CGS48 VCO Problems

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

Moderators: Kent, luketeaford, lisa, Joe.

User avatar
srfnsmy
Common Wiggler
Posts: 108
Joined: Thu Jan 22, 2015 5:10 pm
Location: Los Angeles

Post by srfnsmy » Wed Mar 30, 2016 6:50 pm

guest wrote:does the trim pot do anything? it should be able to get a good sinewave at this point if you turn the amplitude down. does turning the trimpot all the way down make the sinewave disappear?
yes turning it down makes it disappear, but it is again mutating back to the original form (the sine pictured in my last post) once i stop trimming with amplitude up past half way.. :goo:

User avatar
srfnsmy
Common Wiggler
Posts: 108
Joined: Thu Jan 22, 2015 5:10 pm
Location: Los Angeles

Post by srfnsmy » Wed Mar 30, 2016 7:02 pm

tojpeters wrote:Are you using a good quality name brand trimmer?
I'm using these that cost $3.14 a pop, yet I've always thought they feel unstable and cheap!

http://www.mouser.com/ProductDetail/Bou ... H-1-253LF/

:confused:

User avatar
guest
Super Deluxe Wiggler
Posts: 4512
Joined: Mon Aug 19, 2013 11:49 am

Post by guest » Wed Mar 30, 2016 9:34 pm

how does the amplitude of the output change as you turn the trimmer. does it stay about the same, and then just disappear when you get to the end? or does it slowly decrease? is the original video you posted still about the same, except without the large dip in the middle?

also, the ideal point is probably with the trimmer very near the completely off position, so move slowly about that point, and see how it responds. from your video, it looked like it took a few seconds to settle out, so be sure to give it that amount of time after adjustments.
openmusiclabs.com

User avatar
srfnsmy
Common Wiggler
Posts: 108
Joined: Thu Jan 22, 2015 5:10 pm
Location: Los Angeles

Post by srfnsmy » Wed Mar 30, 2016 10:23 pm

guest wrote:how does the amplitude of the output change as you turn the trimmer. does it stay about the same, and then just disappear when you get to the end? or does it slowly decrease? is the original video you posted still about the same, except without the large dip in the middle?

also, the ideal point is probably with the trimmer very near the completely off position, so move slowly about that point, and see how it responds. from your video, it looked like it took a few seconds to settle out, so be sure to give it that amount of time after adjustments.
Wow! That is looking more like it!

So in the vid i first trim from off all the way to full. then back to off and bring it in slightly.. you were right - very close to the off position is best!


User avatar
guest
Super Deluxe Wiggler
Posts: 4512
Joined: Mon Aug 19, 2013 11:49 am

Post by guest » Wed Mar 30, 2016 11:16 pm

cool, is it all working as you want now?
openmusiclabs.com

User avatar
srfnsmy
Common Wiggler
Posts: 108
Joined: Thu Jan 22, 2015 5:10 pm
Location: Los Angeles

Post by srfnsmy » Thu Mar 31, 2016 12:31 pm

guest wrote:cool, is it all working as you want now?
Yeah thats it! :sb:

Sub and Staircase are working too.

Really have to thank you Guest, tojpeters, yan6, and BugBrand for reaching out and helping me on this. The whole time I was thinking it was a lifted pad or cracked trace, but it was just a simple adjustment!

:party: :party: :party: :party:

I do have a question about the 2 PWM inputs on the pcb though:

With the wiring diagram I have, I've wired it so its the CV jack is going to the pot and then to the first PWM input. The second PWM input isn't being used.

There is also a second way for this knob to be wired so that it controls the PWM and the CW and CCW are connected to +/-V.

Is there a way to have the pot controlling the PW at the first PWM input and have the CV jack going directly (without a pot) to the second PWM?

User avatar
guest
Super Deluxe Wiggler
Posts: 4512
Joined: Mon Aug 19, 2013 11:49 am

Post by guest » Thu Mar 31, 2016 2:52 pm

yes, you can just wire the pot and jack to the two inputs. they resistor mix via 100k resistors on the board, so the pwm level you get will be a sum of whatever the knob is at, and what the input CV is at.
openmusiclabs.com

User avatar
srfnsmy
Common Wiggler
Posts: 108
Joined: Thu Jan 22, 2015 5:10 pm
Location: Los Angeles

Post by srfnsmy » Thu Mar 31, 2016 5:13 pm

guest wrote:yes, you can just wire the pot and jack to the two inputs. they resistor mix via 100k resistors on the board, so the pwm level you get will be a sum of whatever the knob is at, and what the input CV is at.
Great! Thanks again Guest :banana:

Now for the VC Divinder :hihi:

User avatar
finlayshakespeare
Learning to Wiggle
Posts: 30
Joined: Fri Jan 24, 2020 3:42 am
Location: Bristol, UK
Contact:

Re: CGS48 VCO Problems

Post by finlayshakespeare » Mon May 04, 2020 2:57 pm

Hopefully not hijacking this, but had some really odd things going on with my builds of the CGS48. Caveats:

-I'm running them off a +/-12V supply, but had made the 20k->14k tweak at the 5V test point node.

-I've used two 2N3904s as the matched transistors.

Otherwise, all is as standard. The problem I ran into was that both VCOs just ran at far too high frequencies. Even changing the timing cap to higher capacitance values (to the point where it was becoming a bad idea), I couldn't get a sensible range out of it.

If you encounter this, here's what I changed:

-56k between pin7 of IC1 and the matched transistor pair was replaced with 30k.

-The Span trimmer was replaced with 100k and the 100k feedback resistor was then changed to a 10k resistor, essentially giving the trimmer a -far- bigger range.

I also replaced the 100pF cap at the Sync input with a 10nF cap, which changes the behaviour from a light phase locking effect to a really far more pronounced sync effect.

User avatar
Bodo1967
Veteran Wiggler
Posts: 655
Joined: Wed Aug 10, 2016 2:27 pm
Location: Germany

Re: CGS48 VCO Problems

Post by Bodo1967 » Mon Jul 06, 2020 5:26 am

finlayshakespeare wrote:
Mon May 04, 2020 2:57 pm
If you encounter this, here's what I changed:

-56k between pin7 of IC1 and the matched transistor pair was replaced with 30k.

-The Span trimmer was replaced with 100k and the 100k feedback resistor was then changed to a 10k resistor, essentially giving the trimmer a -far- bigger range.
Thanks for that info!

Mine (v 1.0) also was impossible to calibrate properly to 1V/oct. I also replaced the 56k with 30k according to your recommendation, and I finally ended up with changing the feedback resistor to 10k as well. But I actually had to use a 200k trimmer: The original configuration gave me about a fifth (C - G) per 1V; 10k resistor and 100k trimmer about a seventh (C - a bit below B), until using a 200k trimmer finally allowed me for a nice octave.
... why buy it for $100 when you can build it yourself for $150?

ZTX500
Learning to Wiggle
Posts: 45
Joined: Sun Jun 21, 2020 3:27 am

Re: CGS48 VCO Problems

Post by ZTX500 » Mon Jul 06, 2020 6:43 am

finlayshakespeare wrote:
Mon May 04, 2020 2:57 pm
If you encounter this, here's what I changed:

-56k between pin7 of IC1 and the matched transistor pair was replaced with 30k.

-The Span trimmer was replaced with 100k and the 100k feedback resistor was then changed to a 10k resistor, essentially giving the trimmer a -far- bigger range.
Bodo1967 wrote:
Mon Jul 06, 2020 5:26 am
Mine (v 1.0) also was impossible to calibrate properly to 1V/oct. I also replaced the 56k with 30k according to your recommendation, and I finally ended up with changing the feedback resistor to 10k as well. But I actually had to use a 200k trimmer: The original configuration gave me about a fifth (C - G) per 1V; 10k resistor and 100k trimmer about a seventh (C - a bit below B), until using a 200k trimmer finally allowed me for a nice octave.
Any idea why this change was needed? The theory is that the voltage at the transistor base (i.e. voltage across the tempco resistor) should change by -17.8mV per octave at 25°C, independent of the type of transistor is used. This should be what the 56k resistor, 1k tempco and 10k trimmer provide (the trimmer should allow the range -17.5mV to -19.2mV per octave).

With a 30k resistor and 200k trimmer, it sounds as though the tempco being used may be closer to 500R than 1k, at 25°C?

User avatar
Bodo1967
Veteran Wiggler
Posts: 655
Joined: Wed Aug 10, 2016 2:27 pm
Location: Germany

Re: CGS48 VCO Problems

Post by Bodo1967 » Mon Jul 06, 2020 7:07 am

ZTX500 wrote:
Mon Jul 06, 2020 6:43 am
Any idea why this change was needed?
Nope :despair: . I'm glad I finally got it working (sub oscillator works as well).
Mine also has that spike or offset in the sine wave, but I decided I'll just call it a feature ;).
... why buy it for $100 when you can build it yourself for $150?

Gandalf
Common Wiggler
Posts: 137
Joined: Thu Nov 29, 2018 8:59 pm
Location: Australia

Re:

Post by Gandalf » Mon Jul 06, 2020 8:35 pm

BugBrand wrote:
Wed Mar 30, 2016 12:21 pm
BugBrand wrote:That was key to all downwind shapers - changing it brought the saw to be correctly centred around 0V.
Actually I may well be wrong about the sine & TSO sections needing the saw to be balanced around 0V - they have input caps and use the LM3900 (which I never got my head around)

I still think you should have the 240k changed to bring the DC saw down to centre around 0V.
The 240K shoul dnot be changed. The output of the TL072 here should be swinging from 0V to +5VDC and so creates the SAW DC output. That goes through the 47uF to create the SAW AC output. The SAW DC signal is fed to the SINE Shaper circuitry. if you change the 240K to give you a 0V centered output at SAW DC then the shaper won't work properly - it is SUPPOSED to have a 0V/+5V input signal.
The Staircase Generator MUST have the link between SQ and STI fitted.
With the recommended 1K TEMPCO fitted, the56K resistor is correct although the 100K in the feedback (connected to SPAN) should be about 95K(ish)

User avatar
finlayshakespeare
Learning to Wiggle
Posts: 30
Joined: Fri Jan 24, 2020 3:42 am
Location: Bristol, UK
Contact:

Re: CGS48 VCO Problems

Post by finlayshakespeare » Mon Sep 14, 2020 11:35 am

ZTX500 wrote:
Mon Jul 06, 2020 6:43 am
finlayshakespeare wrote:
Mon May 04, 2020 2:57 pm
If you encounter this, here's what I changed:

-56k between pin7 of IC1 and the matched transistor pair was replaced with 30k.

-The Span trimmer was replaced with 100k and the 100k feedback resistor was then changed to a 10k resistor, essentially giving the trimmer a -far- bigger range.
Bodo1967 wrote:
Mon Jul 06, 2020 5:26 am
Mine (v 1.0) also was impossible to calibrate properly to 1V/oct. I also replaced the 56k with 30k according to your recommendation, and I finally ended up with changing the feedback resistor to 10k as well. But I actually had to use a 200k trimmer: The original configuration gave me about a fifth (C - G) per 1V; 10k resistor and 100k trimmer about a seventh (C - a bit below B), until using a 200k trimmer finally allowed me for a nice octave.
Any idea why this change was needed? The theory is that the voltage at the transistor base (i.e. voltage across the tempco resistor) should change by -17.8mV per octave at 25°C, independent of the type of transistor is used. This should be what the 56k resistor, 1k tempco and 10k trimmer provide (the trimmer should allow the range -17.5mV to -19.2mV per octave).

With a 30k resistor and 200k trimmer, it sounds as though the tempco being used may be closer to 500R than 1k, at 25°C?
No idea whatsoever - the tempcos were genuine parts from Thonk too, of the correct value and spec.

Post Reply

Return to “Music Tech DIY”