||904B control input 'bleed' through...
| br>Hi folks,
Question for any Wigglers here who either own a Moog 904B HPF or has built/cloned one.
On the two I built I am getting bleed through of the control inputs onto the audio. This results in the EG 'thumping' on the output quite badly and when I patch a sinewave into the control inputs I can hear it on the output almost as loud as patching it to the audio in..!
I'm having trouble getting rid of it and was wondering actually how much 'bleed' is acceptable or 'normal'..?
Apart from this they are actually working perfectly, I just seem to be struggling with this one thing.
Tom br> br>
| br>The first thing to check is whether or not you have any DC offset happening on the audio input. Ground the input (use a shorting plug or whatever method is convenient), put a slow LFO into the control input, and watch the output with a voltmeter. The output should remain at zero. If it tracks with the LFO, you have a DC offset somewhere on the input side of the circuit. There may be a calibration adjustment for it. br> br>
|cornutt wrote: |
|The first thing to check is whether or not you have any DC offset happening on the audio input. Ground the input (use a shorting plug or whatever method is convenient), put a slow LFO into the control input, and watch the output with a voltmeter. The output should remain at zero. If it tracks with the LFO, you have a DC offset somewhere on the input side of the circuit. There may be a calibration adjustment for it. |
Thanks, yes there is an output zero trim, but I cannot get the output to stay at 'zero'..!
I can trim it to zero no problem - but moving the fixed control voltage pot alters it. Looks like something isn't quite right...
Also the LFO does modulate on the output... br> br>
| br>I'm not familiar with the 904B but it could be that the two internal control voltages, +E and -E on the original Moog schematic, are not following each other properly. -E should be the same voltage as +E but inverted. If not, that could produce excessive CV breakthrough.
The whole filter is AC coupled bar the last stage, so perhaps a significant mismatch between Q25 and Q26 would also cause CV breakthrough. As to what you match Q25 and Q26 for I'm not sure. Vbe I would think - but that's not always that easy since one is NPN and the other PNP.
The circuitry around Q1 to Q10 could probably be reproduced with just a TL072 and some resistors for improved performance.
Tony br> br>
| br>Just an update guys, thanks for your suggestions - it all seems to be working now. A few points about my findings...
The notes on the schematic are misleading... it's not Q's 13/17/21/25 & Q's14/18/22/26 that need matching. The wording could be better there.
They actually need matching on the pairs which work together: 13/14 - 17/18 - 21/22 - 25/26.
Also C19 is wrong... it should go to ground and not the -6v rail. That seemed to be my main problem - when I moved it, everything cleared up..!
I also 'upgraded' the circuit with a slight addition.
I a/c coupled the audio amp and also put in a trimmer to clamp the base of Q27 to zero.
I think there was also an upgrade which added a couple of caps on the rear of the board around Q28, which I also performed.
So it's all working now... br> br>
Good job! br> br>
|noddyspuncture wrote: |
|So it's all working now... |