Random Crackling Pop Generator - Art MP S. Preamp almost fully restored

[reinw33]

This Preamp has been almost fully restored, but random crackling remains and Red LED flashing.
About 75% - 100% the time the red Limiter light turns on, because of internal loud crackling and pops and otherwise there's always some much quieter (I guess -25 dbfs) distorted oscilaltion there, perhaps 50hz.
But when you turn on the +20 dB tube boost, it 75% goes crazy, also randomly when +20 dB gain is off, on itself with Pops and crackling and the quieter hum noise you hear more when on.

The output voice sounds absolutely fine otherwise, so this should not be an opamp problem.

The snapshot of oscilloscope is the Line Output without any input attached. Same on Balanced Output.

Flow-Chart is something like Input Gain ---> Discrete Diff. Amp. ----> TL072 ----> Tube ---> NE5532 ---> Limiter ---> Output Gain. Please correct me IIAW.

Schematic is Attached

All of your help or your suggestions to help me restore this are much appreciated, if you have experience in restoring things like that, before I have to inject some voltage into it.

Thanks.

 

[shabtek]

bad connection/solder/broken copper trace or bad tube, have you tried a different/known quiet valve?

 

[reinw33]

bad connection/solder/broken copper trace or bad tube, have you tried a different/known quiet valve?

Hi CJ,

thank you for your reply.

So far I only heard the preamp from inside the DAW.

Today I soldered a few dozen joints.
Not all joints, but the flappy wings and ribbon cable joints.

What annoyed me was two things,
ROHS compliancy.... obviously this annoys many sane people in electronics, the solder joints just look worse and are worse in comparison.
But also what the manufacturer did annoyed me, even though the boards are through hole design, they just soldered the ribbon cable on the surface

It still turns on, but still flashes red because of noise.

I laso checked for shorts and continuity of some traces but not of all traces and measured some components voltage levels in circuit nothing that sticked out to me as unordinary except some voltage fluctuations that with just a digital multimeter, are not much. I wish I had an analog one too though,
I could just start replacing parts but I think I have a better idea.

I'm not gonna smack the preamp against the wall yet, as I it is my go to tool for learing audio atm, but I see no point using my multimeter or oscilloscope or injecting voltages for error tracing this preamp.

I need an audio signal tracer preamp, or something that I can drive or connect to my loudspeakers to and see where the errors stems from, does someone have or do you have any suggestions?

What do you use for this?

Is there a schematic for an error signal tracer?

I'm not gonna turn on my oscilloscope for things like that.

Thank you for all the help.

 

[NoisyIndividual]

I think the first steps should be removing the tube. Is it still noisy?

Then, does the output volume pot turn down the noise?

Removing the tube gives easy access to the first grid, you can easily scope that pin of the socket to see if the noise is there.
Or even use the DMM on that grid point. If the voltage jumps around, your noise is probably present.



Edit: except I forgot you said the noise is much worse with the pad switch lifted, so it is probably somewhere in the initial mic pre, not the tube onwards.

I would clean/check very carefully the gain pot and the wiring to it, and the electrolytic C19 attached to it, for some reason these caps are a point of failure in these types of circuits, creating monstrous noises.

 

[reinw33]

bad connection/solder/broken copper trace or bad tube, have you tried a different/known quiet valve?

CJ, yes, I did try it with another tube, a 5751, 70% gain of 12ax7, both of them really shouldn't go faulty simultaneously.

 

[reinw33]

I think the first steps should be removing the tube. Is it still noisy?

Then, does the output volume pot turn down the noise?
...

NoisyIndividual, you brought up some really insightful and great points that I'm gonna follow asap as time allows.

Thank you for your awesome reply!

Let me please do some quoting NI (Noisy Individual)
Your reply is totally appreciated and so spot on to solving this!

Yes, you can turn down the noise and signal at the output completly IIRC, and turning down input gain, does nothing to the noise IIRC, it keeps flashing RED at the clip/limiter LED.

Another possible source you can rule out is the cable and mic, both are toally fine, I checked with another preamp, so that is out of the way, it's just the preamp that is faulty.

Removing the tube gives easy access to the first grid, you can easily scope that pin of the socket to see if the noise is there.
Or even use the DMM on that grid point. If the voltage jumps around, your noise is probably present.

Yes this is basically the solution to what I suspected as possible noise origin, that the noise almost has to come from or before the stube stage.

Because why would the noise increase unless it was increased going into the +20 db switch tube stage, which runs the tube hotter into grid 1 and 2 and incerases signal with noise together.

Both tubes 5751 and 12ax7 shouldn't go bad simultaneously. Tested with both several times, both have the same noise behaviour.

Edit: except I forgot you said the noise is much worse with the pad switch lifted, so it is probably somewhere in the initial mic pre, not the tube onwards.

Those are great points NoisyIndividual,
Right now phantom power 48V is always on literally soldered to on.
Are the "first stages of the mic pre" before and leading up to Q2 and Q1 a voltage doubler with rectifier? I mean all those diodes and caps before those TIP112s amplifiers.
Do you suspect the noise coming from there?

If it's coming from that end or beginning and I unsolder 48V phantom to off position, it should stop the noise, shouldn't it?

If only I had done that first, I soldered so much today and unwound 2000 turns of transformer wire, and we have night.

But that phantom switch would really narrow down the source, as you can see perhaps in the dark image in OP,
in the picture from left to right SW2 is missing.
The push button link became electronically/damaged unstable and I don't want to destroy the attached mic by pulsing 48V phantom into it, so I made a "trace" joint by a wired joint soldered on for 48V phantom, it has 6 trough holes, I only soldered the pins the have contact when the buttons gets pushed in, so 4 pins have been contact with each other, I know how the push button looks and connects from the inside.
But will after having tried your suggestions first, remove it completely.

Because the clicky push button failed on reassembly, I managed to put this fiddy thing together, but humpty dumpty is not 100%.
so I discarded it for now and just bridged the joints of SW2 (i.e. Phantom Power SW2), they count the switches from left tot right, with solder and wire, so I will also have to unsolder that possibly, and see if the noise source is phantom power.
But it stayed at 46,5Vrespective to ground, the whole time steadily. But with a digital multimeter, and mine is a cheap 10€ one, you can't see sudden 100mv fluctuations that easily I believe, there could definitely be noise there. going in to phantom 48V.

I really learned something from you today by probing the grid with tube unmounted!
I will totally remove the tube and scope on Grid 1 when time allows, so far I only checked the cathodes LT sitting 6,7 V IIRC earlier today.
The +33V was doing sth. at 29,7V IIRC at Q1 in circuit and fluctuating a bit by like +-0.4V, but I vaguely remember this, but I 100% saw it on the digital multimeter.
I have an older siemens DMM that is more sensitive and has more resolution in the Voltages possibly that would work better error search/fluctuating voltages than a cheap 10€ modern DMM.

I would clean/check very carefully the gain pot and the wiring to it, and the electrolytic C19 attached to it, for some reason these caps are a point of failure in these types of circuits, creating monstrous noises.

Definitely, thank you!
Input gain pot has become a bit wobbly or less sturdy, it should be replaced probably.
Will resolder that too.

 

[NoisyIndividual]

Yes, you can turn down the noise and signal at the output completly IIRC, and turning down input gain, does nothing to the noise IIRC, it keeps flashing RED at the clip/limiter LED.

Well this indicates the noise is after the gain stage.

If you can easily get to it, try removing C23 and see if this stope the noise.
Pin7, the output of U1, should be at DC of around +16.5v. While C23 is out you can see if there is any noise from there, C23 is there to block this DC which may be leaking past it.

I also notice Q14, which seems to be a mute/limit FET J113. These are known to go noisy.

You could remove this too, the circuitry will still work as normal, it just won't limit, but this will rule the J113 out as a suspect.
Q13, another FET, may also be removed, just on the off chance it is faulty and affecting Q14, which it is directly connected to.

 

[NoisyIndividual]

The +33V was doing sth. at 29,7V IIRC at Q1 in circuit and fluctuating a bit by like +-0.4V, but I vaguely remember this, but I 100% saw it on the digital multimeter.
I have an older siemens DMM that is more sensitive and has more resolution in the Voltages possibly that would work better error search/fluctuating voltages than a cheap 10€ modern DMM.

I'm having a little trouble translating that, but the +33v should be totally stable as it supplies the op-amps.

Power supplies are the first things you should check, and make sure they are clean.

 

[reinw33]

I'm having a little trouble translating that, but the +33v should be totally stable as it supplies the op-amps.

Power supplies are the first things you should check, and make sure they are clean.

My bad. Your feedback is appreciated! Thank You NI!
Tomorrow is a national Holiday here "Fronleichnam". Excited to probe the pads for sure

 

[reinw33]

Here are the results and pics of todays investigation, I (try to) post in chronological order of time.

This is the scope on Grid 1, which is pin 2 on the tube (tube removed ofcourse) in respect to ground, e.g. Pin 9 which is the CT to ground.
No signal going in. Extremely noisy like 400mV.

I keep posting the results with a short delay between each post.

Today I wasn't able to do everything NoisyIndividual instructed me to, thanks for that!
Am still in recovery mode so to speak.
I keep posting shortly...

 

[reinw33]

After that result, which means it wasn't the tube itself, also the output stage with NE5532P can be ruled out, though I had to replace it way before in the beginning it was actually faulty along the lines of what noise remains present here.
I then immediately went on further and measured the emitter on Q1 which supplies the +48vcc

Also very noisy, like 300mV ac.
Stay tuned, some more results are coming.
EDIT: This is still grid 1, pin 2, just looking at 50mv scale instead of 0.2V in preavious, I confused this with Q1 which further down.

 

[reinw33]

Now in hindsight I realize I should have kept vlog notes along the way. It would have made it easier later on putting the pieces together and to report back to you NosyIndividual.

 

[reinw33]

Regardless of previous admission.
I proceeded to remove the SW2 bridged/soldered, no mic or line no signal going in to Grid 1 which is pin 2...

and then scoped it again!

We have narrowed down the source again.
This wasn't the most brightest or most helpful approach but out of principal of a working SW2, this is the ultimate SW2 switch, there's no Mic/Line going to the tube.

 

[reinw33]

This seems to be the emitter of Q1, still way too noisy of a power source.

Even though no signal goes to the output limiter, we have monstrous red flashing LED.
So I think we pretty much found the culprit somewhere in the 9v ac 0.4 amps supply side.
Still about to measure U1 as NI advised.

 

[reinw33]

....
Pin7, the output of U1, should be at DC of around +16.5v. While C23 is out you can see if there is any noise from there, C23 is there to block this DC which may be leaking past it.

...
NI, I was totally flabbergasted when the aka "pin 7" showed 1-4 volt-ish fluactuations on the DMM.

and I thought "what the heck, huh" - I took you by the word atm in time, and thought: "how in the heck did I hear anything out of this thing at all ?!"

it turned out, according to datasheet, pin 7 is Out 2 of dual opamp, pin 8 is DC supply. Duh.
DC/Vcc+ it's actually pin 8, but at the beginning I didn't realize that.
I missed to take a picture, but it showed +30.9V on the DMM on pin 8. For sure it showed +30V.
The datasheet says max +15V Vcc+
Do we have a problem here?
I can go back and confirm one more time.
EDIT: Don't mind the red croc. cable, I took what was at hand, red.

 

[reinw33]

this is some pin 7 action of U1 IIRC.

 

[reinw33]

Now to pin 8, it looks like 50hz ac is on pin 8. This is probably not supposed to be on Vcc+.

on the 5ms division you get every 4th cell a full cycle, so you got 50hz ac on pin 8, but it's rather low voltage.
Perhaps this is just induced noise on the probe?
What do you think NoisyIndividual?

 

[reinw33]

@NoisyIndividual,
will try to do all of your suggestions as best to my abillities, as time permits.
Perhaps even redo some scopes.
Thank you for your replies.
Good night everyone.

 

[NoisyIndividual]

Well if you have noise on the emitter of Q1 then that is your problem.
Or at least the biggest of them.

I would forget everything else for now, until you fix that issue, the rest may follow.

 

[NoisyIndividual]

Remove Q1 and scope the voltages on where its collector and base are connected.

Measure particularly the voltage and noise on the 47v Zener, including scoping its ground connection, in case you have some kind of scope reference issue.
You could have problems with the voltage doubler capacitors in this circuit, or indeed the Zener.