If it's the box I'm thinking it is, the Vactrols are for automixing, i.e. suppressing channels that are not the loudest.

What I would do is send a 1KHz sinewave at about 20mV from a generator to the mic input on channel 1 (panned centrally) and then chase through where it gets blocked on its way to the output. Use the working channel as a comparison.

Check the voltage across the diode pins on the pairs of Vactrols, comparing with the working channel as you go. The diodes should read something like 1.5V d.c. for a constant input level from the generator and then show a small variation for a step change in input level (indicating a larger current change). If there is a very small voltage (<700mV), then the automix controller is faulty. If there is a relatively large voltage (e.g. 10V), then there is a break in the diode chain, which could indicate a blown LED in one of the Vactrols.

That’s very interesting Boswell. Thanks for the reply.

Of course i tried so far with the 1khz sine signal but it’s pretty tricky to trace that pcb. 
It seems like the initial op amp nakes the balanced - unbalanced and somewhere towards the out it becones banced again. Many op amps, doingbthevsame thing.

There is also an LM 339 where strange things happen - i suspect that ic or around that area (that’s where the vactrols are too) and there are the ever suspicious pnp -npn transistors that i may have to remove for testing (same area too).

The op amp pins and voltages are great. 0v on all in-out pins stable -14v and +14v on - and + voltage sides

I think you can be a little more confident about what you see on the board. Leave aside the balanced-unbalanced conversion, and look at the unbalanced section between the conversion. However, before you do that, it's worth checking that the balancing amp on the non-working output is not the cause of the lack of signal there.

The LM339 is a quad comparator, and my guess is that it is used to compare the smoothed rectified output of a channel with that from the current active channel. If it is greater for a given length of time, then the current through the Vactrol diode for that channel is increased, lowering the resistance of the Vactrol's light-sensitive variable resistors for that channel and hence increasing its output. Conversely, the diode current of the previous channel is reduced, raising the resistance of that channel's Vactrol and hence reducing its output. The Vactrol resistors would be used, for example, as the signal inputs to a summing amp. It's possible to achieve in excess of four decades of resistance value change by varying the Vactrol diode current. Zero diode current results in an output resistance of many MegOhms.

If you are sure that the behaviour you are seeing is due to a fault, then a signal comparison between the faulty channel and the known working channel can be very informative.

As an aside, Vactrols are unfortunately no longer available in most countries as they contain very small amounts of the element cadmium. This falls foul of waste material regulations for end-of-life disposal. Go figure.

Thanks Boswell for the very informative advice.

i think i have to take it in a more step by step way once again i think. 
i will keep your advice in mind and inwillbtry to see if i will be able to figure out the vactrols workings.

Actually i’ve used photoresistor vactrols in other circuits ( vibrato, filter) but they cost for good things. The good thing is that the ones used in the preamp are cheaper ones, easy to find ones.

so i will first try the opamps, then the transistors, then the comparator and finally the vactrols. 
This is a link of the manual to see the block diagram. I suspect seriously the muting circuits.

https://storage.goo…

I've had a look at the manual and the block diagram you linked in your last post. I have to say that it does not appear to describe the model I was expecting from your initial description, i.e. a manual mixer with auto-mixer capability. However, the text says "each channel has its own volume control and 'VOICE' function to adjust speech signals to optimise intelligibility". This may well be the feature that I have been describing as auto-mixing, which would more accurately be auto-switching in this model.

In the block diagram, there are conventional pots shown, marked "volume". There are also wavy lines with a slanting arrow through them, which could well be the Vactrols. The block diagram shows no circuit block to control the diode currents through the Vactrols. Unless it's a part of the box marked "MUTE", the manufacturers may have regarded that as proprietary information and not included it.

I still think that the course of your investigation should be as I mentioned earlier, but with one initial difference. This is to see if there is signal at both the "REC" phono/RCA plugs, or only at one of them. If you find only one channel active here, there is no need to check the balanced driver output. Proceed from there backwards towards the inputs, making a note of the diode voltages on the Vactrols. I would expect a low voltage (<700mV d.c.) on the Vactrols attached to input channels that are not being fed with signal.

As i’ve told it’s way more complicated than a simple pre-amp because of the muting and the switching but i will follow your advice and hopefully i will come to a conclusion….how much easier things would be with a schematic

Well, I think the area you need to investigate can be shrunk by applying some constructive thinking.

The block diagram shows a MONO switch. If setting that switch to the mono position results in both outputs producing signal, then the whole of the output section feeding the XLRs must be working.

If the STEREO OUT phono/RCA sockets both produce signal, you really have narrowed down the fault area to not very much, possibly only the contacts on the MONO switch. The caveat here is that the stereo outs are shown as having Vactrol-set level controls. These are not involved in the signal flow to the main outputs, but unless the secret mute circuitry is allowing signal to flow to the stereo outs, they will show no signal.

There is an additional secret box labelled BASS HIGH MASTER. This is probably the tone control circuitry, and your fault may well be in one of the two routes through it (L or R). Try increasing the level of your mono signal through the mixer and see if both PEAK leds activate together.

Thanks for the advice Boswell.
i will try again as soon as possible - sometime next week- and i will keep you posted.

Bass high master is the tone control circuit and it seems to work well. 

Hi Boswell and everyone else. After some time I found the time to carry some tests on the preamp

Firstly the peak leds don’t work. Even if i feed them about 2v of signal.

Secondly all 3 vactrols without any signal fed show a value of around -400mV on the led+ pin. All other 3 pins show 0V.

My signal experiments results are as following.

 Using 20mv on the xlr inputs:

On the left xlr out side the signal seems capable of reaching very high output but the main out left potentiometer don’t work in either mono or stereo. All the attenuation can happen at the stages prior to that.

The same for the left rec RCA out

The stereo RCA out works well and the stereo output potentiometer works as it should.

On the right xlr out side the signal doesn’t seem capable of reaching very high output but the main out right potentiometer work in either mono or stereo. 

The same for the right rec RCA out

The stereo RCA out works well and the stere output potentiometer works as it should.

Mono-stereo doesn’t affect things at all.

Rca line inputs using about 100mV:

On the left xlr out side the signal seems capable of reaching very high output but the main out left potentiometer don’t work.

The mono / stereo switch attenuates the signal (about 50%) when pressed on the stereo position.

All the attenuation can happen at the stages prior to that.

The same for the left rec RCA out

The left stereo RCA out shows very low signal (way lower than the input one) and the stereo output potentiometer works.

On the right xlr out side the signal doesn’t seem capable of reaching very high output but the main out right potentiometer work in either mono or stereo. 

The same for the right rec RCA out The output potentiometer works here as well.

The right stereo RCA out shows very low signal (way lower than the input one) and the right stereo output potentiometer works.

Please take your time to think what these results tell you and let me know if you have any idea.

Kind Regards

ps. I ordered vactrols from aliexpress, i will also buy some transistors. I will change the comparator. I think the problem must be around there. A lot of conflict

Comparator changed. Nothing changed. 

Will try to change all 9 transistors soon.

Discovered that both in+ pins of half of the op amps are shorted to ground. 

I guess that's not normal and they are the faulty ones. And it means that there is a switching problem.

I'm having a bit of difficulty understanding exactly how you tested this box, but it does sound as though a 20mV signal into channel 1 arrives at both of the Stereo Out RCA jacks, with working amplitude control. That's good news, as it means that there is a path from at least one front panel input as far as that point.

You also say that the signal can pass through to the left XLR  output, but at a high level without any amplitude control, and also to the right XLR output, but with very reduced amplitude. So there is an imbalance that needs investigation.

You should make further measurements to deduce what is working and what isn't before randomly changing components. Don't change any op-amps yet. It's very likely that the +ve inputs to some of the op-amps are intentionally grounded, as current summing is done at the -ve input.

You say that your measurement of the Vactrol diode voltages were done with no input to the unit, which means those measurements are not going to tell us much. You need to re-do the Vactrol LED diode measurements, this time with the input signal applied that gives the high level at the left XLR output.

Firstly, measure the change in LED diode voltage on either the L or R Vactrol associated with changes of the level control on the (working) RCA Stereo Out. Make a note of the LED voltages at zero level, full level and few in-between settings of the Stereo Out level control. Note that the observed d.c. voltage change over the full range of levels will be quite small (maybe 100mV), as it is the LED current that defines the change. The measurements will give you an idea of the sort of static voltage and changes in voltage you would be looking for at other Vactrol diodes.

Now look at the voltages on the L and R Vactrol LEDs of the main output level (rec RCAs and XLR) controls. My guess is that the L one will be at the high end of the range and the R one near the low end. If that is the case, it means you probably have a fault in the LED current control circuitry. Unfortunately, this does not appear on the block diagram.

I wouldn't worry too much at the moment that there is no indication on the peak level indicators. It's possible that the lack of indication is related to the main output's gain fault.

Come back here when you have got to this stage, and we'll decide what to do next. A good-resolution photo of the Vactrol control area of the board would be useful.

Incidentally, did you manage to find a good source of Vactrol replacements?

Thanks for the reply Boswell.

Yes the source of vactrols is good. They are considered the cheapest vactrols VTL5C2. Simple things. 
Nothing like analog effect unit vactroks.
I was never disappointed with aliexpress components. At the end of the day they are all made in china. It just takes time to deliver.

The thing is that i can’t really tell which of the 3 vactrols does what because of the messy pcb.

I will try all 3.

Just to clarify a bit further:

When i say xlr input - it applies to all 4 xlr inputs. Not just one.

And when i say rca input - it applies to all 4 pairs of rca inputs

i will carry the tests later tonight and i will also attach a couple of photos.

cheers

If i didn't do anything wrong the vactrols led sides show 0v whatever the level is and whatever the input used. 

The stereo rca worked as it should.

The opposite side from the vactrol led side on just one pin showed 2.5mv for one vactrol and -4.1mv for the other. 

here are the pictures of the unit and the pcb.

https://ibb.co/pfkp…
https://ibb.co/fMJb…
https://ibb.co/Hn3J…
https://ibb.co/7Qcr…
https://ibb.co/wQyD…
https://ibb.co/xHzd…
https://ibb.co/HVw1…

Hopefully they could clarify things a bit

Excellent photos, thanks. It's nicely-made unit.  However, I could see only 3 Vactrols in your 7 photos. Are there more out of shot?

One point of interest is that your photos show the Master R peak front panel LED is lit. Your description seemed to say that there was only a weak signal on the R channel output. I can't see any connectors plugged into the inputs in your photos, and hence there should be no signal going through the unit, so it's possible that there is a large d.c. offset in the R channel.

In your last post, you said that you never saw any voltage across the Vactrol LED terminals. That's at odds with the Master L output producing a large amplitude signal in your tests, as it would need to have full current through its LED to achieve that. It may point to there being a second fault in a different part of the circuit.

I thank you Boswell for the help

No there are no more vactrols. Just 3. Inguess for switching.

No led is really lit. Just that one permanently very dimly lit.

I took all connectors out before the photo.

If you have the time please read a bit more carefully what i’ve written before. I ‘ve mentioned everything about 3 vactrols, no led etc etc. I don’t know if my writing makes any sense to you or if it really matters, of course :-)

i guess it needs a lot of components change. Luckily enough o have good op amp tester. I bet it’s either vactrols, transistors or op amps. There isn’t much else in there that could h]go wrong since the power supply works.

It's beginning to make a bit more sense now. You've confirmed that the Vactrol count is only 3. This would correspond to one in the microphone 2,3,4 mixer path (in the small MUTE box in the block diagram) and one each for the L and R channels in the stereo inputs (in the larger MUTE box).

If this is indeed the case, then the lack of any LED volts on the Vactrols when a full-level signal is sent through the unit from the microphone 1 input would be correct. It points to there being faults in the R master mix output and also the PEAK LED circuitry. What is harder to explain is how the stereo out RCA jacks can output equal amplitude levels on their L and R channels when there is some permanent illumination (a d.c. level) on the R PEAK LED. However, there are many coupling capacitors visible on the board, so it's likely that a separate d.c. fault in the R channel round the peak LED area would not affect (a.c.) audio being sent to the output amplifiers.

If we concentrate on the difference between the L and R main outputs, you said earlier that the L output was permanently at a high amplitude level and that the R output was smaller but controllable using the R output level control. One explanation for this could be a broken (dry) joint in the ground connection to the L output potentiometer, either at the circuit board solder side or inside the L pot body itself.

Could you re-state the effect on the outputs of changing the position of the STEREO/MONO switch? This would be when there is a relatively large signal going through from Microphone 1 input to the main outputs.

Ok. Once agdin where we go.

For xlr in the stereo RCA out works well and the stereo output potentiometer works as it should

left xlr out high -  right xlr out low
Mono- stereo switch has no effect

Rca in

Xlr outs exactly the same behaviour like the in one of the xlr in.

The mono / stereo switch attenuates the signal (about 50%) when pressed on the stereo position.

The left and right stereo RCA out shows very low signal (way lower than the input one) and the stereo output potentiometer works.

ps. Pot grounding works just fine and all pots (and the pcb itself) are in great (like new) state. 

Let's take the case of a test signal fed into chan 1 XLR input. This mono input is fed equally to L and R channels at the summing nodes. The XLR outputs have L out at a high level and R out at a low level (but not zero). As I understood your previous posts, the REC RCAs mirrored the XLRs, i.e. L high and R low, but the STEREO OUT RCAs were the same level. The crucial observation here is that the mono/stereo switch makes no difference to the output levels. That means that the switch is receiving equal levels on L and R.

This narrows the problem area down to the L and R amplifiers associated with the master volume control. The block diagram shows an amplifier, maybe connected as a buffer, with a variable resistor floating above it. If this is implemented as a conventional potentiometer to ground followed by a buffer amp that drives the RCA outs and also the amplifiers that form the differential XLR output, then the problem is in the buffer amp or the potentiometer itself.

What I would try here is to connect the L and R REC outs together using an RCA-RCA cable, and observe what happens at the XLR outs. It could be that there are isolating resistors feeding the RCA outs, in which case, it would not cause much change in the XLR outputs, but if the L and R XLRs suddenly have the same output level (high or low), then you have confirmed that the trouble is in the individual output level pots with their associated amplifiers.

Until you have the output problem fixed, there is no need to be testing the stereo RCA inputs. My guess is also that the STEREO OUT connectors are intentionally at a low level (-10dBV versus +4dBu at the XLR outs).

I wasn't sure what you meant by "Pot grounding works just fine".

So you encourage me to create a loop between the 2 rec outs.

Wouldn’t that cause any damage because of feedback?

By pot grounding I mean that the ground on the potentiometer pin that goe to ground is properly connected and working as it should.

The +4-10db does make sense that’s why i consider rca as line outs- line ins.

It's just one of the sort of things that test engineers automatically think of to narrow down a fault. Don't do it if you are not confident that there is no loop and therefore there can be no feedback.

I suppose you would also baulk at another thing I would do, which is setting the L and R main output volume controls to mid-position and interconnecting the two wipers.

At least do the mid-position setting of the two controls and separately measure the a.c. level at each wiper. It will be in the tens of millivolts when you input your 20mV sinewave to channel 1 with the input level turned up.

What we are trying to achieve here is to work out whether the problem is the potentiometers or the buffer amps. Going beyond this gets more difficult without an oscilloscope.

Ok Boswell.

Don't worry about that. I'm not that sensitive with equipment. I will try the loop.

I also have to remind you that I'm using an oscilloscope. Not a modern one. A 70s one. But for signal tracing it serves me well.
 

Ps. Tried the loop trick. No difference on the xlr outs. Exactly the same results.

Good news.

While i was replacing the transistors (and just while i was replacing the last one) i realized that the left output pot was broken. It was cut but in a way that it was not obvious at all. You had to remove allthe front panel to realize that.

So i guess now it’s a matter of a new b50k mono pot. 
So now the preamp will have new transistors, a new comparator and a new pot.
I will replace the pot on Monday and i will let you know. I guess it will be good news.

Actually you had guessed it, that there may be something INSIDE the pot.

Thanks for the help so far Boswell.

It's good that you found a problem. It's possible that getting the left channel to be balanced in level with the right may unearth other defects that were being masked by the pot problem.

I'm surprised by your saying that the faulty pot is a B50K, as that would be lin law. Check what the right channel pot says on it (if you can see it) before ordering.

Strangely enough all master pots are 50k linear (the stereo ones too and the other “master” volume ones on the stages before. Only the mic pre amp ones are 5k log.

Unfortunately i bought one, i asked for a 50k linear one but the guy behind the counter put a 2.2k linear in the bag.
Waste of time. Always useful have another pot around for other possible uses.

I’m sure that by changing the pot i won’t sort out the issue of controlling the master out.
Since both master pots were not functioning. 
But at least it may balance the overall output level on each channel. Hopefully.

Bad news. Nothing really changed with the potentiometer change. It doen’t affect the function at all. Same state as before.

Comparator changed, all transistors changes, broken pot changed. Voltages tested. 
So now it’s the vactrols and after that testing of all op amps from the output backwards.

what else?

Can you operate the unit while you have access to the solder side (underneath) of the board? What you should check is that the track from the low side of the pot is continuous to signal ground, i.e. no breaks or cracks in the copper. Looking with a scope at the three pot terminals should show full signal at the pot high, a variable amplitude signal at the wiper connection and no signal at the low end.

I would strongly advise you not to change any Vactrols until you have sufficient evidence that one or more of them is faulty. From what you have said so far, the evidence is not there.

Success. Yes it was the potentiometer. But because i was testing without the chasis being firmly placed, some of the input pot pins were touching it/. When i took it apart for testing everything worked as it should.

Then i insulated with tape nost sensitive parts, as i always do on my projects and bingo.

I think it was a matter of broken potentiometer right from the start (and maybe because of a serious fall someday in its life) the front panel had moved a bit and it was causing trouble with the pot pins. 
I gave it a bit of new life with the transistors etc. And of course I wouldn’t have achieved anything without your guidance and advice Boswel.

Really grateful

Hopefully the good state it is will last. I think it will.

Thanks again. 

Yes, it would be useful additional information to have a comparison between right and left a.c. pot voltages when a sinewave is put through the unit via channel 1 microphone input.

I should have explained in my past post that solder joint cracks are the most common faults on board-mounted pots, as they often have to take significant shear force shocks.

I recently got a far-eastern made amplifier in for inspection and official report. It was brand new, but had not worked from initial unpacking, despite it and its box being covered in "Tested" stickers. On inspection, there were many solder joints that simply were so badly done that it could never have worked, let alone pass any sort of production test. The supplier had refused to accept that it was faulty, even after my report. He cited the Tested stickers, so would not take it back. I fixed it - user happy.