DUSG Help!

From circuitbending to homebrew stompboxes & synths, keep the DIY spirit alive!

Moderators: lisa, luketeaford, Kent, Joe.

groove
Common Wiggler
Posts: 222
Joined: Fri Jul 19, 2013 10:21 pm
Location: NY
Contact:

Post by groove » Thu Oct 12, 2017 7:31 am

cygmu wrote: Do the other controls (CV in, 1V/Oct) affect the rise time?
No, nothing seems to change the rise time, which is very fast(haven't scoped it, yet).

User avatar
cygmu
Super Deluxe Wiggler
Posts: 1501
Joined: Mon Feb 15, 2016 3:33 pm
Location: UK

Post by cygmu » Thu Oct 12, 2017 10:02 am

groove wrote:
cygmu wrote: Do the other controls (CV in, 1V/Oct) affect the rise time?
No, nothing seems to change the rise time, which is very fast(haven't scoped it, yet).
OK, it sounds like something is amiss with the PNP differential pair. (Take a look at Tim Stinchcombe's page to see a clearer drawing of the schematic for this section:
http://www.timstinchcombe.co.uk/index.php?pge=vcs )

I would check the connections around there, make sure the components are correct and properly installed, and measure voltages at the bases of those transistors (and the other pins too, why not?) to get an idea of what is going on.

groove
Common Wiggler
Posts: 222
Joined: Fri Jul 19, 2013 10:21 pm
Location: NY
Contact:

Progress!

Post by groove » Thu Oct 12, 2017 12:28 pm

cygmu wrote: I would check the connections around there, make sure the components are correct and properly installed, and measure voltages at the bases of those transistors (and the other pins too, why not?) to get an idea of what is going on.
Thanks, it was helpful to have a better idea where to look. The Tim Stinchcombe page is really helpful, too. Now I have a working rise pot! I found a place where I'd tried to fix a trace but bridged to the wrong pin by mistake.

Now I have uncovered a slightly different issue with the other half of the board: the rise pot doesn't have the same depth as the fall, or the other rise. The rise time only lengths to about halfway as far as it should. Since I have one working perfectly, now, I should be able to compare and figure out what's happening.

groove
Common Wiggler
Posts: 222
Joined: Fri Jul 19, 2013 10:21 pm
Location: NY
Contact:

Fixed

Post by groove » Thu Oct 12, 2017 1:32 pm

I removed and replaced the 2N5089 on the other half of the board with a matched pair of 2N3904 and now my rise pot has the expected range. Very excited to get this working - it's definitely one of the more fragile boards I've tried to rework. Pads and traces lift really easily.

User avatar
tojpeters
Super Deluxe Wiggler
Posts: 1266
Joined: Mon Feb 06, 2012 10:34 am
Location: Mt.Shasta, Ca

Post by tojpeters » Thu Oct 12, 2017 6:05 pm

Yes they do

animalman
Common Wiggler
Posts: 57
Joined: Mon Mar 18, 2013 2:35 pm
Location: Back alley

Post by animalman » Sat Oct 21, 2017 1:38 pm

Got my stripboard working!! After having flaming hot transistors I swapped them to no help (and in the process learned super handy trick how to check pinout with my DMM).
After tracing the circuit for the umpteenth time I finally found that one of my stripboard CUTS passed signal. WTF.. :zombie: one churn with the drill bit and everything came to life.
Hang on slopy, slopy hang on :sb:

User avatar
cygmu
Super Deluxe Wiggler
Posts: 1501
Joined: Mon Feb 15, 2016 3:33 pm
Location: UK

Post by cygmu » Sat Oct 21, 2017 4:32 pm

animalman wrote:Got my stripboard working!! After having flaming hot transistors I swapped them to no help (and in the process learned super handy trick how to check pinout with my DMM).
After tracing the circuit for the umpteenth time I finally found that one of my stripboard CUTS passed signal. WTF.. :zombie: one churn with the drill bit and everything came to life.
Hang on slopy, slopy hang on :sb:
APPLAUSE!!!

User avatar
arrmcbain
Common Wiggler
Posts: 218
Joined: Sat Oct 02, 2010 7:02 am
Location: Adelaide, Australia

Re: DUSG Help!

Post by arrmcbain » Mon Feb 17, 2020 7:30 am

Well done to everyone who has managed to build a working CGS114! I had some issues too, which were thankfully just down to me being daft and missing the one of the wire links in the bottom half of the board. Funny thing was I did the same thing on both the boards I built for a quad slope on a Serge panel. I just could not understand why the bottom half wasn’t working on both boards — when I pitched the output to the trigger in, the LEDs would either light up with not much happening on the scope, or just not cycle at all. So if anyone has those symptoms I would recommend checking that you didn’t miss the small links that can sometimes be hard to see between the resistors.

Another thing with any CGS board is to actually check your transistor datasheets to make sure the pins match the diagrams on the PCB (wrt arrow btwn emitter and base). Even if you have the right part number you might be surprised how often the pins can vary.

And yes, any single-sided CGS board needs to be treated very gently — I spent literally years trying to troubleshoot a failed CGS29 build only to discover (by poking around with my finger) that it was a lifted pad.

User avatar
breadman
Common Wiggler
Posts: 138
Joined: Mon Jun 18, 2012 6:22 pm
Location: PDX yo
Contact:

Re: DUSG Help!

Post by breadman » Tue Mar 31, 2020 4:00 pm

Just got two DTGs fully working after a few hours of troubleshooting, thanks largely to this thread! They're working and quite stable all around. That definitely seems to be the major hurdle, and I'm very happy just to get this far. However, I'm noticing that all four slopes are on the slow side--maybe 2.4k max with no calibration and ~1.5k once they're set to 5Vpp. I stupidly used (matched) BC547B and BC557B for everything as per Ken's BOM instead of reading threads or listening to the build doc instead (would have used 3904/3906 if i had) and my theory is that the inferior power dissipation of the BC's is slowing the cycle. Does this sound reasonable to assume, or should I look elsewhere? One other note is that I'm using bright LEDs that draw about 20mA, but I'm not noticing any strange behavior related to the power rails or anything, and swapping the current limiting resistors dimmed the LEDs but didn't speed up the cycle.

If the transistors are in fact the culprit, I likely won't rip them out just for the sake of faster oscillation since again, they seem to work great otherwise.
Need a synth built? Get in touch!

email: jschnvr@gmail.com
IG: @jackssynthbazaar

User avatar
cygmu
Super Deluxe Wiggler
Posts: 1501
Joined: Mon Feb 15, 2016 3:33 pm
Location: UK

Re: DUSG Help!

Post by cygmu » Tue Mar 31, 2020 4:51 pm

breadman wrote:
Tue Mar 31, 2020 4:00 pm
However, I'm noticing that all four slopes are on the slow side--maybe 2.4k max with no calibration and ~1.5k once they're set to 5Vpp.
What value did you use for the integrating capacitor? 22nF on the original BOM. If you change this to 10nF everything will go 2.2x faster so you'll get 3.3kHz. It still seems a little slow -- mine maxes out at 5kHz with 10nF caps there. According to my understanding of the circuit, it shouldn't be the fault of the transistors, and I'd be pretty sure the LEDs are not the issue. If you used a value larger than 8k2 for the resistor at the emitters of the expo transistors, that would slow it down too.

I made the changes suggested in this post on all the DUSGs I built and I am pretty happy with them.
viewtopic.php?p=1561757#1561757

User avatar
breadman
Common Wiggler
Posts: 138
Joined: Mon Jun 18, 2012 6:22 pm
Location: PDX yo
Contact:

Re: DUSG Help!

Post by breadman » Tue Mar 31, 2020 8:20 pm

Exactly as you say, swapping in a 10n cap in that position has produced a new maximum of 3.3 kHz when calibrated. I also had used 82k going to the Rise/Fall pots so having swapped those with 68k, presto, it can still go very slow. (I already had 1% 8k2s at the emitter position.) I've still got a 1n cap going to A instead of 10n, plus my other resistor values are "stock." I wonder how much of the difference that might account for...

One thing I want to know is, how on earth was Ken getting 5k with the values on his site?

In any event, this was exactly the info I was after. Maxing out at 3.3k is entirely acceptable for my purposes (unless you think it's reason to worry or something) since these are just DTGs to be used alongside real oscillators. Thank you!
Need a synth built? Get in touch!

email: jschnvr@gmail.com
IG: @jackssynthbazaar

User avatar
Jonachi
Super Deluxe Wiggler
Posts: 1155
Joined: Mon Jul 22, 2013 5:19 pm
Location: Sweden
Contact:

Re: DUSG Help!

Post by Jonachi » Tue Jun 23, 2020 3:46 am

So loong story short. Used my corona isolation days to resteror and refine some old builds. Opened up the Quad Slop (sporting 2 cgs114 pcbäs) to fix one broken channel, clumsy mistake I made many years ago, swapped all leds, installed cyckling switches etc. When putting it back together I must have somehow damaged another channel on the other pcb what completly broke. First rise wasnt working, swapped all Ics and got it working again, then it broke again dieplaying the following behavior. Contantly cyckling )with or without switch or cables plugged in), output around 4v. The LF353 outputs 12v on leg 6 (which should be around 0,52v) comparing with a working channel, In the transistor mirror all emitters have the same voltage (0.16v) until the very last one (before the 353) which also outputs 12v (same as leg 6 on the 353).'

Ive swapped all trannsistors in the mirror to new ones changed ICs several times,check for continutiety between most parts and found nothing. Removed all cabeles and tested it without anything plugged in with (first) power and then then (secondly) with A patched to M to check for voltages, same results as abive (so not the wiring it seems). Before this malfuntion it had been working for at least 6-7years)

Let me know your thoughts.

User avatar
Jonachi
Super Deluxe Wiggler
Posts: 1155
Joined: Mon Jul 22, 2013 5:19 pm
Location: Sweden
Contact:

Re: DUSG Help!

Post by Jonachi » Tue Jun 23, 2020 8:48 am

Here is a picture
FE4C059A-EBEE-4610-8878-C755C3457A45.jpeg
You do not have the required permissions to view the files attached to this post.

User avatar
cygmu
Super Deluxe Wiggler
Posts: 1501
Joined: Mon Feb 15, 2016 3:33 pm
Location: UK

Re: DUSG Help!

Post by cygmu » Tue Jun 23, 2020 11:01 am

If you are running on +/-12V power, the 12V you are seeing in various places can only be a short to the power rail really.

The emitters of the four transistors in the current generation section -- I think that is what you mean by "mirror"? -- are supposed to be connected so it is impossible that there are no continuity errors and also a different voltage on one of those emitters.

I don't know for certain which op amp you are referring to when you say "the LF353" because there could be two, but the one marked LF353 on Ken's schematic, pin 6 is the signal input. So if that is reading 12V, what do you see at the input jack?

User avatar
Jonachi
Super Deluxe Wiggler
Posts: 1155
Joined: Mon Jul 22, 2013 5:19 pm
Location: Sweden
Contact:

Re: DUSG Help!

Post by Jonachi » Tue Jun 23, 2020 11:28 am

Input as in C on the board? Yes I use Ken’s schematic as refrence. So that’s the right Ic
Last edited by Jonachi on Tue Jun 23, 2020 11:48 am, edited 1 time in total.

User avatar
Jonachi
Super Deluxe Wiggler
Posts: 1155
Joined: Mon Jul 22, 2013 5:19 pm
Location: Sweden
Contact:

Re: DUSG Help!

Post by Jonachi » Tue Jun 23, 2020 11:34 am

Sorry messed up when counting the pins. Pin 6 is 0v and so is C and all resitors that are connected to it. Pin 7 however is the one that outputs 12v constantly.

Sorry for the confusion.

User avatar
cygmu
Super Deluxe Wiggler
Posts: 1501
Joined: Mon Feb 15, 2016 3:33 pm
Location: UK

Re: DUSG Help!

Post by cygmu » Tue Jun 23, 2020 5:56 pm

Is there a short from that pin to the +12 power supply? If not it seems very hard to see where 12V would come from there.

User avatar
Jonachi
Super Deluxe Wiggler
Posts: 1155
Joined: Mon Jul 22, 2013 5:19 pm
Location: Sweden
Contact:

Re: DUSG Help!

Post by Jonachi » Wed Jun 24, 2020 12:59 am

nope, no apprent short bwetween this leg and the +12v power supply at least.

User avatar
cygmu
Super Deluxe Wiggler
Posts: 1501
Joined: Mon Feb 15, 2016 3:33 pm
Location: UK

Re: DUSG Help!

Post by cygmu » Thu Jun 25, 2020 3:28 am

Well that's odd. When you say you measure 12V there, you really mean 12V and not 11.4V or something like that?

The other thing you wrote which was suspicious was that the emitters of the four transistors have different voltages. They are all connected together so that should be impossible. Can you double check that?

User avatar
Jonachi
Super Deluxe Wiggler
Posts: 1155
Joined: Mon Jul 22, 2013 5:19 pm
Location: Sweden
Contact:

Re: DUSG Help!

Post by Jonachi » Fri Jun 26, 2020 11:32 am

Found a CGS75 pcb and decided to build that one as a replacement (after spending days troubleshooting this one). Thanks for the help though, might have another look another time. Probably a broken trace or lifted pad unvisable for the human eye.

Post Reply

Return to “Music Tech DIY”