Hey,
I finally got it working!
ACTIONS:
I first did the measurements for the pot suggested by mnats. I had done similar measurements before, but I was hoping that I might find something interesting while doing it. They turned out to be all perfectly fine.
As mnats wrote in his answer and I suggested before too, the symptoms suggested that the voltage at point 7 was getting shunted. For that reason I again followed the yellow path on the diagram I posted. I checked all parts and connections on the bottom of the PCB. Since I had used flux cleaner before to clean of the board most of it looked fine. But of all parts, Q11 had some flux residue left. I cleaned that and realized that there might also be flux and residues on the top side. I checked and found a huge blob of flux stuck underneath Q11. I removed that and also carefully cleaned the whole top with flux remover and a little brush. Especially the whole path connected to 7.
Sure enough after I fired up the unit again it worked. So I guess the post in the FAQ was all the answer I needed after all.
I am a little embarrassed that I did not think about the flux residues on the top side of the board before.
RESULTS:
The warm-up time is reduced to about 5-10 minutes and the needle only moves about +1dB within this time (what is an average warm-up time?).
The release is now doing what it is supposed to do and all other functions of the unit are as expected.
Thank you all for helping me out with this issue!
__________________________________________
On an unrelated note:
While I was searching the internet for similar problems (instead of properly cleaning my board
), I found a bunch of suggestions to swap out some of the adjust-pots for larger values. e.g. R59, Q-Bias ADJ for a 5k or R71, Zero-Set for a 5k. Furthermore, I found a post which suggested that you could swap out R55, Attack for a larger value 50k-250k to be able to get longer release times. I am not really interested in doing any of these mods anymore. The only interesting thing might be to do the attack time mod under the condition that it does not change the original behavior of the unit.
Now, if these parts would just be rheostats (only two terminals A and W), I would agree with the swaps. But since they are pots, they have three terminals (A W B or 1 2 3) with the resistance between A and B always being the maximum value.
So lets say we switch out Q-Bias with a 5k pot. Since R59(A-B) is parallel with R60 and R35 is in series with them the resistance would change from (2k||3.9k)+10k to (5k||3.9k)+10k. Right?
Also and more interesting the attack-pot-mod. The resistance between point 19 and 7 is R55(A-B)||(R54+R55(A-W)). Therefore, when I swap out R55 from 25k to 50k, R55(A-B) would change and with it the total resistance.
Long story short: Which of the adjust pots and control pots can be switched out for larger values and why does or does it not affect the unit.
What I've read so far:
- ATTACK (R55): switch out attack for larger value to be able to have a longer attack time, (without affecting???) original characteristics (https://www.gearslutz.com/board/11855403-post12.html) - I would expect a change in characteristics
- Q-BIAS ADJ (R59): switched for 5k to solve adjustment issues (sorry, can't find the link right now) - I would expect a rather negligible change in characteristics
- TRACKING (R44): switch out for larger value, when running out of range while doing the Gain Reduction Meter Tracking Adjustment (http://groupdiy.com/index.php?topic=20058.msg461269#msg461269) - it makes sense why this mod works and does not change any characteristics (works like rheostat)
- DISTORTION (R85): seen people switch it to smaller value for more precision - personally I don't see the advantage, but I don't see a problem with changing it either to a smaller or larger value (works like rheostat)
- NULL-ADJ: switch for 5k (I think it was because the person did not have 2k pots)
- 0-Set: larger value for larger range - I think it influence the characteristics
If anyone has more information, please let me know.
