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

Jurgen Haible Frequency Shifter (JH FS-1a) Build Thread
MUFF WIGGLER Forum Index -> Music Tech DIY Goto page Previous  1, 2, 3, 4, 5  Next [all]
Author Jurgen Haible Frequency Shifter (JH FS-1a) Build Thread
tommy.york
Posting to my own thread. It's peanut butter jelly time!

Still trying to fix this issue with negative flat cmpout. Did some poking around with the oscilloscope and some amateur schematic reading and such, I'm convinced that the issue is around U15 (tl071) near the bottom middle of the board.

Here are my notes. I've realized today that I was confusing Vp and Vpp, so initial amplitude was half what it should have been. Sending a sine wave into the input such that dry/out is 5v Vpp...

U15 (TL071)

Pin 1 [OFFSET N1] -14.93V
Pin 2 [IN-] probing pin 2 yields input signal, but small, 300 mVpp
Pin 3 [IN+ ] 0V, continuity to ground
Pin 4 [VCC-] -14.92V
Pin 5 [OFFSET N2] -14.92V
Pin 6 [OUT-1] -13.6V, voltage fluctuates a tiny bit, ac-coupled probe shows flat signal
Pin 7 [VCC+] 14.9V
Pin 8 [NC] 0V

I then tried to use that cool circuit simulator app to try and figure out what was going on with U15 and the relevant part of U16A (the LM13600). (simulation link). Btw if that's wrong please lemme know, this is my first attempt at modeling a circuit.

My next guess is to try to replace the lm13600, which I sourced from utsource, with a contemporary lm13700, on the off chance that the vintage chip is broken.

edit: That error was, in fact, a broken lm13600. Replaced all three of them with new lm13700's. The next issue I had was a misunderstanding about the placement about the jumpers. After that, I was able to calibrate as described.

Now, I get something that sounds like half shift, half carrier on both diff out and sum out. Is this what other people with successful builds have? Trying to figure out if there some kind of cancellation magic with one of the other outputs, given the QVCO.
muncky
Something of a necrobump- I stuffed this last year and Charlie Loudest Warning did the hard part of wiring and panel design. As yet unheard, but really happy to have this lovely noise box in my mitts!

sanders
I just thought someone might be interested to know, I connected my FS-1 to a 12v eurorack power supply the other day and it works fine. I didn't get to any elaborate testing of the signals, I simply A/B'd it against a MOTM PS I'd been using and listened carefully while adjusting parameters.

I think it probably sounds better with 15v, but the difference is not dramatic.

In all the years, I'd never read of anyone actually testing the FS-1 on 12v; everyone just seemed to take it for granted that it wouldn't work. But, it does. Good news.
Leverkusen
I would like to start building my FS now - has anyone an idea where to source the dome filter 1% capacitors?

Not sure if I understood it right but measuring and selecting them by hand seems to be a bit tricky, right?

I did not find any 1% caps anywhere that are not way to huge (physically) by now... hmmm.....
Haralds:Werk
Leverkusen wrote:
I would like to start building my FS now - has anyone an idea where to source the dome filter 1% capacitors?

Not sure if I understood it right but measuring and selecting them by hand seems to be a bit tricky, right?

I did not find any 1% caps anywhere that are not way to huge (physically) by now... hmmm.....


What is your problem with selecting by hand? I have done that many times for other projects. Just pick up a bunch and start to measure. OK, you need a decent LCR meter though. PM me if you need help.
Reese P. Dubin
I have been considering getting a second meter to compliment the trusty Fluke 73.

SO somebody suggest a good not insanely expensive one that can measure caps and transistors.

I have a populated FS1 set going back to when Jurgen was still with us that keeps shuffling to the bottom of the deck. I want to finish this thing now.
sduck
I exchanged some emails with Jurgen about this very thing. My fluke 115 doesn't read lower value caps very accurately, or at all below a certain point. So I asked how important it was to do this for all the caps, and his response was that whole approach as he describes for measuring caps and adjusting the resistor values to compensate was probably obsolete considering the relative uniformity of currently available parts. And that I could probably get away just fine using the default values he provides. So that's what I did for the values I couldn't measure with enough accuracy, and my FS1a works fine, no discernible carrier bleed.

As far as actually finding those styrene caps, I don't know - I know mouser still has some stock of at least some of the values. I bought enough for 2 or more of these back then - it was maybe 8 years ago though.
Starspawn
sduck wrote:
I exchanged some emails with Jurgen about this very thing. My fluke 115 doesn't read lower value caps very accurately, or at all below a certain point. So I asked how important it was to do this for all the caps, and his response was that whole approach as he describes for measuring caps and adjusting the resistor values to compensate was probably obsolete considering the relative uniformity of currently available parts. And that I could probably get away just fine using the default values he provides. So that's what I did for the values I couldn't measure with enough accuracy, and my FS1a works fine, no discernible carrier bleed.

As far as actually finding those styrene caps, I don't know - I know mouser still has some stock of at least some of the values. I bought enough for 2 or more of these back then - it was maybe 8 years ago though.


Thank you, Im in the same boat and have put off researching how to measure the smallest values. An excuse to not is welcome smile
Leverkusen
Haralds:Werk wrote:
Leverkusen wrote:
I would like to start building my FS now - has anyone an idea where to source the dome filter 1% capacitors?

Not sure if I understood it right but measuring and selecting them by hand seems to be a bit tricky, right?

I did not find any 1% caps anywhere that are not way to huge (physically) by now... hmmm.....


...OK, you need a decent LCR meter though.


I think that is the tricky point of the endeavor. I hope to find time to write you a pm on the weekend - I always wanted to do so anyway because of your awsome vocoder work.

sduck wrote:
I exchanged some emails with Jurgen about this very thing. My fluke 115 doesn't read lower value caps very accurately, or at all below a certain point. So I asked how important it was to do this for all the caps, and his response was that whole approach as he describes for measuring caps and adjusting the resistor values to compensate was probably obsolete considering the relative uniformity of currently available parts. And that I could probably get away just fine using the default values he provides. So that's what I did for the values I couldn't measure with enough accuracy, and my FS1a works fine, no discernible carrier bleed.

As far as actually finding those styrene caps, I don't know - I know mouser still has some stock of at least some of the values. I bought enough for 2 or more of these back then - it was maybe 8 years ago though.


So you used 2% styrene caps where possible and relied on the expected/printed values? I'll try this too then and see how far it gets me.

Thanks everyone!
Reese P. Dubin
sduck wrote:
I exchanged some emails with Jurgen about this very thing. My fluke 115 doesn't read lower value caps very accurately, or at all below a certain point. So I asked how important it was to do this for all the caps, and his response was that whole approach as he describes for measuring caps and adjusting the resistor values to compensate was probably obsolete considering the relative uniformity of currently available parts. And that I could probably get away just fine using the default values he provides. So that's what I did for the values I couldn't measure with enough accuracy, and my FS1a works fine, no discernible carrier bleed.

As far as actually finding those styrene caps, I don't know - I know mouser still has some stock of at least some of the values. I bought enough for 2 or more of these back then - it was maybe 8 years ago though.


Wish I read this 7 years ago!
Anyhow, great I am just gonna go for it.
sduck
Leverkusen wrote:

So you used 2% styrene caps where possible and relied on the expected/printed values? I'll try this too then and see how far it gets me.



That's pretty much it. I was able to read some of the larger value caps, but if I recall correctly only one (maybe 2) of them required any changes in the default resistor values.
Revok
I have most of my boards stuffed for this one but I'm getting hung up on the three 2SA733 PNP transistors. Outside of pinout differences is there any reason why I can't use 2n3906's here? I plan on heat shrinking the collector and base legs and crossing them.

Here's the area of the schematic if you're curious.

glennfin
I thought this might be the case.... The claim was made earlier in this thread that it absolutely needed 15v. I guess no one actually tried it until now... great to hear.. might breadboard up one now.


sanders wrote:
I just thought someone might be interested to know, I connected my FS-1 to a 12v eurorack power supply the other day and it works fine. I didn't get to any elaborate testing of the signals, I simply A/B'd it against a MOTM PS I'd been using and listened carefully while adjusting parameters.

I think it probably sounds better with 15v, but the difference is not dramatic.

In all the years, I'd never read of anyone actually testing the FS-1 on 12v; everyone just seemed to take it for granted that it wouldn't work. But, it does. Good news.
glennfin
The top post shows an "FPD" file??... don't know what this is...

Does anyone have a link to the full schematic?

Thanks.
marcn
Schematics are somewhere on this page along the old build/calibration notes
http://jhaible.com/legacy/frequency_shifter_fs1a/fs1a

As for the 1% caps, I used Wima. In Europe Reichelt has most of them in stock, takes a while to find the correct ones. I dont recall which ones exactly I used as i find their names pretty confusing but if you head to the Wima site its pretty straightforward to find the ones offered with 1% tolerance and to find them consequently on Reichelt.
glennfin
Got it! thanks. cool

Woah!!!... breadboarding that might take some time.. eek! eek! eek!

I don't suppose anyone has a PCB available?... Mr. Green I see photos of a PCB but can't find any mention about availability.


marcn wrote:
Schematics are somewhere on this page along the old build/calibration notes
http://jhaible.com/legacy/frequency_shifter_fs1a/fs1a

As for the 1% caps, I used Wima. In Europe Reichelt has most of them in stock, takes a while to find the correct ones. I dont recall which ones exactly I used as i find their names pretty confusing but if you head to the Wima site its pretty straightforward to find the ones offered with 1% tolerance and to find them consequently on Reichelt.
Starspawn
Available here:
http://randomsource.net/haible/vintage
keninverse
glennfin wrote:
Got it! thanks. cool

Woah!!!... breadboarding that might take some time.. eek! eek! eek!

I don't suppose anyone has a PCB available?... Mr. Green I see photos of a PCB but can't find any mention about availability.


R*S carries the FS-1A
glennfin
Thanks..... Has anyone built a euro version? (Euro panel ?) Guinness ftw! Guinness ftw!

Starspawn wrote:
Available here:
http://randomsource.net/haible/vintage
marcn
glennfin wrote:
Thanks..... Has anyone built a euro version? (Euro panel ?) Guinness ftw! Guinness ftw!

Starspawn wrote:
Available here:
http://randomsource.net/haible/vintage


I did.. sort of. Pcb's are quite large and frankly the unit is more useful in standalone i find.
Revok
glennfin wrote:
Euro panel ?


It is two 160mm x 100mm pcb's. I'm building mine as a stand alone unit.
sduck
glennfin wrote:
The top post shows an "FPD" file??... don't know what this is...



.fpd is the file format supported by Front Panel Express's software, used for panel design. Go here - https://www.frontpanelexpress.com - download their free software, install it, and then you can open .fpd files, edit or design them as you see fit, and have them made for you.
Revok
I've finished mine using 1% C0G capacitors and the default resistors for the Bode filter. It works as expected. There aren't many demos to compare it to but I would doubt there are any noticeable differences. They fit rather nicely too.

When Jurgen said you *must* use shielded cable on certain connections he meant *must* Dead Banana. After experimenting with regular hookup I found that the Mic Input, Mic Level, and Mic Gain were the most important for this. You might get away with leaving it out elsewhere but I replaced all of it after screwing around.

I would highly suggest using the connectors for the build rather than hard wiring. It made troubleshooting and things like adjusting the switch positions on some of the stuff that isn't immediately apparent until you hear it much easier.

If you are using the built in power supply you should definitely use heatsinks. I may end up swapping my transformer but the Hammond 186C36 I'm using runs a bit higher than the 36V it's supposed to put out so the heatsinks are getting pretty hot. I've run for a few hours so I don't think it's close to putting it into thermal shutdown but I would like it a bit cooler.

One more thing, the V/Oct calibration is not in the adjustment instructions so don't forget about it like I did! It's R8 on the main board.

edit: grammar
Luka
Nice work, noticed your photos in the other thread. Post them here too.

I have mine running from a half wave rectified psu using 15vac wall wort and it runs okay. It gets a bit warm but I have left it on and haven't had any dramas.

What is your measured current draw?

I think I used shielded cable with the mic as per the wiring schematic but interestingly in my vocoder which uses the same mic amp circuit I have gotten away without it. Though it is something I am considering rewiring when my final panel comes.
Revok
Thanks, Luka!





I cant find a good spot to check the current without taking off the heatsinks to lift something but maybe I'll mess with it this weekend. I think it has more to do with the 22.8V AC on each side of the transformer which is hitting the regulators with a measured 26.8V DC. I might give the 28V center tapped version 187C28 with the same footprint a shot.

I think it will work hmmm.....

C = 0.7 * I /(ΔV * F)
0.0004=0.7*.110/(ΔV*120)
ΔV≈1.6V

Vp=14V*√2≈19.8V

As for the shielded cable, it might be the length that's most important! The top of my enclosure is pretty much at the opposite ends of what's going on underneath so some of those cables are as long as 25cm.
MUFF WIGGLER Forum Index -> Music Tech DIY Goto page Previous  1, 2, 3, 4, 5  Next [all]
Page 4 of 5
Powered by phpBB © phpBB Group