V5 White Horse + SBv3 oscillation / squealing sound

[Socrates_156]

 Hello everybody!

I just finished my white horse compressor, which doesn't seem to work very well... Well everything works, but with even the slightest mixed signal add, there is an oscillating squealing sound, which stops only when the Sustain pot is maxed out or the 500KA Drive is maxed out (along the way from min to max, the "oscillation"  changes). When I turn the pre-gain, again I can hear it changing frequency, but no matter where I set it, I can't get rid off it. Sometimes I face the same problem that is already mentioned in another post, the volume gets really loud when not touching the strings, then with the slightest touch the compressor kicks in and drops the volume.

I have checked all solders, component's values (used A10K for volume as mentioned on the V5 ), but I can't find anything really wrong. I checked the voltage on components according to the building documentation, mine are lower than suggested, but still closed to the expected. I honestly don't know what could be the cause, my suspects are a either a ground loop or a component among the NSL32R3, the TLE7024 and the charge pump , so any help suggesting where to begin or how to provide further details with would be deeply appreciated.
On the bypass switching board, the pull down resistor is located on the back temporarily, as I soldered 10K instead of 10M ( so for now I am using 2.2M but I don't see how this could be related to the audio problem). I also left out R5 & R6.

Thanks in advance for anyone kind enough to assist me!

[madbean]

All your parts seem correct on the PCB. I am assuming you have installed the NSL32R3 in the correct orientation (but make sure, of course) and you have also soldered in the bypass LED indicator (critical for correct operation in this particular case).

Slightly lower voltages can easily be explained if your PS has a lower nominal voltage than the one listed in the build doc (for reference, the One Spot I use for my testing rig is 9.42v with no load).

As far as noise floor, all optical compressor (actually pretty much any analog compressor) raises the noise floor some. There's not much getting around that since the basic idea is to reduce a high level of nominal gain with dynamics. So, I wouldn't say that's a red flag unless it's very noisy compared to plucked notes.

For oscillation, that generally points to issues with impedance or too much gain in the circuit itself. Maybe double check that you used 2M2 and not 2k2 for your pulldown resistor sub (the 2M2 value is fine). Other than that, I'm a little uncertain as to what the cause might be. Is there any voltage reading more than 1v off the listed values in the doc? Is ground readings at 0v (or no more than a couple mV?)

[Socrates_156]

Dear MadBean,

This is everything I checked / did :
- Tried different power supply
- Replaced the 2M2 with 4M7 just to be sure ,problem persists
- Soldered the LED, problem persists
- The dot on the NSL is matching the dot on the pcb

The input is 9,15V and after 1N5817 8.83 which after R2 1K delivers a max of 1.843V to the 3mm LED (used 1mH + 16V .5W Zener) -> Q4 (BS170) Gate, so I have 3.7V on Q4's drain and thus 3,7V on Q3's gate (from R19 100K) and 0V on Q3's Drain, and I always have 0V on both 2N5088's collectors!

- Tried 5mm LED, bright, then 5mm diffused, same thing!
- Switched all BS170, same thing!
I also attached an audio file. Around 8 sec, i simply touch the sustain pot and it's like beeping or something!
Any suggestions?

[jimilee]

In my experience, when you have voltages where you should not, you have a solder bridge. I would get a magnifier and go over it, along with reflowing your solder joints.

[Socrates_156]

 Thanks for your advice, checking for sorts and resoldering everything was the first thing I did. In this case however (well as far as I understand), the voltages should be there. According to the schematic the two BS170 work as switches, so with enough voltage on Q4's gate it should go to ON resulting in Q3 switching to ΟΝ and pass the voltage V+ to the 2N5088's collectors, but for this I guess 0V is required to Q4's gate, which I don't have (and this is why the voltage drop of the LED is so important in this particular build).

[Socrates_156]

Regarding the voltages, it was the LED after all (so anyone reading this post and having the same issues check your LED again)... switching it again (4 times), solved the BS170 problem and the voltages now look like a charm! But the oscillation issue persists, so I back to zero...  now for the weird stuff.. If a put multimeter on pin 14 on the tle2074 and I touch the lead with my finger, the voltage goes to 9V.. If I am not grounding anywhere, it get's up to 0.8V...