Dotcomformat

@Bill - The problem is so finicky, I wouldn't be surprised if unit-to-unit component variations make or break it. I hope the fix fixes it!

--- In dotcomformat@yahoogroups.com, Bill Yerazunis <yerazunis@...> wrote:
>
> Fascinating!
> 
> One of my Q109 envelopes exhibits this behavior, and the other doesn't.
> 
> One has a date code of 12-27-04 and doesn't stick on the attack, the other
> is 1-05-05 (both with inspector AA, no idea who *that* might be... :) ) 
> and does stick on the attack.
> 
> Moving the "long envelopes" jumper from one to the other had no effect.
> Nor did moving from one PS to another PS; behavior moved with the 
> PCB and nothing else.
> 
> And thanks for supplying the (easy) fix!
> 
>    - Bill
> 
> --- On Fri, 9/21/12, cobey_0 <cobeyii@...> wrote:
> 
> From: cobey_0 <cobeyii@...>
> Subject: [dotcomformat] Q109 envelope generator attack stickiness + fix
> To: dotcomformat@yahoogroups.com
> Date: Friday, September 21, 2012, 11:45 PM
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
>         
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> Hi all,
> I found an interesting problem with both of my Q109 envelope modules. It was discovered while demonstrating how an envelope generator works (of course!). Anyway, it's an obscure problem, but repeatable. With attack around 2-3, decay 0, sustain 0, release 0, the envelope will get stuck in the attack stage while gate is on. But only if using the gate input; the button works fine. Also, the problem will occur when the EG output connected to PWM input, for example, but go away if the EG output is connected to a 1V/oct input.
> This bugged me, so I did some debugging (hehe). I found that the output op-amp goes into high-frequency oscillation when the problem happens preventing threshold detector from triggering sustain stage. Lowering the op-amp's feedback impedance from 100K to ~10K pushed that pole up 10x thus killing the oscillation and allowing the circuit to work properly. If using the EG in normal 0-5V mode, placing a 10K resistor across R? (second resistor west of the TL-074, see picture) is all that needs to be done. However, if you're using +/-5V mode with jumper at J4, you'll also need to put a 10K across the 100K connected to J4. Or do it right and replace those two 100K resistors w/ 10K 1%.
> Picture of my quick fix: Q109 Oscillation Fix 
> I'm curious if anyone else encountered this problem, or is it just me?
> Cheers,Cobey
>