had to push them out to make room for the power connectors.

- Find a more elegant and durable solution for the wiring.

Other than that, I found it to be a pretty straight-forward build. I opted for 16mm Alphas and found the design accommodated them very well. The only slightly dodgy bit is where the middle on-on switch lies flat to the driver-pcb, once mounted, and covers the driver-out taps. Also, I put a pair of washers on every pot inside the panel to give a little more space between the pot-lugs and the ground rails for the jacks.

With the VCO cores the wiring was pretty straightforward but it took me a little while to puzzle out the driver connections (I'm too stupid to understand the circuit diagram).
I think I've done it correctly (at least it all seems to work) and can maybe save someone some time with this:

The octave leads go to the on-off-on switch but when the and board and switch are both mounted it's the wrong way around, i.e. the topmost connection on the board goes to the lowest connector on the switch, middle to middle and then the two closest connect. Otherwise 'Octave+' drops the pitch and vice versa.

There are 4 Driver-out connections; one goes to the Driver jack and the other 3 go to the 3 on-on switches. When board and switches are mounted on the panel the bottom connector on the switch connects to the Driver-out connections on the Driver, the middle connector goes to the driver tap on each VCO core respectively and the top tap connects to 0V.

Someone has already explained the other taps, which are for the sine wave.

All in all, if I can do this anyone can. Please point out any errors and I will post a couple of pics for your amusement... hihi

A huge thanks to everyone for making this wonderful VCO a possibility for people like me. love



Grumskiz
Great to hear, that yours works perfectly smile Gives me back some confidence in this project.
Your build looks great!
duff
I'm just going through my populating 3 of my VCO boards, but have had a nightmare with R26 on one of them. I made a bad decision to try and remove it to re-seat it better and ended up knackering the holes d'oh! Fixing it on one side is easy enough since it connects to R25, but I can't for the life of me work out where the other side should go. Looking at Juergen's schematic is states -b1 which doesn't seem to help me as it doesn't seem to connecting to anything else also marked with -b1. I can't see any trace on the boards and it doesn't appear to be connecting to the ground plane. Can anyone tell me where it should go?
Benjamin AM
duff wrote:
I'm just going through my populating 3 of my VCO boards, but have had a nightmare with R26 on one of them. I made a bad decision to try and remove it to re-seat it better and ended up knackering the holes d'oh! Fixing it on one side is easy enough since it connects to R25, but I can't for the life of me work out where the other side should go. Looking at Juergen's schematic is states -b1 which doesn't seem to help me as it doesn't seem to connecting to anything else also marked with -b1. I can't see any trace on the boards and it doesn't appear to be connecting to the ground plane. Can anyone tell me where it should go?

I think that you are misreading the schematic. It should be -Vb1. -V means the negative rail. -Vb1 probably means negative voltage on board 1, that is I assume that you're looking at the 1st schematic on his page
duff
Benjamin AM wrote:
I think that you are misreading the schematic. It should be -Vb1. -V means the negative rail. -Vb1 probably means negative voltage on board 1, that is I assume that you're looking at the 1st schematic on his page


That is what I expected, everything else marked -b1 can be seen to have continuity with -ve power header (or at least the bead on the negative rail as I haven't put that in yet), but R26 appears to be hanging on all (or at least 4) or the VCO boards I have as far as I can see.
qfactor
finnurbjarna wrote:
Well, I've just finished building this from the pcb and panels I ordered from LeftyLogic. It's easily the most complex build I've tried but, to my surprise, it actually works. Not only that - this thing sounds amazing.


Great to hear of your "adventures"! thumbs up
One thing I'm curious is about the trimmers.
are you using the top adjusting types (3296W) or the side adjusting ones (3296X)?
If it's the top adjusting ones, how much problems did you have accessing the inner cores when calibrating?

Mine is still far away from completion as I've just started working on the Driver board, while trying to anticipate any space issues I might come up to, as I complete more of it! eek!

Thanks! thumbs up
finnurbjarna
I just used the top-adjusting ones listed in the BOM. Looking back, side-adjusting ones sound like a very good idea.

To get around the calibration/space problem I used jumpers for all the VCO wiring so that removing them is at least easy. I mounted the first and tuned it, then the second, and so on.

I'm astounded at how well this thing tracks - it's good over the complete range. The 3 cores also stayed very well in tune with each other all afternoon/evening yesterday and were still in tune when I powered up this morning. It's a lot better than I was expecting.

BTW it's worth being relatively awake when you start mounting and powering up the boards because the driver-board power-header has negative at the top whereas the cores have it at the bottom.
Benjamin AM
I decided to put some multi turn Frequency pots in. The spacing is tight but definitely worked out. I did have to remove the IC behind the frequency pot but I came up with a nice solution that I would like to share. Basically I mirrored the TL072s footprint by inverting the pins. I threw a DIP socket on the back of the board and plugged the upside down IC. Easy enough but make sure your board works before trying this.


Barcode
Hey folks. I've been MIA from the forums for a bit, only popping in every once in a while. It was brought to my attention that a few people have been having some problems with building the cores. I haven't had a chance to going over this entire thread. I did notice a couple errors that I completely missed during my testing..... By the way, my complete VCO sounds fantastic with a couple hiccups.

First, the PWM and FM pots are set up incorrectly like stated Benjamin Am. This is an easy fix by cutting a couple traces and using oops wires.

Second, the PW pot is the wrong value. Using a 1M pot like shown in the BoM causes the pulse wave to drop out at full CCW. Even more so when using PWM. Looking over the original BoM this pot should be 100k.

I was planning on building up a couple more of the cores over the next year for a stand alone synth but it looks like I'm going to expedite the builds. I will work to document my build process with fixes as best as possible and keep it updated on my blog and here.

Again I apologies for any issues folks have had on their builds and I will do all I can to help out.
logicgate
@Barcode

Nope, you have to use 1M for the PW pot, otherwise it won't work, and I have no clue why. 100K is for the 15V build, for 12V you have to use 1M.

Then you have to socket R39 and keep swapping resistors (while adjusting the trimmer) to find the a value that will give you the PW range from 10% to 90% and it won't drop out.
jules
Barcode wrote:


First, the PWM and FM pots are set up incorrectly like stated Benjamin Am. This is an easy fix by cutting a couple traces and using oops wires.


Could you elaborate which traces to cut, and wires from where to where?

I though it was R39 that was to be socketed...?

Are the schematics from J.Haible page reliable with the 2 batches of PCBs (leftylogic and logicgate/barcode), cause I'm confused to say the least.

Thanks.
finnurbjarna
I'm a bit confused about this PWM/FM pot issue;

I soldered the pots onto the black pcb's as they were and everything seems to be working fine - the pots are attenuating whatever I plug into the PWM/FM jacks.

So is this working in spite of being wrongly connected? And if it is something that you advise I change, is there any chance of a pic showing best practise?
logicgate
@jules

Yes it's R39, sorry for The typo, i corrected my post
logicgate
jules wrote:


Are the schematics from J.Haible page reliable with the 2 batches of PCBs (leftylogic and logicgate/barcode), cause I'm confused to say the least.

Thanks.



You are confused about what, particularly? The VCO core boards are exactly as the schematics (in fact, I even used some scans of the original LVCO PCBs to compare with mine, and make some corrections), and I only added the saw to tri converter to it.
pulse_divider
pulse_divider wrote:
Is there any way to adjust the driver out CV?

I got my first core up and running and it's scaling perfectly over 5 octaves via it's own 1v/oct input. When I hook it up to the driver, it goes flat as i increase the input voltage.

The other driver functions (octave, glide, vibrato) seem to be working fine but the driver itself is not putting out unity voltage.
Has anyone else had this problem?


Just bumping this, mine is all finished and everything works as expected except that the driver is causing the pitch to go flat (~10 cents octave) as I go up the scale. If I bypass the driver and plug directly into the VCOs they track perfectly across 5 octaves.

There's not much builder feedback on the driver circuit, has anyone built theirs successfully and had it scale properly?
Benjamin AM
@ Logicgate & Barcode- I don't think that the value of the Pulse Width pot is that important. In fact I would be willing to bet that it would not change anything if I were swap it out. It is just a voltage divider. I feel that the Pulse width mixer has become a bit unbalanced(Q7 and surrounding resistors). Returning three resistors to values suggested by Haible got me to a balanced PW circuit.
Here is my original post.
Benjamin AM wrote:
Sorry to bring this up so late in the game. I found that the Barcode BOM v1.3 had some values that did not work my build. I feel I should share my experience in case anyone out there has had the same issue. In regards to the PWM section of the circuit, resistors R39 and R36 were troublesome for me. I came to this conclusion because I felt that the PWM wasn't working as advertised because the pulse wave was cutting out even before full CW and CCW. This circuit is not supposed to be like this. The diodes are supposed to protect the pulse from completely shutting off. This means to me that the gain and bias on the transistor at the summing stage of the PWM circuit is to high. So all that I did to change this was return R36's value to the suggested value by Haible(750k opposed to the Barcode BOM 1.5M). Then the PWM mixer resistor(R39) back to the original value 1M, instead of Barcode BOM value(510k). I also found it nice to return one of the biasing resistors(R33) back to original value 20k. Now all is good for me.
R33=20k
R39=1M
R36=750k

I guess there is more than one way to skin a transistor.
logicgate
@Benjamin AM

Glad you got it working.

FIY, the resistor values that were changed (to make it work properly with 12V) were taken from electro music forum, at the LVCOs thread, and were suggested by Haible, if I´m not mistaken. I just copied those values from there...
jules
logicgate wrote:
jules wrote:


Are the schematics from J.Haible page reliable with the 2 batches of PCBs (leftylogic and logicgate/barcode), cause I'm confused to say the least.

Thanks.



You are confused about what, particularly? The VCO core boards are exactly as the schematics (in fact, I even used some scans of the original LVCO PCBs to compare with mine, and make some corrections), and I only added the saw to tri converter to it.


HA! Thanks for the clarification. It's sounds stupid, but I couldn't find this bit of info anywhere. It probably also sounds obvious to you, but it wasn't for me.

And then there's a bit of confusion as to what mods have to be applied to which version of the PCB... There's 3 versions, apparently some do require, and some don't require modifications.
This situation is stirring up a bit of confusion in my simple mind.
I guess finnurbjarna has the lefty logic version of the PCB, and these work fine out of the box (just the R39 to fiddle with).
Altitude909
any reason there isnt a trimmer in R39?
logicgate
Altitude909 wrote:
any reason there isnt a trimmer in R39?


Because the PCBs were already manufactured when I found out that R39 was the cause of my PW issue...

Anyways, you can kludge a trimmer in there, if you feel like thumbs up
qfactor
logicgate wrote:
Altitude909 wrote:
any reason there isnt a trimmer in R39?


Because the PCBs were already manufactured when I found out that R39 was the cause of my PW issue...

Anyways, you can kludge a trimmer in there, if you feel like thumbs up


If it were to be a trimmer for R39, would it be only the wiper and one side being used (of the 3 pins on a trimmer)?
And would a 2M value trimmer be suitable?
Altitude909
qfactor wrote:
logicgate wrote:
Altitude909 wrote:
any reason there isnt a trimmer in R39?


Because the PCBs were already manufactured when I found out that R39 was the cause of my PW issue...

Anyways, you can kludge a trimmer in there, if you feel like thumbs up


If it were to be a trimmer for R39, would it be only the wiper and one side being used (of the 3 pins on a trimmer)?
And would a 2M value trimmer be suitable?


Yes. Technically a rheostat
jules
My inner buddha is gone.
I'm blowing a fuze 'cause of these ridiculously small holes.

1/8W Vishay (RN55D) resistors are too big for them.
And now I've got a kilo of them on my hands.

I don't use the little faces, but angry and very frustrating is appropriate.
Altitude909
1/8 watt RN Vishay parts are the size of everyone else's 1/4 watt parts. Try 1/8 watt xicon
the bad producer
I've tried many different resistors for R33, R34, R36, R38 and R39 - just for the hell of it I think as I didn't really need to!

I've found it works best with what Benjamin AM selected and mentioned earlier in the thread - the values that Haible suggested for R33, R36, R39.

FWIW I've used a 100K pot, I may try a 1M just for a test... If anyone is interested I could do a sound demo of the original Haible PCB version running at 15V and this PCB at 12V?




Benjamin AM wrote:
@ Logicgate & Barcode- I don't think that the value of the Pulse Width pot is that important. In fact I would be willing to bet that it would not change anything if I were swap it out. It is just a voltage divider. I feel that the Pulse width mixer has become a bit unbalanced(Q7 and surrounding resistors). Returning three resistors to values suggested by Haible got me to a balanced PW circuit.
Here is my original post.
Benjamin AM wrote:
Sorry to bring this up so late in the game. I found that the Barcode BOM v1.3 had some values that did not work my build. I feel I should share my experience in case anyone out there has had the same issue. In regards to the PWM section of the circuit, resistors R39 and R36 were troublesome for me. I came to this conclusion because I felt that the PWM wasn't working as advertised because the pulse wave was cutting out even before full CW and CCW. This circuit is not supposed to be like this. The diodes are supposed to protect the pulse from completely shutting off. This means to me that the gain and bias on the transistor at the summing stage of the PWM circuit is to high. So all that I did to change this was return R36's value to the suggested value by Haible(750k opposed to the Barcode BOM 1.5M). Then the PWM mixer resistor(R39) back to the original value 1M, instead of Barcode BOM value(510k). I also found it nice to return one of the biasing resistors(R33) back to original value 20k. Now all is good for me.
R33=20k
R39=1M
R36=750k

I guess there is more than one way to skin a transistor.
the bad producer
Here it is anyway, see comments for comments:

[soundcloud url="https://api.soundcloud.com/tracks/199681143" params="auto_play=false&hide_related=false&show_comments=true&show_use r=true&show_reposts=false&visual=true" width="100%" height="450" iframe="true" /]
the bad producer
Should have mentioned, the first two parts are manual PWM manipulation, the rest are PWM CV and then both (tweaking PW knob with CV)...
FingerTappin
the bad producer

Jeez, that demo sounds lush!


I am excited to build these, but am waiting until more documentation/mods/fixes are out. I am not the best trouble shooter.

Great to see what is going on in this thread though.
Siri
@thebadproducer

waw nice demo w00t

@fingertappin

I guess it starts now Mr. Green Benjamin AM and thebadproducer got their builds working nicely with same tweaks/pot values/resistor values
FingerTappin
Siri

It seems so! Some nice builds and some great sounds coming about.

I like Benjamin AM's idea about the TL072 on the back of the PCB to make room for a 10 turn. Wasn't really looking forward to fly wiring the chip off board.
qfactor
FingerTappin wrote:
I like Benjamin AM's idea about the TL072 on the back of the PCB to make room for a 10 turn. Wasn't really looking forward to fly wiring the chip off board.


Hear, hear!
So i guess this would require flipping all the pins upwards (with a long nose plier) and mounting the topside of the chip on the underside of PCB eh? Would kinda look like when a spider takes a long dive, with its legs all up? zombie

Of course mounting it to a socket on the underside of the PCB would be safer too!! thumbs up
qfactor
Can a thermistor be used as a substitute for Tempco?
hmmm.....

elmegil
my understanding is that the temperature coefficient moves in the opposite direction from what you need in a thermistor.

A tempco increases its resistance as the temp goes up.
qfactor
elmegil wrote:
my understanding is that the temperature coefficient moves in the opposite direction from what you need in a thermistor.

A tempco increases its resistance as the temp goes up.


Oh, so a thermistor is the opposite of a tempco??! d'oh!
elmegil
qfactor wrote:
elmegil wrote:
my understanding is that the temperature coefficient moves in the opposite direction from what you need in a thermistor.

A tempco increases its resistance as the temp goes up.


Oh, so a thermistor is the opposite of a tempco??! d'oh!


Don't worry, I have a bag of tiny glass ones that I did the same thing with smile
ZibraZibraZibra
so been having fun with a completed single vco module.

spent a bit troubleshooting it, i couldn't work out the issue, finally it was the 10 turn pot. it wasn't working at all! the oddest thing of all is the resistance would change on the multimeter but not the actual frequency, so i swapped it out.

my plan was to build a couple of solo VCO's first then build the nice big 3 vco with driver after working out all the things i want to change.

but ok to get to the bottom of this I'm using BOURNS 10 turn pots, i bought 8 of them overall and they are crap! REALLY CRAP. anybody else had experience with them?

the first one like i said didn't work, and the one i have in now does work however if you spin it a little too quick it sounds like it misses a gear and goes down again..... i don't know maybe I'm doing something wrong but it just seems like they are crap?

suggest any better?
qfactor
can someone clarify what does the driver pcb actually do in relation to the 3 cores? hmmm.....
the bad producer
Sounds very surprising! Have you wired them up correctly? The pin numbering is counterintuitive. They shouldn't be crap at all!


ZibraZibraZibra wrote:
so been having fun with a completed single vco module.

spent a bit troubleshooting it, i couldn't work out the issue, finally it was the 10 turn pot. it wasn't working at all! the oddest thing of all is the resistance would change on the multimeter but not the actual frequency, so i swapped it out.

my plan was to build a couple of solo VCO's first then build the nice big 3 vco with driver after working out all the things i want to change.

but ok to get to the bottom of this I'm using BOURNS 10 turn pots, i bought 8 of them overall and they are crap! REALLY CRAP. anybody else had experience with them?

the first one like i said didn't work, and the one i have in now does work however if you spin it a little too quick it sounds like it misses a gear and goes down again..... i don't know maybe I'm doing something wrong but it just seems like they are crap?

suggest any better?
the bad producer
There's lots of info in the original thread over on e-m, in fact I would recommend anyone building this to read through it:

http://electro-music.com/forum/topic-30749.html&postorder=asc

(for eg):

jhaible wrote:
there is some simple octave switching in the driver section (one oct up, normal, one oct down) so you can quickly transpose the whole set of 3 oscillators. Also, a fine tune pot that works on all 3 oscillators in common


jhaible wrote:
BTW, there is an on-board Vibrato LFO, in the Oscillator Driver section.
Sine modulation of the Pitch for all 3 VCOs together. Voltage controlled modulation depth. Manual Rate control that is deliberately limited to 2Hz ... 12Hz, so you get a very fine resolution of the useful range.
You can get any other range by choosing different component values, of course - so you don't have to limit the range.
But here also, this is not intended to replace a standalone LFO for all kinds of weird modulation - it's just an on-board VC-Vibrato, so you don't need to scarifice an extra LFO for this function. (Just plug Aftertouch voltage to the CV input, and that's it for touch controlled Vibrato.)


qfactor wrote:
can someone clarify what does the driver pcb actually do in relation to the 3 cores? hmmm.....
qfactor
Cool! Thanks for the heads up! thumbs up
Will check out the rest of it on e-m
qfactor
I don't suppose the 1k tempco could be substituted with a 2k tempco, eh? seriously, i just don't get it
livefreela
sounds like an incorrectly wired pot to me - the legs on multi turns are often weirdly layed out in comparison to the normal guys. i have some (i think bourns) where the wiper was not the center terminal, i got them surplus so the markings weren't visable either. have a check with your meter...

ZibraZibraZibra wrote:
so been having fun with a completed single vco module.

spent a bit troubleshooting it, i couldn't work out the issue, finally it was the 10 turn pot. it wasn't working at all! the oddest thing of all is the resistance would change on the multimeter but not the actual frequency, so i swapped it out.

my plan was to build a couple of solo VCO's first then build the nice big 3 vco with driver after working out all the things i want to change.

but ok to get to the bottom of this I'm using BOURNS 10 turn pots, i bought 8 of them overall and they are crap! REALLY CRAP. anybody else had experience with them?

the first one like i said didn't work, and the one i have in now does work however if you spin it a little too quick it sounds like it misses a gear and goes down again..... i don't know maybe I'm doing something wrong but it just seems like they are crap?

suggest any better?
livefreela
i'd venture to say that maybe a 2ktempco with a 1k resistor in parallel if your really hard up will probably "go" - but that's probably squarely in the school of adventurous kludgery!
qfactor wrote:
I don't suppose the 1k tempco could be substituted with a 2k tempco, eh? seriously, i just don't get it
elmegil
livefreela wrote:
i'd venture to say that maybe a 2ktempco with a 1k resistor in parallel if your really hard up will probably "go" - but that's probably squarely in the school of adventurous kludgery!
qfactor wrote:
I don't suppose the 1k tempco could be substituted with a 2k tempco, eh? seriously, i just don't get it



Presuming you meant in series.... still a bad idea.

I've accidentally subbed 1K for 2K tempcos (or vice versa, I forget) and it totally blows the tracking. You can calibrate for 2V/Oct or 1/2V/Oct but never will reach 1V/Oct smile
qfactor
LOL!! hyper

Ok, I get it! Not something to mess around with if one wants standard 1V/Oct tracking!! d'oh!
jules
logicgate wrote:
jules wrote:


Are the schematics from J.Haible page reliable with the 2 batches of PCBs (leftylogic and logicgate/barcode), cause I'm confused to say the least.

Thanks.



You are confused about what, particularly? The VCO core boards are exactly as the schematics (in fact, I even used some scans of the original LVCO PCBs to compare with mine, and make some corrections), and I only added the saw to tri converter to it.


I don't know if we are talking about the same schematics, but I'm referring to the ones on JH page for the LVCO.

I just can't find the 2nd TL072 and the TL074 on the schematics of the core, and I'm wondering which TL072 is which (U1 on JHs), plus what's the purpose of the 074. And where do you get the tri wave from?

I've got oscillation but a nasty transformation of the waveforms on one of the cores. What comes out of saw is more or less a ramp, the tri is a ramp with an exponential front and everything reacts to PW...
It's not the 3046 or any of the TLs..
logicgate
jules wrote:
logicgate wrote:
jules wrote:


Are the schematics from J.Haible page reliable with the 2 batches of PCBs (leftylogic and logicgate/barcode), cause I'm confused to say the least.

Thanks.



You are confused about what, particularly? The VCO core boards are exactly as the schematics (in fact, I even used some scans of the original LVCO PCBs to compare with mine, and make some corrections), and I only added the saw to tri converter to it.


I don't know if we are talking about the same schematics, but I'm referring to the ones on JH page for the LVCO.

I just can't find the 2nd TL072 and the TL074 on the schematics of the core, and I'm wondering which TL072 is which (U1 on JHs), plus what's the purpose of the 074. And where do you get the tri wave from?

I've got oscillation but a nasty transformation of the waveforms on one of the cores. What comes out of saw is more or less a ramp, the tri is a ramp with an exponential front and everything reacts to PW...
It's not the 3046 or any of the TLs..


Those ICs are part of the Saw to Tri converter circuitry, it's not on the original schematics.
jules
Anyway to get the schematics?
I'm troubleshooting blind... Which might be considered a sport, but all in all very annoying.
logicgate
@jules

Attached to this message are the schematics, I've placed a red rectangle around the Saw to Tri converter, hope this helps.
jules
Done. Shortcut between atom size pad of transistors.
The oscs sound really good, but I'm sorry to say that it was not a pleasant build.
You need thin solder, thin solder iron tips, and some desoldering braid to eventually remove the excess of solder overflowing on adjacent pads.
Appart from the headache of finding the right size components, I unluckily went through 3 faulty CA3046.
The other thing: you can put the trimmers of the driver board on the other side of the PCB if you want access to them.

logicgate
Aw come on jules, wasn't this rough ride worth it?? hihi lol This is fun!

Congrats, now you have the lushiest of oscillators! SlayerBadger!
sammy123
Nice work Jules. I didn't think this build was very difficult, but I already had thin resistors, solder, and solder tip. I think having a fine tip is essential for SDIY anyway. You're wiring looks great. Nice and neat. The total opposite of mine.
duff
My 100K 0.1% resistors aren't going to go through the holes on this so I was planning to match some 100Ks. My assumption is that I just need to make sure the 3 used for each VCO match and that they don't need to match between VCO's - is that right?

Also, as per my previous post, I still can't see R26 connecting back to the -ve rail. Can anyone else confirm this?
qfactor
duff wrote:
My 100K 0.1% resistors aren't going to go through the holes on this so I was planning to match some 100Ks. My assumption is that I just need to make sure the 3 used for each VCO match and that they don't need to match between VCO's - is that right?


I thought I read in one of the (3) threads for the LVCO that it would be "better" to match the 100k for ALL 3 cores? All of the 9 100ks! hmmm.....
Anyway, that's what I did. seriously, i just don't get it
duff
qfactor wrote:
duff wrote:
My 100K 0.1% resistors aren't going to go through the holes on this so I was planning to match some 100Ks. My assumption is that I just need to make sure the 3 used for each VCO match and that they don't need to match between VCO's - is that right?


I thought I read in one of the (3) threads for the LVCO that it would be "better" to match the 100k for ALL 3 cores? All of the 9 100ks! hmmm.....
Anyway, that's what I did. seriously, i just don't get it


Fortunately it would appear that out of the 30 or so 100K resistors (from 600+) that I have with the right lead width there are at least 9 +/- 0.1% of each other. Shame I spent (what seems like when compared to 1% prices) a fortune on 0.1%ers.
cleaninglady
I have two questions...

The BOM says 1K Tempco , so i actually need 2K Tempco's ? Anyone used SMD ?

Also does the Living VCO Core BOM contain the Driver components too ?
nigel
cleaninglady wrote:
The BOM says 1K Tempco , so i actually need 2K Tempco's ?


hmmm..... No, you need 1k Tempcos. Like it says. SMD is fine.

cleaninglady wrote:
Also does the Living VCO Core BOM contain the Driver components too ?


There's a separate BOM for the driver board.
elmegil
cleaninglady wrote:
I have two questions...

The BOM says 1K Tempco , so i actually need 2K Tempco's ? Anyone used SMD ?

Also does the Living VCO Core BOM contain the Driver components too ?


I apologize, I think I confused things when I said:

Quote:
I've accidentally subbed 1K for 2K tempcos (or vice versa, I forget) and it totally blows the tracking. You can calibrate for 2V/Oct or 1/2V/Oct but never will reach 1V/Oct smile


I meant that I had done it in a different circuit entirely, I have not yet built my Living VCO boards. Bottom line: use what the BOM says, in this case, 1K.
FingerTappin
cleaninglady

Both BOMs can be found on page 2 of this thread, about half way down.
qfactor
elmegil wrote:


Quote:
I've accidentally subbed 1K for 2K tempcos (or vice versa, I forget) and it totally blows the tracking. You can calibrate for 2V/Oct or 1/2V/Oct but never will reach 1V/Oct smile


I meant that I had done it in a different circuit entirely, I have not yet built my Living VCO boards. Bottom line: use what the BOM says, in this case, 1K.


Was just thinking, so if I can squeeze in two 2k tempcos in parallel in there, I would come up with a 1k equivalent tempco, eh?? seriously, i just don't get it
Though having two temperature dependent resistors going at different rates of resistance may throw the tracking in another direction too! eek!
Anyway, just a thought.... d'oh!
nigel
elmegil wrote:
I apologize, I think I confused things when I said:
Quote:
I've accidentally subbed 1K for 2K tempcos (or vice versa, I forget) and it totally blows the tracking.



Ah, right. It's probably worth noting that you can use two 2k Tempcos in parallel in place of a 1k, or two 1k Tempcos in series in place of a 2k. (If you have Tempcos to burn, for some reason.)
cleaninglady
thanks all found on page 2 , i'm a bit tired and jumped from page one to last page.

I've been putting an entire Serge panel together : Panel fabrication planning , BOMs ordering and im a bit zombie Dead Banana
qfactor
I'm trying to match Q1 and Q2 using the Simple Transistor Tester circuit I built off Ian Fritz's schematics (modified a little by Dave Jones).
However, on measuring the voltage, the readings doesn't seem to stop!! help
It would stay at a particular reading for a while and then start jumping again. All this after waiting for half an hour to test 2 BC560C (1 as reference trannie)

I've paired 2N3906/3904s before and never had this issue. Is the voltage fluctuations "normal"? hmmm.....

Worst case, would tracking be really bad if I don't pair Q1 and Q2? seriously, i just don't get it
Randy
The readings on the transistors change as their temperature changes. That's why it is recommended to leave any you are testing in the same room for some period of time before testing them. Since the testing sends some current through them, their temperature changes, which changes the readings.

http://www.musicfromouterspace.com/analogsynth_new/TRANSISTORMATCHER/T RANSISTORMATCHER.html

Randy
elmegil
some folks have talked of putting them under some kind of cover to keep stray air currents from messing with them.
qfactor
From my measurements (out of 50 BC560Cs) there were 5 trannies that measured from 2.8 to 3.4mV, respectively.
Would these be considered "matched" or should I be going to exactly the same measurements? hmmm.....
Out of the 50, I had 2 pairs that matched exact mV measurements, but from what I read in MFOS website, matching could be within 2mV of each trannie.

Oh, and do I also need to match the 550 trannies as well?? eek!
Randy
Where did you read the trannies needed to be matched at all? I must've missed that. At this stage, I'm not going to bother anyways, just curious.

Randy
qfactor
Randy wrote:
Where did you read the trannies needed to be matched at all? I must've missed that. At this stage, I'm not going to bother anyways, just curious.


I believe someone on this thread said something along the lines that it was "good to be able to match Q1 and Q2" woah
Somewhere.... eek!

I saw your message on BST that you're selling your unfinished kit off??!! Why are you giving up?? seriously, i just don't get it
Randy
I'm not having too much luck with some of these more complex builds but since I had no takers I've decided to try and finish this one. Finished the panel and the PCBs yesterday, need to figure out the wiring to the jacks today. Going to get one working at a time and then figure out the wiring for the driver, which so far I'm lost on.

I've probably been staring at the boards too long but where is the sine output on the Barcode PCBs?

Randy
livefreela
hang in there man! sine comes out of the driver which contains the tri to sine wave shaper.. each core sends a tri feed via whatever you're not using as an output jack to the tri-in on the driver board - ergo the 6x wave shaper trims - 2 for each core. the sine out is wired from the driver directly to the jacks.

Randy wrote:
I'm not having too much luck with some of these more complex builds but since I had no takers I've decided to try and finish this one. Finished the panel and the PCBs yesterday, need to figure out the wiring to the jacks today. Going to get one working at a time and then figure out the wiring for the driver, which so far I'm lost on.

I've probably been staring at the boards too long but where is the sine output on the Barcode PCBs?

Randy
Randy
Thanks! I'm going to leave the driver for after the core boards are working so I'll hook all of that together later. I'm still missing a couple of pots for the driver (that damned 1M log pot is one).

Randy
qfactor
Yeah, Randy. Was also going to ask you to hang in there as well! hihi

Let's all build this together and go along asking (in this thread) how things are supposed to be (wired up and such). I'm not sure too! Just that I've not gotten to that part yet!

Perhaps we should upload some images of our builds, now and then just to check with one another! Mr. Green

I tested positioning the pots and placing them on the panel. Does your PCB get in the way with Osc 2's tracking SPDT switch?
I've tested a 16mm pot and a 9mm pot. Both seems to protrude the PCB in such a way that it's tricky fit with that Tracking switch getting in the way!
And even if it does fit, I think the PCB will be touching that Osc 2 tracking switch.

I'm thinking of placing some insulating sheet between the switch and the PCB. Have you tried fitting that in? It's the Core next to the Driver PCB I'm talking about.
qfactor
One other question for the good folks in here:
Is LM3046 the same as CA3046? hmmm.....
I found some LM3046 at a local store today and was wondering if I should get some to test for this build. seriously, i just don't get it

Edit: just came back from the Original LVCO thread when JH had just introduced this....seems like he used LM3046 originally!
Randy
Hopefully I'll be testing the first PCB today. I made my own panel so there have been no issues with clearance anywhere. It's a rough panel but looks fine from a reasonable distance!




Rand
qfactor
Nice!! thumbs up
Though isn't the 10k trimmer backwards? Mine's in the opposite position with the 100k trimmer hmmm.....
Randy
The direction of the trimmer shouldn't matter. At least it never has in the past.

Randy
livefreela
q - pretty sure lm3046 will work; i think its just the ti variant of the intersil chip. i've used lm3080 in place of ca3080 if thats worth anything, but i didn't check the datasheets to see if there was anything mission critical between the manfacturers. i'm fairly certain nothing will explode unless there's a totally unruly swap of pinouts between the manufacturers thumbs up re the trimmer, the orientation shouldn't matter. i would however strongly suggest swapping them out for side adjust types. i've built two living vcos for euro and trimming's a breeze provided you can reach that infernal fucking screw.

also yes, re: the selector switch hitting the pot, if you use 9mms there will be problems. better use 16 on the drivers if you're mounting the pots to the board - the switch needs the extra clearance provided via the pot's greater projection (counterintuitive i know, smaller pot, less space - weird)

i won't have access to mine for photos for a few days but are there any questions in particular re: the wiring you were wondering about? I assume you're probably wondering about the driver stuff as the core outs are pretty self explanatory and we've already addressed the sine above.

each core has a single pad labeled drv or driver, or something of that ilk. similarly, there's multi pin pad array on the driver board for driver out. the driver board's outs go to the center lugs of your tracking select switches. the outside lug should go to driver in on the core. that's it for the driver functions - as they're all simply manipulations on the pitch cv itself they're carried to each core on a single wire, received on what's essentially just an additional v/o input. the other driver/core connection just the sine converter that we covered earlier...


i must state that all the above comes with the disclaimer that but for having built two this way that work as should, i have no bloody idea what i'm talking about from any sort of scientific point of view hihi

qfactor wrote:
Nice!! thumbs up
Though isn't the 10k trimmer backwards? Mine's in the opposite position with the 100k trimmer hmmm.....
Randy
livefreela wrote:
re the trimmer, the orientation shouldn't matter. i would however strongly suggest swapping them out for side adjust types. i've built two living vcos for euro and trimming's a breeze provided you can reach that infernal fucking screw.


That's why I'm going to get them working individually first. I don't have any side-adjust versions handy.

livefreela wrote:
each core has a single pad labeled drv or driver, or something of that ilk. similarly, there's multi pin pad array on the driver board for driver out. the driver board's outs go to the center lugs of your tracking select switches. the outside lugs should go to ground and driver in on the core. that's it for the driver functions - as they're all simply manipulations on the pitch cv itself they're carried to each core on a single wire, received on what's essentially just an additional v/o input. the other driver/core connection just the sine converter that we covered earlier...


That's almost simple enough for even me to understand, thanks!

livefreela wrote:
i must state that all the above comes with the disclaimer that but for having built two this way that work as should, i have no bloody idea what i'm talking about from any sort of scientific point of view.


You are way ahead of me so thanks again for the help.

Randy
qfactor
Thanks for your reply, livefreela applause
One other thing is, did you have to match your trannies? Read somewhere in this thread that Q1 and Q2 should be matched? Are there any other trannies that need to be matched?

Thanks, man!
Randy
Tested the first PCB and, for some surprising reason, it works! Tracking is not too bad without adjusting anything. I fooled with R8 and R9 and got it tracking within a few cents, good enough for me. Detune doesn't seem to do anything but I'll test that once I get another PCB going.

Which trimpot is for which adjustment? I need to adjust the square wave a bit and the BOM (Barcode v 1.3) I have does not mention what each trimpot actually adjusts. Found it, reproduced here just in case:

R8 - Frequency
R9 - 1v/Oct scale
R18 - High Freq tracking
R40 - Pulse Width

Also, and I think this has come up before, is there anything I can do to turn the Detune into a Fine Tune?

There is a post on page 5 mentioning that matching the trannies is not necessary. It mentions matching resistors but I didn't even do that. I also did not do the fix for PWM and FM that Benjamin suggested, because they seem to work fine as is.

This thread can be a bit confusing in places!

Randy
Randy
Not quite so lucky with number 2. The TRI and SAW outputs both put out a SAW wave. Any ideas where to look?

Thanks

Randy
qfactor
My story This is fun! , so far... MY ASS IS BLEEDING

Silesius
Hi, I've just ordered the pcb set without panel, because I already had three nice (and expensive confused ) multi turn pots, and I would like to make a nice front panel with them.
Now I'm planning the panel, but I can't find any pics of the boards and I'm not sure how to design it.
Can anybody post some pics of the pcbs or the back side of the finished module or whatever you've got? And how about the size of them? It would be very helpful....
Thanks!
Randy
Silesius, are you talking about a Euro panel? If you go back one page, there is a photo of my panel, with one board installed, should give you an idea of how the boards are mounted. This thread and the panel thread have other panel ideas as well.

The boards are not large but the key dimension, assuming you are mounting the boards in a Euro panel, is the pot to pot measurement.

If I might make a suggestion, it would be to wait on the panel design until you get the PCBs.

Randy
Silesius
Thanks Randy, I've missed that one! And yes, I think I'll wait for the pcb's... But I was drawing this morning and I wanted to know a little more
Randy
PCB number 3 works fine as well, got it to track well and a nice, hollow square wave. Went back to PCB 1 and got the tracking better so that's two working fine.

Argh! Took a closer look at #2 to follow the traces from the tri out and noticed I had not soldered one end of one of the resistors! Quick test seems to indicate that was the problem, just letting it warm up a bit before I calibrate it.

Randy
OrganFixer
livefreela wrote:


each core has a single pad labeled drv or driver, or something of that ilk. similarly, there's multi pin pad array on the driver board for driver out. the driver board's outs go to the center lugs of your tracking select switches. the outside lugs should go to ground and driver in on the core. that's it for the driver functions - as they're all simply manipulations on the pitch cv itself they're carried to each core on a single wire, received on what's essentially just an additional v/o input. the other driver/core connection just the sine converter that we covered earlier...






I'm reaching about the same point in my build as you guys. So I'll be in on any wiring conversations. I guess I'll start with this one. The way you described the track/free switch does not seem right to me. On the driver card the four "DRIVEOUT" pins are all connected. The way you describe it you would ground all "DRIVEOUT" outputs from the driver for all cores if you turned tracking off for one. I think the DRV from the core goes to the center lug. When tracking is turned on it connects to the "DRIVEOUT" connector on the driver. The FREE lug should not be connected to anything so that you don't ground out the v/oct for that core. It just disconnects from the driver for that oscillator. I haven't wired mine yet but I am going to try it this way. I think I'm still probably a week away from that.

I actually edited this for clarity but I'm not sure it helped.
Randy
Can someone tell me how to calibrate the sine waves from the driver?

Thanks

Randy
cleaninglady
Can anyone tell me if using the BC550B/BC560B variants will be a problem ?

I couldn't find BC550C/BC560C at Tayda.
Randy
cleaninglady wrote:
Can anyone tell me if using the BC550B/BC560B variants will be a problem ?

I couldn't find BC550C/BC560C at Tayda.


Yes it will be a problem. You must use the C versions. There is an explanation why in one of the threads, can't remember where 'though.

Randy
cleaninglady
Great , thanks Randy ! thumbs up

I ordered some of the others but there cheap enough so no worries.

Any line on who is stocking the 'C' variants , Element14/Farnell says there discontinued. Dead Banana
Randy
I ordered the B variety of the 550s and installed them, didn't read the original thread or the BOM closely enough. Had to desolder them and get the correct versions. Found mine on eBay.

OrganFixer, the Track/Free switch wiring is correct as described, I just tried it, and it works as expected. Wait, see below.

Randy
qfactor
cleaninglady wrote:
Great , thanks Randy ! thumbs up

I ordered some of the others but there cheap enough so no worries.

Any line on who is stocking the 'C' variants , Element14/Farnell says there discontinued. Dead Banana


As Randy mentioned, eBay was where I sourced my variant "C" for both BC550s and 560s too. I paid around US$2 for 20 BC550C including shipping, so prices are good there as well. thumbs up
Just shop around on eBay a little before buying as there are quite a few sellers there. Furthermore, at those prices, it's a small risk to pay should the trannies come out *duds* seriously, i just don't get it But mine weren't! This is fun!
nigel
I got my BC550C / BC560C from Small Bear. Still in stock, as far as I can tell. (Rather more than 10c each, but probably more reliable.)
qfactor
Randy wrote:
Can someone tell me how to calibrate the sine waves from the driver?

Thanks

Randy


Probably best to use a O'scope for that? Otherwise, I'd just do it "by ear" to get the purest sine tone I can imagine! eek!
Randy
I was more wondering why there are two trimpots for the sine. I usually do it by ear as well. It's an analogue synth, not a piece of lab equipment, if it's not exact, I think I can live with it.

I used a linear pot for the Portamento, instead of a log pot. What this seems to have caused, although I have no idea why, is for the octave up/down transitions back to no transposition to include portamento even with the pot at minimum. Kinda odd.

OrganFixer, now that I have all of the VCOs hooked up, it seems you are correct. Moving any one of the T/F switches causes all of the VCOs to switch. Right now I have only VCO1 hooked to power and with the audio going to a mixer, but if I switch VCO 2 or VCO 3 (even with no power going to them), VCO 1 changes from track to free or vice versa.

I checked OrganFixer's idea and it works correctly. So, I have two things to figure out. 1 is that odd portamento on the octave switch and the other is that the Vibrato Depth pot does nothing for the first half of its travel.

I tossed it in the rack temporarily anyways, just to fool with it for a little while before putting knobs on and making everything permanent. Wow, this is a very fat batch of VCOs! I almost hate to say this but it can sound kinda Minimoog-ish.

Randy
SMonk
That 2.7 nF Polystyrene cap, is that 23PS227 on mouser? Are there any alternatives somewhere that don't have a delivery time of several months and a minimum order of 100?
spacedog
http://www.musikding.de/Styroflex-capacitor-27nF

Guinness ftw!
qfactor
Randy wrote:
Can someone tell me how to calibrate the sine waves from the driver?

Thanks

Randy


Looking around, I found this that might be of help to you?
SMonk
spacedog wrote:
http://www.musikding.de/Styroflex-capacitor-27nF

Guinness ftw!


Thanks. I had actually searched at musikding, but obviously i was outsmarted by the alternative spelling of polystyrene...
sammy123
Randy, do you mean it glides up or down when you flip the switch? I think mine does that as well.

Also concerning your earlier comment about the detune, it is very subtle. I asked that same question when I built mine from the original Barcode run.
Randy
qfactor, thanks for the schematic. I wish it helped but I appreciate the effort. I just did it by ear.

sammy123, it only glides when I switch it from transposed to non-transposed (back to centre). Really odd. Mine is the Barcode run as well. "Tune" on the driver is also a bit more subtle than I figured it would be, transposes a tritone (aug 4th) or so. All of this might be expected behaviour, really don't know.

If there is anyone out there that has one of these VCOs working, could you do a video showing the behaviour of the Tune on the driver and switching octaves, or at least let us know if this thing is actually doing what it is supposed to be doing.

Working or not 'though, it sounds great!

Randy
Benjamin AM
Randy wrote:
I was more wondering why there are two trimpots for the sine. I usually do it by ear as well. It's an analogue synth, not a piece of lab equipment, if it's not exact, I think I can live with it.

One controls amplitude of the signal which rounds off the point of the triangle and the other is a voltage bias to balance the waveform. If the bias is not set correctly one peak of the wave may be rounded and the other not at all.
Quote:

I used a linear pot for the Portamento, instead of a log pot. What this seems to have caused, although I have no idea why, is for the octave up/down transitions back to no transposition to include portamento even with the pot at minimum. Kinda odd.

If this is so, then I would test the pot and the solder connections. The portamento should not work at all when in full CCW.

Quote:

I checked OrganFixer's idea and it works correctly. So, I have two things to figure out. 1 is that odd portamento on the octave switch and the other is that the Vibrato Depth pot does nothing for the first half of its travel.

I had the same problem with the Vibrato depth. I however did end up using a 13700 for the OTA but can't imagine this being the source of the problem. I sort of feel that the suggested 12V values are not right. I found a good balance with R158 being 200k and R160 being 390k. I also changed R150 back to 15k(didn't do much). I removed R159. I think that Haible might have put this there for a footswitch. Your guess is as good as mine. Also make sure that the vibrato depth input is shunted to ground via the normalizing jack as detailed in Haible's wiring diagram. I also found a good mix without normalizing the jack(banana plug option). Leave in R159 and both R158 and R160 are 200k. I settled for the first option.
Randy
Thanks for the reply Benjamin. I set the sine by ear, sounds good enough for me. I don't have a 'scope handy so my ear will need to do for now. The odd portamento is only present when switching octaves, and only from +1 to normal or -1 to normal. I was thinking about swapping some resistors to play with the vibrato depth issue but since it works fine, I'll probably leave it alone for now and continue using the VCO, just in case there are any other issues that pop up. So far 'though, it's been fine.

Randy
Benjamin AM
Randy wrote:
Thanks for the reply Benjamin. I set the sine by ear, sounds good enough for me. I don't have a 'scope handy so my ear will need to do for now. The odd portamento is only present when switching octaves, and only from +1 to normal or -1 to normal. I was thinking about swapping some resistors to play with the vibrato depth issue but since it works fine, I'll probably leave it alone for now and continue using the VCO, just in case there are any other issues that pop up. So far 'though, it's been fine.

Randy

Your correct! I've never noticed this behavior. It glides only when returning to the center position. Funny!
ZibraZibraZibra
WOOOOO. that was a long one. started with 1 vco as a tester, then built the 3 big oscillator but 1 wouldn't work, for the life of me i couldn't source the problem, so had to build another VCO form scratch ill sort the other one out one day. but finally! all 3 working and in there with the driver.

also managed to add a mixer to the module because its always great not having to patch up them to an external mixer in my opinion. however skimpt on switches and decided just to offer square/saw/off because if they were all put into a mixer usually means its going to a filter... and triangles well you know..... i guess?

but overall frikking awesome!!!

except for the bourns 10 turn potentiometers sometimes skipping annoyingly its the bomb.

THANKS

Randy
Cool panel!

Randy
livefreela
sorry for the delayed reply on this - my transistors were hood-matched wink aka no test circuit of that ilk, just found ones that matched the readout via the diode tester on my meter - the same thing i do with trannies in my filters. the tracking on mine is fine for me - they're hardly motm300s but that isn't really what this circuit is about. still, i would imagine if approached via the fritz method or something similar tracking could be improved upon. i'm so stoked to hear you guys are making some headway on this, congrats guys!

qfactor wrote:
Thanks for your reply, livefreela applause
One other thing is, did you have to match your trannies? Read somewhere in this thread that Q1 and Q2 should be matched? Are there any other trannies that need to be matched?

Thanks, man!
ZibraZibraZibra
Randy wrote:
Cool panel!

Randy


Thanks! Black spray paint. Silver sharpy and lacquer haha. The quickest way i think!

All of the pcb's are parallel to the front. It looks Pritty neat! And easy to tune but I guess more space consuming! But it's not eurorack so I don't mind razz
livefreela
true - don't know where i got that from! don't ground the switch lug wink (op edited)

OrganFixer wrote:
livefreela wrote:


each core has a single pad labeled drv or driver, or something of that ilk. similarly, there's multi pin pad array on the driver board for driver out. the driver board's outs go to the center lugs of your tracking select switches. the outside lugs should go to ground and driver in on the core. that's it for the driver functions - as they're all simply manipulations on the pitch cv itself they're carried to each core on a single wire, received on what's essentially just an additional v/o input. the other driver/core connection just the sine converter that we covered earlier...






I'm reaching about the same point in my build as you guys. So I'll be in on any wiring conversations. I guess I'll start with this one. The way you described the track/free switch does not seem right to me. On the driver card the four "DRIVEOUT" pins are all connected. The way you describe it you would ground all "DRIVEOUT" outputs from the driver for all cores if you turned tracking off for one. I think the DRV from the core goes to the center lug. When tracking is turned on it connects to the "DRIVEOUT" connector on the driver. The FREE lug should not be connected to anything so that you don't ground out the v/oct for that core. It just disconnects from the driver for that oscillator. I haven't wired mine yet but I am going to try it this way. I think I'm still probably a week away from that.

I actually edited this for clarity but I'm not sure it helped.
Randy
Here's how it turned out:



Randy
qfactor
Nii-iice!! See? Weren't you glad you didn't sell it?!
applause
Randy
Yup!

Randy
qfactor
OrganFixer wrote:
livefreela wrote:


each core has a single pad labeled drv or driver, or something of that ilk. similarly, there's multi pin pad array on the driver board for driver out. the driver board's outs go to the center lugs of your tracking select switches. the outside lugs should go to ground and driver in on the core. that's it for the driver functions - as they're all simply manipulations on the pitch cv itself they're carried to each core on a single wire, received on what's essentially just an additional v/o input. the other driver/core connection just the sine converter that we covered earlier...




I'm reaching about the same point in my build as you guys. So I'll be in on any wiring conversations. I guess I'll start with this one. The way you described the track/free switch does not seem right to me. On the driver card the four "DRIVEOUT" pins are all connected. The way you describe it you would ground all "DRIVEOUT" outputs from the driver for all cores if you turned tracking off for one. I think the DRV from the core goes to the center lug. When tracking is turned on it connects to the "DRIVEOUT" connector on the driver. The FREE lug should not be connected to anything so that you don't ground out the v/oct for that core. It just disconnects from the driver for that oscillator. I haven't wired mine yet but I am going to try it this way. I think I'm still probably a week away from that.

I actually edited this for clarity but I'm not sure it helped.


Hey, so I'm at wiring stage now and need a bit of help with some connectors.
On the panel, Driver Core side, there are two V/oct jacks. Are these connected in series with one another? I see a "v/o" on the Driver PCB as well. So how do I connect these to the 2 jacks? hmmm.....
ZibraZibraZibra
[quote="qfactor"][quote="OrganFixer"]
livefreela wrote:


Hey, so I'm at wiring stage now and need a bit of help with some connectors.
On the panel, Driver Core side, there are two V/oct jacks. Are these connected in series with one another? I see a "v/o" on the Driver PCB as well. So how do I connect these to the 2 jacks? hmmm.....



that one got me too. i thought it was V/o and a ground (the square pad) turns out the square pad is just another V/O............ so they do go in seperately. just the connection is quite misleading eek!
ZibraZibraZibra
aaah got the drill out again... well for one i forgot to put jacks in for the sine outs. didnt read about them! i wondered why there were so many variable pots on the drive board. pritty cool about overdriving a triangle wave into a sine pritty cool.... never knew it did that.


however heres another thing ive come to. its annoying there is no coupled PWM jack. and looking at the original living vco it has! im just going to do a simple tl072 mixer i guess and use that to sum the pwm's together.... anyone got a better idea? i would love it to be passive but how would i do it?
qfactor
[quote="ZibraZibraZibra"][quote="qfactor"]
OrganFixer wrote:
livefreela wrote:


Hey, so I'm at wiring stage now and need a bit of help with some connectors.
On the panel, Driver Core side, there are two V/oct jacks. Are these connected in series with one another? I see a "v/o" on the Driver PCB as well. So how do I connect these to the 2 jacks? hmmm.....



that one got me too. i thought it was V/o and a ground (the square pad) turns out the square pad is just another V/O............ so they do go in seperately. just the connection is quite misleading eek!


So just to be sure, you mean each v/o pad goes to one v/o jack,eh?
thumbs up
qfactor
ZibraZibraZibra wrote:
aaah got the drill out again... well for one i forgot to put jacks in for the sine outs. didnt read about them! i wondered why there were so many variable pots on the drive board. pritty cool about overdriving a triangle wave into a sine pritty cool.... never knew it did that.


however heres another thing ive come to. its annoying there is no coupled PWM jack. and looking at the original living vco it has! im just going to do a simple tl072 mixer i guess and use that to sum the pwm's together.... anyone got a better idea? i would love it to be passive but how would i do it?


Yes indeed! Didn't realize this till you mentioned it too! eek!
Would be super to have a summed Pwm to the 3 Oscs.
Randy
I'm thinking the other way with the PWM. Although a bit resource intensive, having a different PWM source for each VCO certainly sounds cool.

Randy
ZibraZibraZibra
[quote="qfactor"][quote="ZibraZibraZibra"]
qfactor wrote:
OrganFixer wrote:
livefreela wrote:


Hey, so I'm at wiring stage now and need a bit of help with some connectors.
On the panel, Driver Core side, there are two V/oct jacks. Are these connected in series with one another? I see a "v/o" on the Driver PCB as well. So how do I connect these to the 2 jacks? hmmm.....



that one got me too. i thought it was V/o and a ground (the square pad) turns out the square pad is just another V/O............ so they do go in seperately. just the connection is quite misleading eek!


So just to be sure, you mean each v/o pad goes to one v/o jack,eh?
thumbs up


yep
ZibraZibraZibra
[quote="qfactor"][quote="ZibraZibraZibra"]
qfactor wrote:
OrganFixer wrote:
livefreela wrote:


Hey, so I'm at wiring stage now and need a bit of help with some connectors.
On the panel, Driver Core side, there are two V/oct jacks. Are these connected in series with one another? I see a "v/o" on the Driver PCB as well. So how do I connect these to the 2 jacks? hmmm.....



that one got me too. i thought it was V/o and a ground (the square pad) turns out the square pad is just another V/O............ so they do go in seperately. just the connection is quite misleading eek!


So just to be sure, you mean each v/o pad goes to one v/o jack,eh?
thumbs up




yep
ZibraZibraZibra
[quote="qfactor"][quote="ZibraZibraZibra"]
qfactor wrote:
OrganFixer wrote:
livefreela wrote:


Hey, so I'm at wiring stage now and need a bit of help with some connectors.
On the panel, Driver Core side, there are two V/oct jacks. Are these connected in series with one another? I see a "v/o" on the Driver PCB as well. So how do I connect these to the 2 jacks? hmmm.....



that one got me too. i thought it was V/o and a ground (the square pad) turns out the square pad is just another V/O............ so they do go in seperately. just the connection is quite misleading eek!


So just to be sure, you mean each v/o pad goes to one v/o jack,eh?
thumbs up


yep
ZibraZibraZibra
[quote="qfactor"][quote="ZibraZibraZibra"]
qfactor wrote:


So just to be sure, you mean each v/o pad goes to one v/o jack,eh?
thumbs up


yep
ZibraZibraZibra
@qfactor yeahhh each pad to both vo inputs!
ZibraZibraZibra
aaah shit how do you delete duplicates?

soz
qfactor
As I'm slow and steadily building this up, bit by bit, I just noticed for the Osc Cores, there are duplicate pads for each waveform.

Now I know that I'll need 2 of the Triangle out pads, one to go directly out to the Tri jack while the other go to the Driver board "Tri In" to get converted to Sine Out, but what are the duplicate pads for the other waveforms for?? hmmm.....

Oh and not forgetting duplicate V/O pads too on the Osc Core? seriously, i just don't get it
qfactor
Benjamin AM wrote:
I solder a green wire at the middle leg tap and jump it over to the resistor leg that has been lifted off the board(R2-FM R38-PWM).


Benjamin AM: just it be doubly sure I get the right end of R2 and R38, I'd like to get your help help
From your image, I think you lifted the R2 end that is on the side of pins 5-8 of the TL072, right?
How about R38? hmmm.....
Randy
It is interesting that Benjamin and Barcode mention various problems but I didn't notice anything in my build, it works fine. The PWM pot and FM pot seem to work as expected so I really don't know what to think about these recommendations for mods.

I realize it's a hassle to go back and fix something after you've built it, but qfactor you might want to consider building it as is before doing a bunch of mods.

Here's my Living VCO, with PWM:

http://soundcloud.com/piscione/moog-filter-demo

Randy
logicgate
qfactor wrote:
As I'm slow and steadily building this up, bit by bit, I just noticed for the Osc Cores, there are duplicate pads for each waveform.

Now I know that I'll need 2 of the Triangle out pads, one to go directly out to the Tri jack while the other go to the Driver board "Tri In" to get converted to Sine Out, but what are the duplicate pads for the other waveforms for?? hmmm.....

Oh and not forgetting duplicate V/O pads too on the Osc Core? seriously, i just don't get it


They are the same thing... Only that Barcode arranged them side by side, for the sake of organization...

The "original" pads (scattered through the PCB) are the ones very close to the outputs of the ICs, just because I´ve learned that the signal is less prone to interference/noise if the pads are placed this way... Well, at least it´s what I´ve been told, I´m no EE.
qfactor
Randy wrote:
I realize it's a hassle to go back and fix something after you've built it, but qfactor you might want to consider building it as is before doing a bunch of mods.


According to them, the mod makes the pots into attenuators? Is your currently attenuating?? oops

But you're right, I should just do it "as is" first and see how it goes without over thinking things! thumbs up

Will keep y'all in the loop the next few days when my 1st (and maybe 2nd) core would be ready! nanners
nigel
Randy wrote:
The PWM pot and FM pot seem to work as expected so I really don't know what to think about these recommendations for mods.


They will "work", in that the signal gets larger and smaller as you turn the knob. The problem is that when the pot is turned all the way down, the input signal is shorted to ground, which is generally Not A Good Thing.
Randy
qfactor, when I turn the PWM knob clockwise, I get more PWM. Same for the FM.

nigel, not a good thing in what way? I haven't noticed any ill effects so I'm wondering what the issue really is? Can I damage something eventually?

Randy
nigel
Randy wrote:
nigel, not a good thing in what way? I haven't noticed any ill effects so I'm wondering what the issue really is? Can I damage something eventually?


Most module outputs have a series resistor, so they shouldn't be damaged. But for example, if you're using a passive mult to connect a signal to this input and other inputs at the same time, those other inputs will receive no signal when this module shorts it out.
Randy
nigel, you are correct, I gave it a try just to see what happens. Since I don't really understand Benjamin's fix, perhaps someone can do some sort of diagram for this, or I'll just put up with it.

Randy
nigel
I've just finished fixing my boards, managed to get quite a clean result. I'll try to post some photos when I get home.
qfactor
nigel wrote:
I've just finished fixing my boards, managed to get quite a clean result. I'll try to post some photos when I get home.


That would be great! thumbs up Do remember to make the images clear as to which end of the resistors are being lifted too? d'oh!

Thanks!
nigel
Disclaimer up front - I haven't finished this build yet, so this hasn't been tested. However, I'm reasonably sure that it is correct. If anyone sees a problem, please let me know.

I decided to cut the PCB tracks rather than lifting resistors, using a Dremel with a little ball-ended cutting bit (I think it was the 107 Engraving Cutter). I found it easier to mount the Dremel and move the board up to it, rather than the other way around. You could probably use a sharp knife, but the Dremel and a steady hand made the job quick and easy.

We need to rewire the two centre pins and the two top pins from the FM and PWM pots. First we need to get the signal from the pots. The centre pins both run to a marked connection point immediately above them, so we'll use that. For the end pins, I'm using 16mm pots, so I used the 9mm pot connections to the same pin.



Now, cutting the tracks. The centre connection to the PWM pot runs along the front edge of the board, on the top side. Cut it next to the pot, and scrape clean a short section of the track on the far side of the cut. The top connection from that pot runs on the back of the board. Again, cut it next to the pot and clean the far side of the cut. The centre connection to the FM pot is on the back of the board - for this one, we need to cut on the far side of the connection marked FM. (The marking is on the front, we need to cut on the back.) The top connection is on the front, and is cut next to the pot like the PWM. For both, scrape clean a short section of track on the far side of the cut.



Now grab a couple of bits of wire. Connect the centre connection from each pot, to the far side of the track which used to go to the top connection. Connect the top connection of each pot to the far side of the old centre connection. You may need to curve some of the wires a little to find a clear path.

And you're done. Note that the wire to the 9mm holes will be sitting very close to the pot terminals when you fit them. You may need to trim a bit of metal from the centre terminal of the pot to ensure that it doesn't short out.
Randy
Thanks nigel. I may try to figure out what Benjamin did only because I don't want to desolder the pots. Desoldering 6 pots would not be fun.

Randy
qfactor
So would that be a total four cuts on the PCB?
nigel
Randy wrote:
Thanks nigel. I may try to figure out what Benjamin did only because I don't want to desolder the pots. Desoldering 6 pots would not be fun.

Actually, I had already soldered the pots on one of my boards. You just need to shift the cuts on the front of the board a little further along the tracks. The trace that runs along the front of the board can be cut (carefully!) in the gap between the pots. I wired it directly to the leg of the pot, since the extra hole is inaccessible.
qfactor wrote:
So would that be a total four cuts on the PCB?

Yes. Four cuts, four links.
qfactor
Thanks, nigel!
applause
Randy
I did the mod for FM and PWM inputs as Benjamin suggested, easy mod, although made a bit trickier because my Living VCO is already wired up. No traces to cut, works great. Here are two pics showing the two resistors and the wires going from the extra pot solder tab holes (eventually making their way to the relevant input jacks). The wire from the exposed end of R38 goes to the PWM hole and from the exposed end of R2 goes to the FM hole (under the red wire, sorry). Some heatshrink tubing went over the connections on the resistors.




Randy
qfactor
Alright! This was super clear showing which end of the resistors to lift! thumbs up
Did you use the 10-turn 10k pot for the Freq as well?
Randy
Nope, no 10-turn pot. I haven't had any trouble tuning it with a normal pot.

I have done the mod on all three VCOs and it works fine. While I was in there, I also changed some stuff on the driver. I replaced the linear 1M portamento pot for an audio version, much better, and I tried another of Benjamin's recommendations, replacing R158 with a 200k and R160 with 390K to get more use out of the vibrato depth. That mod did not work, all I got was constant vibrato, so I replaced R160 with a 200K and it's much better.

That leaves me with one last bit of oddness, that I'll just put up with for now, and that is the portamento that happens when returning the octave switch to the centre position, even with portamento off.

Randy
qfactor
Randy wrote:


That leaves me with one last bit of oddness, that I'll just put up with for now, and that is the portamento that happens when returning the octave switch to the centre position, even with portamento off.

Randy


Do you mean after returning the octave switch to center, the portamento remains ON for all the notes you play even though the knob is at "0"? hmmm.....
Randy
Nope. If I hold a note and switch from normal to +1 (for example), it sounds fine. If I hold a note and switch from +1 to normal, the note glides down to normal. After that the rest of the notes are fine. Same for -1.

Randy
jules
Thanks for the FM&PW mod.
I was wondering why my FM signal was disappearing when mult-ed.
Edit: Wrong.
Still I would consider that a critical thing to do.

Question:
Does someone knows the way to tame the PWmodulation?
If I barely move it 9am with and LFO 10vpp plugged in, I go from 0% PW to 100%PW (and I loose sound). Or should I use 10k pot instead of 100k?
Note that I do not loose sound when turning the PW pot...
Edit2: that was due to the lifting of the resistors.

Then the vibrato: I get it when the vibe depth pot is past 12:00.
between 8am and 12:00, no effect.
Randy
Jules, I'm not sure about the PWM because mine didn't have that issue. I thought I read something about it in this thread 'though so you may need to read through the entire thing.

For the Vibe depth, I took a hint from Benjamin, with a small change, and changed R158 and R160 to 200k (on the driver PCB). Benjamin had suggested 200K for R158 and 390K for R160 but when I did that, I got a constant vibrato.

Randy
Benjamin AM
Randy wrote:
Jules, I'm not sure about the PWM because mine didn't have that issue. I thought I read something about it in this thread 'though so you may need to read through the entire thing.

For the Vibe depth, I took a hint from Benjamin, with a small change, and changed R158 and R160 to 200k (on the driver PCB). Benjamin had suggested 200K for R158 and 390K for R160 but when I did that, I got a constant vibrato.

Randy

Hold on, I make two suggestions in my previous post. One for the input normalled to ground( r158 is 200k and r160 is 390k). For non-normaled inputs, 200k for both resistors(which you did). Curious if your input is normaled to ground. If it is not, this would explain why to had constant vibrato with r160 being 390k.
Randy
Benjamin, I was hoping you would chime in! I did not normal the input to ground, so I'm glad you were able to explain that, thanks! The two 200K resistors seem to work quite well 'though, at least good enough for me.

Randy
qfactor
Benjamin, when you say the input normalled to ground, why does only the Vib Depth input have to be normalled? hmmm.....

Randy, I did the fix using the "lifted resistor" version that Benjamin had mentioned as well! thumbs up Will be testing the 1st Core with the Driver, later! thumbs up
Randy
qfactor, I wired, tested and calibrated each core one at a time. When I had all three working fine, I added the driver last.

Randy
Benjamin AM
qfactor wrote:
Benjamin, when you say the input normalled to ground, why does only the Vib Depth input have to be normalled? hmmm.....

Well, short answer. I wouldn't normal the vibrato input to ground.
Real answer: I did normal the vibrato jack because Haible does in his schematics. I was curious why he did and I have come to the conclusion that he did this for the option of being able to use a foot pedal in the vibrato input jack. Notice that this is just a theory that I came up with while looking at the schematics. R159 keeps the sum of vibrato pot and input jack at a high voltage which turns on the OTA. With the jack normalled to ground the R159 has virually no effect on the sum. If a N.C. footswitch is plugged in, the input connection is grounded until the pedal is step on. Then when the ground is lifted(pedal stepped on), there is a floating input and R159 does its thing(turns on OTA=adds vibrato). Make sense? One may also plug in a voltage source as usual. This gives the input a dual function. Maybe someone else can shine some light on the subject, verify/tear down theory. I don't believe this function is mentioned on his web page but maybe it is mentioned somewhere deep n the E-M thread.
Personally, I cut out R159 and kept to value in that worked with the normalling for no real good reason but just out of pure laziness. I would suggest the 200k for both resistors discussed earlier without normalling.
qfactor
Randy wrote:
qfactor, I wired, tested and calibrated each core one at a time. When I had all three working fine, I added the driver last.

Randy


I understand. Just that I had done up the Driver board 1st as there was only 1 Driver. d'oh!
In any case, tried it out and happy to say that the Core and the Driver seems to work alright based on what it can do with only 1 Core, ATM.
Still need to calibrate the tracking as it does go a little off over the octaves. Which trimmers work with which frequency to calibrate again? hmmm.....
I used a multi-turn 10k pot and works perfectly!! Bought this from some eBay guy in China, so was taking a risk as to the quality, but it worked! nanners

As for the Portamento, I too, encounter that same issue you mentioned when switching the octaves. It slides each time the octave switch is changed.
Logically speaking, a glide would mean that each time the switch is changed, a voltage slowly changed to the original, that is connected to the switch. Is there any cap around that area, doing this? hmmm.....

Do note that I'm no electronics expert and only DIY based on "color by number" mindset! zombie
But that's my logical sense.... eek!
the bad producer
I'd never noticed that (slide when switching octave switch) but it's present on the original too - just thought I'd mention that, that it is not a feature just on this board set, but also on the original JH PCB's...

Randy, I've just PM'd you!
gbiz
qfactor wrote:

As for the Portamento, I too, encounter that same issue you mentioned when switching the octaves. It slides each time the octave switch is changed.
Logically speaking, a glide would mean that each time the switch is changed, a voltage slowly changed to the original, that is connected to the switch. Is there any cap around that area, doing this? hmmm.....


Yes. My guess is it's C28. It's easy enough to test, lift one leg.

I can't tell this on mine here yet, i'm still battling with getting all the other rework finished.
jules
After a few painful hours of swapping resistors, that's the half working combination I got:
R33: 22k (tried 15,18, 20, 24 and 27k)
R36:800k (500, 750 and 1M)
R39: 1M (500k to 2M)
R38: 400k (1M to 270K)

That's a lot of combinations, as you can imagine.
It didn't solve the uselessness of the PWmod potentiometer.
I go from no modulation to full modulation in less than 1/8th of a turn.
The way to go is to lower R33, but then the modulation doesn't swing very far. I've also check what's around, and I can't find anything wrong.

I'm done wasting time with all of this.
I don't know what the PW modulation sounds on a CS80 sounds like -and now I've got no desire to- but if I want a triangle to go trapezoid as a modulation source, I'll overdrive it upstream the LVCO.
gbiz
jules I feel your pain. I'm getting seriously fed up with all the rework required to get these finished to an acceptable level Dead Banana

I've got the PWM & PWMmod working reasonably well now on one of the two LVCOs i'm building (so thats three core VCOs). The PWM pot swings from 10% to 90% pulse width across it's travel on all three. The PWMmod C/V modulates to the same widths.

This is with the resistor values in the original JH schematic. I started with resistor values from the published BOM & was seeing similar behaviour to you.

So my working resistor values are:
R33 20K
R36 750K
R39 1M

I tried 12K in R34, but it doesn't make a huge amount of difference compared to the 10K from the published BOM, so i'm leaving that as is. One less resistor to unsolder on these fragile PCBs is a good thing.

I matched Q4 & Q5 from the start though. If you've not done that, it's probably worth giving it a try, at least on one board
gbiz
I just replaced C28 on the driver board with a 100pF MLCC & that's cured the slew when switching octaves. (OK, technically it's still slewing but 1000 times faster now).


For anyone who's not had enough of reworking these boards, i can recommend swapping the power pins on the driver board header so you can use a dasiychained ribbon cable across all the boards without the need to twist the driver board. I cut the tracks from the header pins to the two beads, replacing each track with a wire from the opposite header pins. Obviously if you're using a shrouded header you'll need to spin the shroud through 180 degrees to take care of the keying.
qfactor
gbiz wrote:
I just replaced C28 on the driver board with a 100pF MLCC & that's cured the slew when switching octaves. (OK, technically it's still slewing but 1000 times faster now).


For anyone who's not had enough of reworking these boards, i can recommend swapping the power pins on the driver board header so you can use a dasiychained ribbon cable across all the boards without the need to twist the driver board. I cut the tracks from the header pins to the two beads, replacing each track with a wire from the opposite header pins. Obviously if you're using a shrouded header you'll need to spin the shroud through 180 degrees to take care of the keying.


Thanks for the tips, gbiz, jules
If not anything else, your effort and time testing all of these and sharing them here so that others can learn from it makes all your trouble, not in vain!! thumbs up applause applause ..and most appreciated too!!

Fortunately, I've been following this thread over the last few months(!!) and also used these resistor values you mentioned as was mentioned/suggested by Benjamin and Randy too. So my resistor values for those you mentioned are similar as well AND they seem to work fine! thumbs up

Will try out the C28 change to 100pF as well. As for daisychaining the Core power, another way could be using these shrouded right angle headers?
Unfortunately, I found these only AFTER soldering non shrouded, vertical headers to all 3 Osc cores a week before, so I don't think I'm in the mood to desolder all the 3 cores now! d'oh!
Oh well, as I said, for others to benefit...

qfactor
Though I've got another set of LVCOs that's still not started that would look like this with the right angled headers...
The triangle marking that is where the -ve should be makes it necessary to mount this at the bottom of the PCB! sad banana

leterell
hey, could the ones with more wisdom tell me if these transistors are the C-type (the more favourable for this particular design)?

mouser part neek!'s

512-BC550CTA

512-BC560CTA


i tried reading up, yeah, also in the datasheets, but to no end. there are a couple of C's in the part number...? d'oh! i guess my theoretical knowledge of trannies is insufficient. probably for the best for me tho Dead Banana
qfactor
Er....does someone have an image of the unstuffed Driver PCB? hmmm.....
Can't figure out where are R158 and R160 on the Driver Core now that I've soldered everything on both my sets!! d'oh!
gbiz
leterell

Yes those are the ones. It's best to order extras as you'll need to match them.

If you're only just getting around to ordering parts, & you've not already done it, i'd suggest you look at getting correct dimensioned resistors so you don't end up with resistors that don't fit the holes, & stacked on top of each other. Make sure they're 3.5mm length, 2mm dia. I used TE Connectivity LR0204F from Farnell but it looks like Mouser sell these too.
gbiz
qfactor

Even with those you'll still need to twist the power cable at the driver board.

With a daisychained cable that doesn't need a twist in it, you only need a keyed header on one board to ensure the cable is correctly orientated.

I hate to admit it, but as luck would have it I used a unbranded Tayda shrouded header. I lifted the plastic shroud off of the pins whilst they were still soldered in the board, turned it through 180˚ & pushed it back on the pins. No soldering required. I've done this before with unbranded headers, but never managed it with a branded one.
the bad producer
Yes, that's what I use!

Attached scan of driver PCB too:




leterell wrote:
hey, could the ones with more wisdom tell me if these transistors are the C-type (the more favourable for this particular design)?

mouser part neek!'s

512-BC550CTA

512-BC560CTA


i tried reading up, yeah, also in the datasheets, but to no end. there are a couple of C's in the part number...? d'oh! i guess my theoretical knowledge of trannies is insufficient. probably for the best for me tho Dead Banana
gbiz
I've now reworked all 6 of my VCO cores with values for R33/R36/R39 that match the original JH schematic (20K, 750K, 1M). All 6 have now consistent 10% through 90% swing for pulse width across the range of the PW pot. This is with matched transistors for Q4 & Q5.

The PWM CV input hits it's full range before the end of the travel of the PWM pot, but i can live with that.
qfactor
the bad producer wrote:
Yes, that's what I use!
Attached scan of driver PCB too:


Thank you so much!! I've got to remember to take a picture of all my unpopulated pcbs for future DIY builds!! d'oh!
Sometimes the excitement of getting to build them makes me forget all about the possible modifications/fixes that come with DIYs!
Thanks! thumbs up thumbs up
jules
gbiz wrote:

The PWM CV input hits it's full range before the end of the travel of the PWM pot, but i can live with that.


Can I ask at what point of the travel does it hit full value?
I was borderline thinking about trying a 10k there.
Randy
My PWM CV pot is about the same, no change past 4:00. I'm also willing to live with it.

I replaced C28 on the driver board with a 100pF as suggested above and it cures the odd glide habit. Now my tracking when switching to +1 is screwed up, gotta go and find the driver calibration again. The individual VCOs track fine 'though.

I don't think I have ever pulled a module out and put it back in again so many times!

Hey Bad Producer, replied to your PM.

Randy
qfactor
Hey Randy,
Where'd you get the Driver calibration routine from? Only saw an Osc calibration on the JH website. hmmm.....
qfactor
Duplicate
Randy
I may have imagined it! I thought I remembered reading something about the OCTTUNE01 and OCTTUNE02 trimmers but it could've been in a dream.

Randy
gbiz
jules wrote:
gbiz wrote:

The PWM CV input hits it's full range before the end of the travel of the PWM pot, but i can live with that.


Can I ask at what point of the travel does it hit full value?
I was borderline thinking about trying a 10k there.


With a +/- 5V modulating CV, it's about 12 o'clock (half way through the travel), which seems to be as designed. Past that, it's into the overmodulation discussed in the other LVCO thread (here) & demonstrated in kvitekp's video linked in that thread. That video pretty much shows what i get here.

A 50K pot might be a better bet, that way you don't lose some of the overmodulation.
Randy
I think I'm done with the Living VCO, finally. Everything appears to work, and a bit of fooling with the OCTTUNE01 and OCTTUNE02 trims on the driver returned the octave switching back to normal.

Even if I find something else wrong, I'm not touching it for a few weeks.

Randy
Benjamin AM
Congrats! I'm not sure why the slew on octave shifting is bothersome to people. I personally enjoy the quirk. To each their own.
qfactor
So, like, what's the best way to calibrate and tune these Oscillators? d'oh!
Spent the whole afternoon trying to get one of the Oscs to get the octaves right and still trying! very frustrating
I used mainly the V/O trimmer and the Freq trimmer on the Osc core to do the tuning. What I'm getting is if I hit,say, a middle C, the octave below sounds right but the octave above that is hitting C#.
I'm doing this by ear as I don't have a Freq meter nor tuner around.
Any tips you guys use to tune and calibrate?
The other Osc seems to be fine though, so it's basically that the octaves aren't! seriously, i just don't get it
Randy
Benjamin, it didn't bug me that much either but since gbiz found an easy fix I figured I might as well put it in.

qfactor, are the octaves screwy when you switch octaves with the switch on the driver? If that's the case, you need to use the OCTTUNE01 and OCTTUNE02 trimmers on the driver. One is for the upper octave and the other is for the lower, I can't remember which but you'll hear it. It is very subtle so either use another synth for reference or keep switching back to centre for reference. I have a tuner for my iPad that I usually use, but I did the octave switch tuning by ear this time.

Get the VCO cores calibrated first so they are tracking properly in "free" mode. When they are fine, don't touch them again. Use the trimmers on the driver for the octave switch.

Randy
qfactor
Randy wrote:
Benjamin, it didn't bug me that much either but since gbiz found an easy fix I figured I might as well put it in.

qfactor, are the octaves screwy when you switch octaves with the switch on the driver? If that's the case, you need to use the OCTTUNE01 and OCTTUNE02 trimmers on the driver. One is for the upper octave and the other is for the lower, I can't remember which but you'll hear it. It is very subtle so either use another synth for reference or keep switching back to centre for reference. I have a tuner for my iPad that I usually use, but I did the octave switch tuning by ear this time.

Get the VCO cores calibrated first so they are tracking properly in "free" mode. When they are fine, don't touch them again. Use the trimmers on the driver for the octave switch.

Randy


Thanks, Randy.
Not sure how you mean by "screwy octave" while switching the Oct switch but mine sounds pretty much the same whether I physically play the octave higher or use the octave switch to switch an octave higher.

Tried the Octtune 01 and 02 and you are right in that I don't hear much of a difference. But maybe because I wasn't switching the Oct switch back and forth while adjusting. Besides the pitch being sharper on the octave up, the switch was pretty consistent (consistently SHARP, that is!!) hihi

So you tuned the Osc while in "free" mode? Did you send a known voltage in "free" mode, check the pitch, then send another voltage twice the original and tune for the octave? How would the V/O jack work in "free" mode then?? hmmm.....
setsun8
Following this thread makes me not want to start my LVCO...it seems like quite a pain to get it working right. Please tell me it was all worth it in the end!?! Was it?! seriously, i just don't get it
mush
I found it to be an easy build. can't really understand all the struggle posts... hmmm.....
Randy
mush, you were really lucky, you haven't used it much or you are an exceptional builder, there have been all sorts of issues, some common, some individual.

setsun8, just read through the thread carefully, note the changes that need to be made, make those changes while you are building instead of after, and you should be fine.

qfactor, I use a MIDI keyboard and qMI interface, and used the V/O in of each core to calibrate each individually. I suppose you don't really need to, you could use the Driver v/o in instead, no difference, but I would not use the driver octave switch to calibrate the tracking of the cores because the octave switch may need to be calibrated too.

Randy
Zaibach
setsun8 wrote:
Following this thread makes me not want to start my LVCO...it seems like quite a pain to get it working right. Please tell me it was all worth it in the end!?! Was it?! seriously, i just don't get it


Do it, build it, you won't regret it cool Mine worked right away (which was more a matter of coincidence). I still have the pwm issue which I want to fix at some point (and switch to 10-turn pots for frequency) with the info found in this thread. Nevertheless, everything else is working fine and there's no reason you shouldn't start building yours. I've had way more problems with other modules, so swapping a few resistors in this case is child's play.
setsun8
Yeah I have read through the thread but there is a lot of information and it is not all clear. I wish we could gather these things together to make it easier for people sifting through 20+pages. I just did another quick glance through the thread and here is what I noted. (People who have built this thing, please feel free to add/comment to this list so we can try to keep things together).

Recommended changes:
-Benjamin’s PWM resistor recommendations:
(can socket for easy testing)
R33=20k
R39=1M
R36=750k

-Benjamin's PW and FM pot swap fix here: https://www.muffwiggler.com/forum/viewtopic.php?t=117137&postdays=0&po storder=asc&start=125

Optional changes:
-10 turn freq pots
OrganFixer
Ok here are the fixes I have done or I am planning to do. Hopefully they apply to a +/- 15v build.

In no specific order

1) To fix the portamento glide when using the octave switch on the driver board, Replace C28 on the driver board with a 100pF capacitor.

2)Due to update BOM replace portamento pot with 1M Log pot. (driver board)

3) To fix the vibrato depth problem replace R158 and R160 with 200K resistors. If you have a normalled jack one resistor should be 390K but you will need to go back and find that post. I'm using banana jacks so there is no normalling in my build.

4) To fix the dropping out of PWM when the pot gets near the end of the rotation I have replaced R39 with a 1.5M resistor but I believe the preferred solution will be to use these values
R33=20k
R39=1M
R36=750k

5) raise the two resistors and rewire the PW and FM pots to be attenuators



Does that seem to cover everything. I have completed 1,2, and 3. I'm about to do 5 and 4 ( I think in that order.) I'm hoping this will eliminate the need to wire thing up, test, take apart, fix, repeat. I'm hoping I can put it all together and call it done. I'm building two of these so each fix will need to be done six times. I guess what I'm asking is if I can just do all these or should I do some test on each board and customize values etc?

Also, are there one or two L's in Normalled and Normalling? For some reason it looks wrong both ways to me.
qfactor
setsun8, OrganFixer

Yes just about cover all the mods and fixes that has been discussed so far in all these 20+ pages. I guess through our discussions we had to ultimately arrive at all of these (some 20 pages later seriously, i just don't get it) and in the spirit of sdiy, nobody else (but this thread) would have been able to "just provide all the details" until it was discussed, thrashed out, reasoned out, tried out and tested, before we collectively arrived at the conclusion (hopefully that's the last of it!!) on this page!

I've found this build, the most challenging to date over the last 3 years since I started DIY in MW. I also felt that going through this has provided me more "go-get" to even more difficult builds simply because I persevered and achieved building this! thumbs up

So, like what Zaibach said, setsun8, go for it!! nanners We've all managed to get this far, helping one another out! thumbs up

And OrganFixer, it does seem like that is spelt with a "double L" in Normalled!! hihi
setsun8
Thanks for the encouragement qfactor ! Yeah I definitely appreciate all the help people have provided with the hours that went into their builds. I'll be getting the components soon and will grab the Tayda resistors so that they fit the PCB holes. Thanks for the help everyone!
we're not worthy
jules
OrganFixer wrote:


5) raise the two resistors and rewire the PW and FM pots to be attenuators



Edit: I was wrong. But lifting the resistors didn't work for me.
cleaninglady
Do the 2N3904s Q1-Q6 need to be matched in the Driver ?

Also , do the BC550C's and BC560c's in the core and driver need to be matched ?

Thanks !

Edit: Sorry for my laziness , found Q4-Q5 in the core need to be matched.

But the others ?
gbiz
cleaninglady I matched Q1/Q2 & Q3/Q4. I'd do that as a matter of habit whenever i see transistors paired like that. Been bitten too many times by using unmatched transistors.

setsun8 if you haven't got the resistors yet, get ones that physically fit (2mm dia, 3.5mm len). If you use conventionally sized 0.25W resistors, you have to stack them on top of each other to get them to fit.
cleaninglady
Thanks gbiz !

Will match those trannies.

I've done my resistors already. Stacked a few... hihi
qfactor
cleaninglady wrote:
Thanks gbiz !

Will match those trannies.

I've done my resistors already. Stacked a few... hihi


If you see this link on Pg 5 of this thread, the 9th message (by Siri) talks about the trannies not necessarily be matched more than the resistors. I guess if you matched the trannies, it wouldn't do any harm either, so that's optional! thumbs up

Here's the link to the page:
https://www.muffwiggler.com/forum/viewtopic.php?t=117137&postdays=0&po storder=asc&start=100
gbiz
qfactor wrote:
cleaninglady wrote:
Thanks gbiz !

Will match those trannies.

I've done my resistors already. Stacked a few... hihi


If you see this link on Pg 5 of this thread, the 9th message (by Siri) talks about the trannies not necessarily be matched more than the resistors. I guess if you matched the trannies, it wouldn't do any harm either, so that's optional! thumbs up

Here's the link to the page:
https://www.muffwiggler.com/forum/viewtopic.php?t=117137&postdays=0&po storder=asc&start=100


Definitely thumbs up
qfactor
jules wrote:

Once again: have a look at the schematics and the boards, that fix will work for the first (green/logicgate/barcode) run of PCBs. For the subsequent ones (black) you do have to cut traces.


Wait a minute! You mean those who have been active on this thread for the past 3 pages were NOT building this off the black PCB (Leftylogic's) version??!! eek!
Me included. d'oh! I've got one green PCB version that I've been building on and have not started on the Black PCB version yet.
But I thought Leftylogic mentioned that his black PCB version was identical to Logicgate/Barcode's green PCB? hmmm.....
Anyone here did the PWM/FM fix for the black PCB version by lifting the 2 resistor legs?
Randy
Mine are the Barcode versions. I also recall Leftlogic mentioning he didn't change anything. If you have both versions, you could always compare the two.

For what it's worth, I didn't match trannies or resistors and it sounds fine. I would be interested in comparing the sound of mine to the sound of a version that was matched, although I'm not sure what to listen for that would differentiate the two.

Randy
elmegil
whether matching matters or not in some ways depends on what kind of transistors you've got.

Many of us have found that if you have transistors on tape, typically one and its neighbor are already pretty well matched.
gbiz
Seeing as i'm the only one who seems to have used matched transistors, the only reason i used them is that i would always do that when i see a pair of transistors in a circuit. As i say, i've been bitten too many times by unmatched pairs in the past, i now just do it as a matter of course.

The only reason i emphasised that i used them was that others seemed to be having difficulties getting the PWM pot to function correctly across it's full range. The only differences i had on the 6 boards that all worked correctly were (a) the 3 resistors at values from the original schematic (and which Benjamin AM also seems to have had success with), and (b) matched transistors.
qfactor
I too have matched my transistors but only Q1 and 2. Also changed the 3 resistors as Benjamin's thread. Didn't match Q3 and 4.

So far finished 2 Osc cores and the Driver core and seem to get 1 Osc pretty much 'tamed' but still have the 2nd one that has a bit of issue calibrating over the octaves.
Probably will finish my 3rd core by Wednesday or Thursday.
cleaninglady
OK , thanks all. Really appreciate the help.

I've just got so much on the go right now.

I need to finish dozens of CGS pcb's for a Serge Custom panel , RYO optodist and Living VCO's.

The BOMs are killing me. Dead Banana
cleaninglady
One more question about C3 in the Core.

I have a grab bag of styrenes that has some 2.49nF's in there.

I've tested them all and one comes as close as 2.6nF and the rest are a bit more off but no more a than around 10%.

Do you think these would be OK to sub for the 2.7nF styrenes ?

I'm assuming yes... hmmm.....

PS

Does anyone have a scan or photo of the Living VCOs Core logicgate V1.3 and same of the. Driver V 1.1 ?

I'm doing some of the resistor changes and can find a picture I can actually make out values on.
qfactor
Did a visual check as well as a connectivity check between the green pcbs and black pcbs versions and they seem to be identical! seriously, i just don't get it
Anyone build the black pcbs version and did the resistor lifting of the 2 resistors to fix the PWM and FM issue?? hmmm.....

Also, here's a little doodling of one of 2 of my lvco. This is the green pcb version I completed:

duff
qfactor wrote:
Did a visual check as well as a connectivity check between the green pcbs and black pcbs versions and they seem to be identical! seriously, i just don't get it
Anyone build the black pcbs version and did the resistor lifting of the 2 resistors to fix the PWM and FM issue?? :hmm


I haven't finished mine but I lifted the resistors to do the fix. You can follow the traces to see these boards have the same issue as the green versions.
qfactor
jules wrote:
Once again: have a look at the schematics and the boards, that fix will work for the first (green/logicgate/barcode) run of PCBs. For the subsequent ones (black) you do have to cut traces.

duff: my thread was in relation to this by jules.
If you do have the black ones and have lifted the resistors and they work, then I guess it can be confirmed that BOTH green and black pcb versions can be fixed this way, coz cutting traces is more troublesome than just lifting the resistor legs. seriously, i just don't get it
duff
qfactor wrote:
jules wrote:
Once again: have a look at the schematics and the boards, that fix will work for the first (green/logicgate/barcode) run of PCBs. For the subsequent ones (black) you do have to cut traces.

duff: my thread was in relation to this by jules.
If you do have the black ones and have lifted the resistors and they work, then I guess it can be confirmed that BOTH green and black pcb versions can be fixed this way, coz cutting traces is more troublesome than just lifting the resistor legs. seriously, i just don't get it


I hadn't seen that post, and I must admit I have not tried my cores yet - I had to re-order a few resistor as I had none that fit in my component stash and got side tracked building something else. I did do the fix though and was of the understanding that the LeftLogic boards were identical to the green ones. I followed the traces to see where they went to check I was doing it right and all appeared as expected on that basis.

Unless I've missed something, which could well be the case, what were the inputs are just connected to the middle leg of the pot so running that to the relevant resistor leg is not an issue and should not require a cut. The traces that went to the other leg of the pot only go to the end of the resistor that was lifted, so I don't see why a cut would be needed there either. With the bonus extra holes available on the pots in goes straight to the relevant leg, or could even go to the spare leg hole of the lifted resistor if so desired. I think this should then match up with the core schematic + the wiring option diagram , although my schematic skills are not the best.

If that is wrong it would be helpful if Jules could be a bit more explicit regarding which traces need cutting.
duff
Posting issues... move along
jules
So with the black version (lefty logic) of the PCB, lifting one side of R2 and R38, and reconnecting them to the middle pin of the pots:
if you "multiple/divide" a modulating signal at least twice (before going to FM or PWm), it doesn't disappears and everything works? It just didn't work for me.
And what's the behaviour of your potentiometer for the FM and PWMod when you've done that?
SMonk
If one would want to drastically cut the upper range of the Frequency knob, would it be as simple as putting a lower range pot there or is it more complicated than that?

Right now i have a range from 16Hz to 16kHz or so. Since i also have fairly stable tracking over at least 6 octaves, it's well beyond human hearing. A pot range of say 3-4 octaves, from 16hz to 128-264 or so would be wholly sufficient for me, and make it much easier to dial in a correct pitch. If i want to harass dogs or young folk i can use external CV.
nigel
Is there a schematic for the sine converters somewhere? I looked through the old threads but couldn't find anything.
the bad producer
I have a coarse tune pot and a fine tune pot, same value pot but the fine tune goes to the CVSN via a 4M7 resistor off the wiper

SMonk wrote:
If one would want to drastically cut the upper range of the Frequency knob, would it be as simple as putting a lower range pot there or is it more complicated than that?

Right now i have a range from 16Hz to 16kHz or so. Since i also have fairly stable tracking over at least 6 octaves, it's well beyond human hearing. A pot range of say 3-4 octaves, from 16hz to 128-264 or so would be wholly sufficient for me, and make it much easier to dial in a correct pitch. If i want to harass dogs or young folk i can use external CV.
qfactor
nigel wrote:
Is there a schematic for the sine converters somewhere? I looked through the old threads but couldn't find anything.


Yes there were several times this got posted up on this thread! The latest being on Page 17 (6th thread from the bottom) Guinness ftw!
nigel
qfactor wrote:
nigel wrote:
Is there a schematic for the sine converters somewhere? I looked through the old threads but couldn't find anything.


Yes there were several times this got posted up on this thread! The latest being on Page 17 (6th thread from the bottom) Guinness ftw!


D'oh! Thanks.
cleaninglady
Does anyone have a bare / empty PCB scan or photo of the Living VCOs Core logicgate V1.3 and same of the. Driver V 1.1 ?

I'm doing the resistor changes and can't find a picture I can actually make out values on.
Siri
@cleaninglady
cleaninglady
You are awesome Siri !! SlayerBadger! SlayerBadger! SlayerBadger!
Siri
@cleaninglady

no worries thumbs up

just finished mine!





Worth the trouble love
qfactor
Wow!! Fantastic, Siri!! applause applause

I agree with you on it being worth the challenge of going thru it all! nanners
One thing I noticed when I used all 3 Sawtooth waves together, going into a mixer. They seem to cancel one another out and I get *almost* no sound? hmmm.....
It's definitely going out of phase with one another because when I play around with the mixer volume for each Osc, the levels would come back in when I turn one down, but the moment all 3 are turned up full, they seem to cancel out the sound.
Is anybody getting this? Only the Sawtooth seems to do this, though the other waves beat against one another too but not so out of phase as the Saw.
colorbars
I saw a mouser cart for the core pcbs does anyone have one for the driver?? w00t
leterell
https://www.mouser.de/ProjectManager/ProjectDetail.aspx?AccessID=c63ef 68f45

EDIT: so please omit the resistors here, since they are too big.

i will order the ones kindly suggested earlier by gbiz, LR0204F from TE connectivity. from farnell, mouser has a kind minimum quantity of 4000.

for me this will not include the ones over 1Mohm, and the 0.1% ones, these i will just cram on there somehow Guinness ftw!
colorbars
Thanks!! SlayerBadger!
nigel
qfactor wrote:
One thing I noticed when I used all 3 Sawtooth waves together, going into a mixer. They seem to cancel one another out and I get *almost* no sound? hmmm.....
It's definitely going out of phase with one another because when I play around with the mixer volume for each Osc, the levels would come back in when I turn one down, but the moment all 3 are turned up full, they seem to cancel out the sound.
Is anybody getting this? Only the Sawtooth seems to do this, though the other waves beat against one another too but not so out of phase as the Saw.


The Saw wave goes from 0V to 10V (roughly), instead of -5V to +5V like the other outputs. I suspect that when you mix the three, the signal is clipping at some point.
LeftyLogic
Hey all, sorry I've been away for so long, I haven't been able to get on the forum much lately.

After getting caught up with the discussion over the last few pages, there are a couple of things that I'd like to point out:

Firstly, there are absolutely no differences between my PCBs, (the black ones) and Barcode's (the green ones). For my run I used Barcode's GERBER files exactly as posted, so any mods used on his PCBs should also work perfectly on mine. thumbs up

Secondly, to colorbars and leterell, I would personally not recommend using resistors from Mouser on this project. It turns out that the holes for the resistors are a little too small, and most brands physically won't fit. I've used resistors from Tayda on my builds, and due to their extremely thin lead size they work perfectly. Also, Tayda's having a 15% off sale right now, so now would be a great time to stock up!

Anyway, I finally have some free time, so I'm working on getting my eurorack build finished up and will try to get some pictures of that posted soon.

Fantastic job to everyone who's built this so far! applause

Seth
leterell
Quote:
It turns out that the holes for the resistors are a little too small, and most brands physically won't fit.


thanks for the heads up!

i dont have the boards at hand now, and wont in a couple of weeks, does someone have an exact diameter of the holes?

the resistors in this cart are KOA speer, and according to the datasheet those have 0.6mm, the Vishay's, of which there also are a couple, have 0.6mm. and the TT electronics ones that there are, are 0.6mm.

then again, someone mentioned, that normal body lenghts of resistors are too big, so a 7,1mm length is too much? the KOA speers are this length...

i cant find the info on tayda's site what the lead diameters there are, they dont give that info on the resistors.

i also received some resistors from mouser at some point that were a bit more robust. i'd rather not order resistors from a separate place since the total price would not be in the 'free postage' range, so the info would be helpful.

ahhhh damn so nice to have the euro friendly pcb size but there really are some things to consider as a downside!
gbiz
leterell If you use resistors with a 3.5mm length body, they'll have correct diameter legs, and you won't have to stack them in order to get them to fit. Check back through the thread. I know that at least both myself & Altitude909 have posted types that are suitable. You can buy these at Mouser, Farnell etc.
yan6
I got the three vco's up and running just fine, but I dont think my driver is doing anything? I have LM13700N's in place of the LM13600, is that ok?

I have checked power on the chips and it looks good, and the triangle to sine converters are working fine.

any suggestions where to start digging on the rest of this
qfactor
yan6
How is the driver not working?
1V/Oct not pitching the Oscs? Vibrato Depth/Speed? Portamento not gliding?

If anything, it could be the 3 Track/Free switches soldered wrongly perhaps?

Oh and the 13700 is fine. thumbs up
STOJ
logicgate wrote:
LektroiD wrote:
Anyone got a wiring diagram for the 4 boards?



Here you can find 3 ways to wire up the boards:

http://www.jhaible.com/living_vcos/jh_living_vcos.html


This link is dead. Does anybody know where can I find schematic and wiring diagram.
elmegil
presumably somewhere here

http://www.oldcrows.net/~jhaible/living_vcos/jh_living_vcos.html
yan6
qfactor wrote:
yan6
If anything, it could be the 3 Track/Free switches soldered wrongly perhaps?


No they are soldered correctly but this made me double check them and I found the solder tips of the pcb were touching the outer, metal case of the switch, essentially grounding the driver outputs. Put some tape over the switch case and all seems much better now hihi
STOJ
elmegil wrote:
presumably somewhere here

http://www.oldcrows.net/~jhaible/living_vcos/jh_living_vcos.html


Thank you elmegil.
audiohawk
leterell wrote:
Quote:
It turns out that the holes for the resistors are a little too small, and most brands physically won't fit.


i cant find the info on tayda's site what the lead diameters there are, they dont give that info on the resistors.


Answer from Tayda...

What is the body lenght? ( about 5.70 mm.)
What is the legs diameter? ( about 0.40 mm.)
Siri
elmegil wrote:
presumably somewhere here

http://www.oldcrows.net/~jhaible/living_vcos/jh_living_vcos.html


How did you know? RandomSource should put a link to this site while they are rebuilding the site ...

Concerning the resistors i got mine from tayda too, works nice.

Also the resistors sold at dipmicro fits. Cheap price & cheap delivery (1 week). I don't know if they carry all the values though.

qfactor
So....with the announcement of the "official" LVCO pcb being able to do Sync'ing of its 3 Oscillators, I was wondering, would this be something that can be modified on our version of the LVCO as well?? hmmm.....
Altitude909
I've decided to pass on this project so if anyone wants my kit, send me a PM. I have one stuffed board and all correct size resistors, some other parts for the other boards, pots, and panel. Will sell everything at what I paid for it (~$100)
Grumskiz
qfactor wrote:
So....with the announcement of the "official" LVCO pcb being able to do Sync'ing of its 3 Oscillators, I was wondering, would this be something that can be modified on our version of the LVCO as well?? hmmm.....


BLASPHEMY!

Just kidding, I actually think this would be a great addition! I might leave out the Sine wave outputs to have individual sync inputs if that is in fact a possibility
leterell
so, i finished my build, got all 3 cores plus the driver going. hyper

HUGE thanks to everyone sharing information in this thread, and to the people who did all the hard work to make the PCB's happend!

the sounds that come out of this thing are pure gold. when they come out. osc 1 is still weather-dependent or something on the pulse, just puts out -5v on some days.

so i can definitely recommend this build! i have a year of SDIY experience, no scope (ableton + line in for tuning duties) and a 10€ multimeter and i managed to pull it off.



here's a pic (please excuse the horrific quality, my htc is 4,5 y old Dead Banana ) of my custom panel (works great, even the 3590s's fit back there, although not even a millimeter between the top rail and them), decided to attach a noise source and a mixer on the panel. i really like the MFOS noise cornucopia, with the options i crammed on the panel especially. and the RYO ampmix brings a nice grit to things when needed. might change up for the newer version tho, that seems to work more versatile.

the rotary switches for waveform selection are a personal preference, i really do that rather than repatch a cable each time.

i flew the v/oct and hi-freq trims to a separate stripboard, and this i can highly recommend. i did this after i had assembled and trimmed everything, only to note that there was prob a quarter of a semitone discrepancy in the oscs over 2 octaves. i mean i want it to 'live' but that was too much for playing anything. after retuning it kept in tune nicely for 3-4 hours of playing yesterday evening, i mean over 4-5 octaves with 3 parallel oscs.

also, the pw goes from what sounds like 10% to 90% in 1/10 the pw pot's movement with osc 2 and 3, go figure... but i like the 50% pulse the most anyway...


would have some questions tho:

1. i get out a very nice sine shape optically, but it always has a sharp peak at one end, cannot get rid or that on any of the converters. could i trim something to iron this out?

2. is there a simple way to bias the sawtooths even a bit? getting any volume out of them is quite hard mixing all 3... of course i could just ass a -v from my MVP in the mix but that's also omitting one of my sources from the mix waah

3. the vibrato depth pot starts to work at 12 o'clock, ccw from that no effect. any easy way to try to trim this down? i played with the resistors close to the pot to no effect, not even magic smoke.

4. how could i go about setting the rate of the vibrato lower or increasing the range? it's now 1-2Hz at its slowest, i'd like something like 0,1-0,2Hz for the low end.
Randy
leterell, impressive build, congratulations! I need to get back into mine and trim the v/oct again, but unlike you, I left the trimmers on the board so I need to take it apart to do it. Argh!

If you read (backwards) through the thread you'll find fixes for some of the issues you are experiencing.

Randy
leterell
Quote:
If you read (backwards) through the thread you'll find fixes for some of the issues you are experiencing.


thanks Randy. Yeah, i guess there are suggestions as to the vibrato depth will have to follow those.

but on the other issues i couldnt find fixes. only the mention that vibrato rate is deliberately limited to 2-12hz. well, with the vib depth pot i can live, but if someone could point me to a component it would make sense to tamper with to get slower low end, i'd be very grateful!

i have the feeling that if i go on my gut feeling on 'now this resistor seems like it'd have something to do with the LFO rate' i think i WILL see magic smoke MY ASS IS BLEEDING
sammy123
Nice build leterell. I love a shot of the backside if you have it out of the rack.

I would describe my sine as looking ok on the scope as well, but buzzy. Not that I mind that much. I actually was doing an experiment for a friend (more of a joke) and you can clearly hear the buzzy quality in the first three sines in this video. https://instagram.com/p/4efk4JJNmd/
leterell
Quote:
I would describe my sine as looking ok on the scope as well, but buzzy. Not that I mind that much.


judging from the vid my sines are right in your ballpark. i also dont mind it that much, just thought about an easy fix, a couple of components or so.

Quote:
I love a shot of the backside if you have it out of the rack.


will snap one for you the next days when i have to tune the thing up again =) judging from experience it wont be long haha
idealer
Hi.. great to see all nice lvcos in this thread
But i bet no one have as cool waveforms as mine do hihi

The question is how i can uncool them a bit... seems like the saw is biased to about +0.5v and is a little triangular.. i post a picture and let you have a look at it
It applays to all three om my cores

The other waveform is the tri

I have been pooking around with r21 and r22 ..no change
So could it be crappy lm311... leaky timing cap?
Well i will turn every stone untill i find it but i find it a good idea to ask you guys for advise first

Ps ... it is the first time i publish a image here so it might not work

idealer
Zaibach
Finally I had the time to work on my Living VCOs a bit more... I installed all the mods and redid the panels. I desoldered all the trimpots (exept for the HF trim) and put them on seperate boards on the back (as recommended by someone on this thread), and moved the trimpots on the driver to the other side for ease of access. I leveled up my desoldering skills several times while doing this, because every single trimpot I desoldered still works hyper
Only a few pads lifted, but nothing serious. The only things left to do are installing 10-turn pots for frequency and kludging tempcos on the VCOs (and cutting out the normal resistors I put in there) ...

Hopefully I won't have to take this monster apart again after this...

qfactor
Nice! applause

In case anyone is looking for alternative sources for the 10-turn pot, I found one on eBay that I bought from that is reliable (bought a few times from him alread and everything seems to work fine!)

Look for seller:michael99514
I've also ask him for components that he doesn't list and most of the time he'd be able to get it as well. They're all from China, of course!

Hope this helps! thumbs up
Zaibach
qfactor wrote:
Nice! applause

In case anyone is looking for alternative sources for the 10-turn pot, I found one on eBay that I bought from that is reliable (bought a few times from him alread and everything seems to work fine!)

Look for seller:michael99514
I've also ask him for components that he doesn't list and most of the time he'd be able to get it as well. They're all from China, of course!

Hope this helps! thumbs up


I just looked him up, he's got the 10-turn trimpots for 5$ plus shipping, which is not bad. Reichelt.de in germany has them for 6,99€...
Siri
just to confirm that I am 15V version and i had to use, as logicgate suggested, 500k trimmers for OCTTUNE1 and OCTTUNE2 on the driver board.

I wish the linear detune knob was less subtle... On JH front panel you can see that his knob can go from flat to sharp. He even says "while the real interesting feature is the low range animation - the "linear detuning" stuff..

But still I'm happy about it It's peanut butter jelly time!
Rod Serling Fan Club
Possibly a silly question:

I believe I read that the synthi oscillators are tuned around .32 v/oct. At that tuning the voltage swing from another oscillator, trapezoid gen, etc. would be a lot more dramatic (approximately a 3 octave shift per volt rather than 1). It seems to me this is part of what is special about the way a synthi sounds. Just looking for other users thoughts on this matter.
bensaddiction
How about a decade counter on the Multi Turn.... Over kill? Would they fit?

Its only a buck more for the counter.

Like these:

Ebay Multi Turn
bensaddiction
Late start to my build. But its all done and its sounds amazing.

One thing though... Two of my oscillators tune and track well enough. But one VCO is out by 1 semitone. As in once its tuned to say C3 I don't get C4 unless I press C#4 on my CV keyboard

The scale trim pot doesn't fix the issue for this oscillator (it did for the others). The tuning gets closer part way through the scale trimmer's range (multi-turn) and then its starts opening up the difference again. I can't get vaguely close at all with VCO#1

This is with both the controller driver, and all the other possible v/oct inputs from . Does anybody know where I should look for a short or other error on the board?

The TL074 on that VCO was also a little warm. What would normally cause this and would that be the cause of the issue or its that chip for waveshaping for other wave forms?

The hardest solder points on the PCB have been the transistors which don't have the usual separated spacing you normally see on DIY boards and are easy to bridge, the pads for the SMT caps are also a little conservative and so I can't guarantee all of those connections.
billsship
All I have left are the transistors, tempcos, and wiring it all up. I'm so afraid I've messed up one little part, but I've been thorough so fingers crossed.

Should I match the transistors? I didn't buy enough to match them, and the BOM doesn't say anything about it, so I'm guessing no. Although others in this thread have said it's a good idea.
Silesius
Hi, I've finished my LVCO's with the Barcode boards. I've used some of the fixes found in this thread, so many thanks to everybody.
Surprisingly, everything seems to work fine: nice waveforms, the vco's track for about 5 octaves and the driver works as expected.

The only thing that bothers me is when you engage the "track" switch: even with the vibrato depth pot at minimum, you still can hear the driver modulation. It's subtle, but it's there. If you move the pot CW, the modulation increases (as expected), but you can never turn it completely off using the pot.

I've searched through this post and it seems to be a fix for the "vibrato depth problem", involving increasing resistors 158 & 160. But it seems that's meant to fix the pot when it has too little effect, and not the other way around.

Is anybody experiencing this? How about normalling my vibrato jack to ground so the circuit sees ground when nothing is plugged? Any ideas would be very welcomed...
unnu_
I am now building the Living VCOs following the schematics only, and from my readings so far some people noticed two differences between the original schematics and the pcb- variant:
1. R40 should be referenced to 0 and not to -vb;
2. there is an extra R6 from the junction of R30/R31 to 0; i would like to ask you what value is this mysterious R6...
qfactor
unnu_ wrote:
I am now building the Living VCOs following the schematics only, and from my readings so far some people noticed two differences between the original schematics and the pcb- variant:
1. R40 should be referenced to 0 and not to -vb;
2. there is an extra R6 from the junction of R30/R31 to 0; i would like to ask you what value is this mysterious R6...


From the BOM by Barcode, it says that R6 in the vco pcb is 10k.
Are you referring to this same one?
unnu_
thanks, qfactor, that's it. Somehow i've missed the BOM on the very first post of this topic meh
qfactor
Hi,

For those of you who built this version of the LVCO, would you happen to know how much current it uses?

Today, I hooked up one (of 2) of my LVCOs along with 2 TH Mega Percussive Synth and used the LVCOs pulse wave (at its lowest freq setting) to trigger the 2 MPS and left them to "burn in" for a few hours. All went well.

Then I decided to hook up the 2nd LVCO. So had 2 LVCOs and 2 MPS powered up via a uZeus and this resulted in one of the LVCOs Not being able to work and when I touched the uZeus panel, it was burning HOT!!
This is the 1st time I'm experiencing a uZeus not being able to power up 4 modules (2 LVCOs and 2 MPS)!! I'm not powering any other modules but these 4 only.

Do the LVCOs take up so much current? seriously, i just don't get it
Grumskiz
I suggest you measure the current draw yourself, because this way you can rule out any faults in your LVCOs. At least one may have developed a fault or short somewhere.
medbot
Well each LVCO is like four modules in one, so if you think about it like that maybe it's not such a stretch.
Jop
I did measure the current and got below results, quite a load per panel (3x vco & driver):

qfactor
Jop wrote:
I did measure the current and got below results, quite a load per panel (3x vco & driver):



Thanks for the measurement, Jop! Guinness ftw!

Another question:
When I mix 3 of the Saw waves for a 3 Osc sound, I seem to get cancellation of the waves that makes me think they are out of phase with one another?
Do any of your LVCOs have this issue?
This is not the Random Source version, btw.
sduck
I don't get that with my r*s version, either with an external mixer or the on board one. What version are you using? (edit: doh Drunken Homer Simpson didn't see which thread this was, never mind)
qfactor
sduck wrote:
I don't get that with my r*s version, either with an external mixer or the on board one. What version are you using? (edit: doh Drunken Homer Simpson didn't see which thread this was, never mind)


As mentioned, mine is not the r*s version but LogicGate's pcb.
I'm trying to identify if this is the case for those who built this version, or I messed up in my (possibly) inverting/non-inverting op amp part of the schems d'oh!
cleaninglady
I still haven't finished my build, i've had the finished PCB's sitting in my backlog for two years. If i can get onto it soon i'll let you know what i can see.
qfactor
cleaninglady wrote:
I still haven't finished my build, i've had the finished PCB's sitting in my backlog for two years. If i can get onto it soon i'll let you know what i can see.


Haha! And I thought my completing two of these over a 1.5year period is long as it is! hihi
thumbs up
nigel
qfactor wrote:

When I mix 3 of the Saw waves for a 3 Osc sound, I seem to get cancellation of the waves that makes me think they are out of phase with one another?


I built an older version, and the Saw was from 0V to 10V instead of -5V to +5V. When you mix them together they will very quickly clip against the positive power rail, which can sound like phase cancellation. Not sure if that is the problem here, but it seems possible.
qfactor
nigel wrote:
qfactor wrote:

When I mix 3 of the Saw waves for a 3 Osc sound, I seem to get cancellation of the waves that makes me think they are out of phase with one another?


I built an older version, and the Saw was from 0V to 10V instead of -5V to +5V. When you mix them together they will very quickly clip against the positive power rail, which can sound like phase cancellation. Not sure if that is the problem here, but it seems possible.


Hmm... That might be possible!
Looks like I've got to take it out of the case (been trying to avoid that!) and hook it up to an Oscilloscope to check this. thumbs up
Thanks!
glennfin
Is there a build thread for the Random source Eurorack version?

.... If there is, I can't seem to find it...

I'm trying to locate a part number and a U.S. source for the tempco resistors required for the R*S pcb...
Dogma
leterell wrote:
https://www.mouser.de/ProjectManager/ProjectDetail.aspx?AccessID=c63ef 68f45

EDIT: so please omit the resistors here, since they are too big.

i will order the ones kindly suggested earlier by gbiz, LR0204F from TE connectivity. from farnell, mouser has a kind minimum quantity of 4000.

for me this will not include the ones over 1Mohm, and the 0.1% ones, these i will just cram on there somehow Guinness ftw!


Is this for the Rs living vco?
Synthbuilder
glennfin wrote:
Is there a build thread for the Random source Eurorack version?


This one:

https://www.muffwiggler.com/forum/viewtopic.php?p=2071084

Quote:
I'm trying to locate a part number and a U.S. source for the tempco resistors required for the R*S pcb...


Find them at Mouser, their part number is:

279-LT300014T261K0J

This is the small 3000ppm/K one. You get slightly better compensation by using the KRL 1K 3500ppm/K but these are much bigger and their leads don't go through the holes in the PCB.

Tony
glennfin
Thank you Tony.

Quite the amount of components to source on this project. Mr. Green

Synthbuilder wrote:
glennfin wrote:
Is there a build thread for the Random source Eurorack version?


This one:

https://www.muffwiggler.com/forum/viewtopic.php?p=2071084

Quote:
I'm trying to locate a part number and a U.S. source for the tempco resistors required for the R*S pcb...


Find them at Mouser, their part number is:

279-LT300014T261K0J

This is the small 3000ppm/K one. You get slightly better compensation by using the KRL 1K 3500ppm/K but these are much bigger and their leads don't go through the holes in the PCB.

Tony
Starspawn
Do you still need the tempcos if you opted for the supermatched smd transistor pair? They wont be in thermal contact anymore for one ...
sduck
Yes, you need them still. And they still can be in contact. The smaller variety of tempcos are easier to use in this application.

Edit: I didn't notice which LVCO thread this was - so my answer may not be correct. In general though, if a tempco is needed, you at least need some kind of resistor in that spot for it to work.
Silver
Does anyone still have any panels available for the Barcode edition? I got the pcbs, but I'm hesitant to build until I know I can get a panel (without making it myself. Preference would be for lower hp, but at this point, any would be good. Please pm me if you have one to sell. Thanks
nd595
Just finished mine - it took some time as I had a very busy period.
I made all possible mistakes - applied a wrong polarity to driver board AND to one of VCO's - nothing serious, just had to replace all chips and electrolytic caps; strangely, all transistors are fine.
Tracking is as perfect as it can be - excellent over 7 octaves. Mutable Module Tester is a great tool, as it saved me lots of time.
The only thing left is that spike on triangle and sine converters; even Robert Moog himself had that problem with his triangle converter and he sorted it with an extra capacitor. Will have to deal with that, otherwise - this is a great module, not easy to assemble, but it was enormous pleasure.
Million Thanks to logicgate - hooray!!!
pix
I have a problem with my vibrato circuit. When increasing the amount nothing happens until a point where pitch goes super low.

Vibrato rate has no effect.

any hints on where I should start looking for problems?

thanks!!
pix
help
elmegil
I don't have the Euro schematics handy, so I don't have proper reference values.

In the original JH schematics, page 4 of the schematics for the driver board shows that the vibrato is coming off of an LFO circuit made up of U14A and U14B. The rate is part of that circuit.

Then there is an LM13600 (or LM13700) OTA that's used for the depth to operate as a VCA on the output of the LFO.

So

1) verify the LFO is oscillating, measure the output of whatever the equivalent is of U14B.

2) if it is oscillating, verify that your OTA is functional (maybe just swap the chip if you have spares).
shiftr
After almost 3 years I'm nearly finishing my barcode edition living vco s
I also have a problem with the vibrato on the driver. It does seem to do something but it's very weird and subtle. I'm not sure if I'm expecting something different or if it doesn't work correctly.
sduck
There's a different thread for that version - try either this one

https://www.muffwiggler.com/forum/viewtopic.php?t=93892

or this one

https://www.muffwiggler.com/forum/viewtopic.php?p=1704049
shiftr
Thanks.. But I've searched those threads and they point back to this thread wink

There also seem to be a lot of pictures of similar PCB's of mine in this thread.
I guess I'm already lucky having a pretty good working vco. I'll have a look if I can fix the vibrato or maybe leave it like this. This is a project I'd really like to finish. Maybe I'll just use the module "as is" .
sduck
Oh nuts I misread the thread title. Yes, this is the place for your version. Woops, and sorry.

I've only built the random*source version, but it should be similar circuitry. And I seem to recall there was a resistor that needed to be tweaked to get a decent amount of depth from the vibrato - but I'm not sure which one it was, sorry. Maybe that'll point you in the right direction at least...
shiftr
hihi w00t
thanks! SlayerBadger!
shiftr
I found an error.
I put a 620k resistor in the vibrato circuit where there was supposed to be 510k. I found a whole collection of 620k resistors wrongly marked as 510k in my drawers.
I took out the 620k and the circuit works fine without it. I tried putting in a 510k resistor but i liked it better without. The vibrato runs slower without.
Now i've got a problem with one of the sine shapers. One om the generates a clipped sine.
This project seems more work then building a TTSH smile
cleaninglady
Because I am quite busy and lazy too, could any kind Wiggler point me to a wiring diagram for the the Living VCO's.
I have completed Barcode edition V1.3 pcbs sitting here that I finished about 5 years ago. I'm planning to just ratshack it all into one box so I won't need a panel. Actually as the pots are all PCB mounted and the outputs are obvious, all I really want to know is :

1. How do the Driver pads on each OSC pcbs connect to the Driver PCB ?

2. The switch wiring. (I'm going to look at the Option 3 Diagram on Jurgen Haible's legacy site).

EDIT : Problems above solved. Option 3 diagram was great ! SlayerBadger! SlayerBadger! SlayerBadger!
cleaninglady
More questions...

So, (first question), I was just checking over the BOM on page 1 of this thread, i'm sure that i never matched R1,R2 and R3.

Is there a high res image of the unstuffed PCB around ? (I have the
Logicgate 2014 Barcode Edition v1.3 PCB)

I can find R1 and R2 from making the first image above bigger but i can't find R3 anywhere. Dead Banana

Also, (second question) i have a very badly misshapen TRI wave on one of the OSC PCB's. It actually wen't from being just a Triangle with a spike to a bad TRI / SAW combination with a spike after i changed C7 and C8 to BP electrolytic.

Is a clean TRI wave heavily dependent on R1,R2 and R3 being matched ?
oscarthesquirrel
Not sure if anyone can help with this but here goes...


Finished building my Living VCO today, but I've got a problem. No output from any VCO...

OK - so I already have a LVCO (built by someone else) and I've used that as a reference during my build.. so I started by measuring a few voltages


The power header for the VCO section is reading -10.66V and 11.17V

I've also checked all 3 IC1's and IC7's (I'm using OPA2134s here for better tracking) and all 3 IC1s show 10.7V at pin 3 , with IC7's showing 12.48V at pin 3. The values from the working LVCO ICs are 11.9V and 12.2V respectively.

Voltage at the driver pin header seem fine at -11.95V and 12V

So I've checked for solder bridges, cold joints, IC orientation, transistor orientation, diode orientation and also cross referenced against the working LVCO.

I'm stumped as to where to start looking...I'm no electronics engineer.

Any immediate suggestions as to a potential culprit ??? The fact that all 3 VCO circuits have the same issue leads me to think I've installed the same thing incorrectly in all 3 circuits....but I can't figure out what that would be.
oscarthesquirrel
After a week of checking I'm no further forward.

Have literally checked every single component for value and placement, I've reflowed all the joints, i've checked IC sockets to make sure they're all seated OK..but still I have a non-functioning module.

Any kind-hearted soul out there able to help ?

Dead Banana very frustrating
Muff McMuff
I have noticed many times on muffs people posting photos and somebody points out potential issues straight away. I would post some good images.
elmegil
After you post photos, start with a single VCO.

Sounds like you've verified the voltages and components. Have you verified your panel wiring?

Are you sure the driver voltages are reasonable and not pushing you up into ridiculous inaudible rates or down into near-DC? If you can detach them you probably should.

When you adjust the frequency knob(s) do you get voltage changes at the output of the summing amp? I don't recall Barcode's designators and whether they match Juergen's original schematics, but talking about the output of U1A from the original.

Are you seeing any oscillation on the output of U1B? I hope you have a scope of some sort....
oscarthesquirrel
Attached is a pic of the board. Not the prettiest I know... if a higher res is needed, I can dig out my Nikkor 105 macro and send a monster file...

1. Yes, I have a scope
2. I have a working LVCO to compare voltages, components etc and that helped me spot 2 ceramics that were wrongly positioned in the driver section - that's now fixed..driver section works (I fed in quantised voltages from an A-155/156 combo and the signal through 1V/O seems fine)
3. Voltage at the pin header is weird +11.2V / -10.6V What might cause this voltage imbalance ?
4. Saw output steady +7.6V,across all 3 outputs (ie flat line)
5. Pulse output was 4.66V when I measured it, then I noticed it was slowly dropping..almost as if it was cycling very slowly..same across all 3 outputs+0.0008, +0.0041)
6. Sine outs -1.01, -0.4056, -0.2267 (flat line)
7.R33B is on the other side of the board - long story, but the solder simply wouldn't flow out when replacing R33B, so I ended up placing it on the other side. >,<
8. No panel wiring - am using the R*S bard set, so its a simple jacks, pots 'n' switches solder job and no complicated wiring.

Now the saw is the feed for the other waveforms via wave shaping, so I guess I need to start at IC1A/B/C. I've checked the transistors - correctly positioned, correct components

So where to go next ?

oscarthesquirrel
High res shot front of the board. Back to follow - for some reason can't upload - maybe file is too big.

Picture file
oscarthesquirrel
And the (ugly) back as an attachment
oscarthesquirrel
Latest update:

I pulled all of the ICs from the VCO section and measured the pin header voltage...perfect +12 V / -12V

So i started re-introducing the ICs one at a time and measuring...

All the OPAs - fine. Voltage is good at the header +/- 12V

The problem seems to be with the 311P chips I have - I suspect they're a fake batch from China. With each one I added in, the negative voltage dropped by 0.2-0.3V, and the positive voltage dropped too. I'd ordered them locally in Malaysia, so its a possibility they're duff...I've tried several from the batch in various locations and they all give the same (or worse) results.


The TL074CN ICs I've used also seem to be pulling down the voltage, although with them in place (and no 311Ps) the voltages are + 12v and -11.3V at the header.

So there are still issues, but hope this helps someone more knowledgeable than me to point in me a suitable direction

thanks !
elmegil
Quote:
4. Saw output steady +7.6V,across all 3 outputs (ie flat line)


This means you're right, the comparator section isn't working. The cap charges up, but never gets reset. I would suggest you find an alternate source of 311 chips.....
sduck
By the way, this isn't the build thread for the R*S version, this one is for a different flavor of LVCOs. The actual R*S thread is at - https://www.muffwiggler.com/forum/viewtopic.php?t=150988 - but it seems this one is working for you.
oscarthesquirrel
sduck wrote:
By the way, this isn't the build thread for the R*S version, this one is for a different flavor of LVCOs. The actual R*S thread is at - https://www.muffwiggler.com/forum/viewtopic.php?t=150988 - but it seems this one is working for you.


Yeah - sorry... I went for the thread that has [BUILD] in the title !

If you need to move this bit of thread across that's cool with me. thumbs up
oscarthesquirrel
elmegil wrote:
Quote:
4. Saw output steady +7.6V,across all 3 outputs (ie flat line)


This means you're right, the comparator section isn't working. The cap charges up, but never gets reset. I would suggest you find an alternate source of 311 chips.....


This makes sense...at least to a novice like me...thanks for the explanation.

New 311 chips already on order from Farnell.
elmegil
Sounds good. We can pick it up in the other thread if if continues to be an issue... smile
sduck
oscarthesquirrel wrote:
[

If you need to move this bit of thread across that's cool with me. thumbs up


Doesn't matter to me. Seems like it's doing fine here. And with this ancient software you can't really move sections of threads around anyway. Carry on!
oscarthesquirrel
Update..

All sorted...the new 311 ICs did the job. I must have picked up a bad batch (ie fakes)...let that be a lesson for anyone thinking of buying from eBay if they can't find the chips on Tayda, Mouser or Farnell.
elmegil
I buy nothing on ebay any more unless I can find absolutely no other source, and if it's something that ought to be expensive, not even then.

311's are not expensive, but like original SSM/CEM chips and such, would not touch ebay for the world. Not to mention a lot of those are getting re-manufactured again....
oscarthesquirrel
elmegil wrote:
I buy nothing on ebay any more unless I can find absolutely no other source, and if it's something that ought to be expensive, not even then.

311's are not expensive, but like original SSM/CEM chips and such, would not touch ebay for the world. Not to mention a lot of those are getting re-manufactured again....


I agree 100%.... I'd only ventured to Ebay because Farnell didn't have any stock locally, nor did Tayda and Mouser wanted to charge me a gazillion dollars to chip 3 chips to Malaysia ....anyway lesson learned. Dead Banana
MUFF WIGGLER Forum Index -> Music Tech DIY  
Page 1 of 25
Powered by phpBB © phpBB Group