Crackle in Allen & Heath PA120 Mixer

[obiwanjacobi]

Hi,

I have a mixer that crackles. Even when all the channel's faders are off (lowest position) you still hear it.
(On the scope the signal makes small short jumps)

I tried to disconnect the channel cards from the bus one by one. I found one channel card that seems to be the source of the crackle. If I connect it to the bus the crackle re-appears. Any other combination of connected channel cards is ok as long as this card is not part of it. So I removed that channel card for some further investigation.

In the schematic (attached) on page 12, the channel circuit is shown. My first thoughts were that it may be a bad opamp.
I lifted on leg of both R54 and R56 (bottom left). It still crackled.
Then I lifted one leg of R55 and R57. It still crackled.
Then I lifted one leg of C8 (right side - last C before bus). It still crackled.



I have also checked for bad solder joints and carefully flexed the PCB to see if there was any change. No change, it still crackled...
Activating any switches or turning any pots, does not change anything. It still crackles.

I cannot think of anything else to try- probably because I do not understand the type of problems that cause this symptom.

Any help or suggestions would be welcome.

[Mick Bailey]

My first thought is to scope the supply rails to eliminate the possibility of any noise that may be caused by that particular channel card. If the supply rails aren't clean you won't be able to troubleshoot the rest of the circuit.

[obiwanjacobi]

My first thought is to scope the supply rails to eliminate the possibility of any noise that may be caused by that particular channel card. If the supply rails aren't clean you won't be able to troubleshoot the rest of the circuit.

The rails are probably affected -but I have not tested that yet.
Then again, how is this one card making that happen when I disconnect basically the entire circuit from the + and - rails by lifting those resistors?

[Enzo]

If fader to zero does not stop it, and I am assuming now that the pan pot pans the noise left to right. But just knowing the fader has no effect, I instantly suspect the op amp, U3. I'd be swapping that out at this point.

But I don't know how it can make noise with power removed from the IC. Did you lift BOTH 17VA resistors together? But still, we can verify this by simply removing U3 and powering the board, still noise?

[obiwanjacobi]

I am assuming now that the pan pot pans the noise left to right.

It does not. The crackling is in stereo but not affected by any of the pots or switches of this channel.

Did you lift BOTH 17VA resistors together?

Yes, both 10 Ohm pairs for +/-17VA and +/- 17VB are lifted - all four resistors R54, R56 and R55, R57.

The power supply, although not as clean as I would like it, does not seem affected.

I experimented some more and connected only the power +/- (pins 1 & 2 on the connector) and ground (pins 15 & 16). I then heard a loud hum. I then also added a connection to pin 3, some sort of virtual ground ..?
That made the hum go away. When I touch that pin - in other circuits call 'GC'- I hear the familiar hum of a human touching an input. I searched the rest of the circuits and found that this GC signal is sometimes used as a reference for opamps in the master section. That explains it.

Then I also connected the 48V phantom power and thought this was somehow causing the problem. It seemed to crackle when connected and fade when I disconnected it - like a cap discharging. I lifted a wire link (of the 48V) right next to the connector to make sure but that did not resolve the symptom.

My attention was brought back to the GC line, I suspect it is somehow involved in this.

So nothing conclusive, yet. It sure is a strange phenomenon.
I think I will be pulling all op-amps (only 3 8-pin packages) next, just to make sure these are not involved in some weird way.

[Enzo]

OK, intriguing. SInce it won't pan, can you see this noise on a scope at the pan pots or the AUX pots?

I don't spot GC on the drawing. Pin 3 is a ground, is zero ohm resistor R67 present on the board?

Lest we assume, what channel is this, and have we moved the card to a different position? I usually move noisy channel cards to the channel 1 position so I can flip it up on the ribbon end connector. Want to verify the issue is on the card rather than in teh connections to the ribbon on a particular channel. Is the "copper ground strip" a flexible item?

[olddawg]

A can of freeze is your friend... especially if you have a scope. Btw... does each channel have breaking jacks?

[obiwanjacobi]

can you see this noise on a scope at the pan pots or the AUX pots?

No. That is why I think the GC is involved. In the last experiment I did not even have these connected to the bus and you could still hear the crackle...

I don't spot GC on the drawing.

Search the PDF with text 'gc' and you'll find several in the stereo channels and master section.

Pin 3 is a ground, is zero ohm resistor R67 present on the board?

No it is not ground. The 0R0 in the schematic is in real life 10k! I did lift this one too, but it produced that awful hum. I did measure it then and it was (on my not too expensive DMM) 9.9xx k ohm, which I called good.

what channel is this, and have we moved the card to a different position? [..] Want to verify the issue is on the card rather than in the connections to the ribbon on a particular channel.

Yes, I did try different connectors on the flat-cable bus. I have the card out now and use the first (most removed from the connections to the master section) connector of ribbon cable, it is normally connected to the 6th (channel 6). When testing I have all other channels disconnected as well as the effects card. Only three master cards are connected and the secondary bus (card) remains connected over the entire master section (incl. effects card).

Is the "copper ground strip" a flexible item?

No it is a stiff copper strip with screw holes. I have it removed now, but bridge that connection on the channel card with a test lead with clips.

About Inserts: Yes there are inserts and I have checked them all (one of the first things I did). If it would be an insert problem on the channel, the crackling would be dependent on the signal path -and it is not- and if it would be a master insert issue, the crackling would always be present and it is not.

[Enzo]

No it is not ground. The 0R0 in the schematic is in real life 10k! I did lift this one too, but it produced that awful hum. I did measure it then and it was (on my not too expensive DMM) 9.9xx k ohm, which I called good.

But your schematic calls for zero ohms, not 10k, and if you look it even says "ground" pointing to pin 3 of the connector, which is connected to R67. Lifting that resistor would open the ground connection so I don't doubt it makes hum. Try shorting across it. Look at one of the other identical channels in that position, is that also 10K or is it zero? I see equivalents on other pages now, also marked zero, but the parts list does say 10k, so ther is a discrepancy.

So this card crackles regardless of which ribbon connector you plug it into?

Your GC is still some sort of ground reference, be it a jumper or 10k.

And I missed earlier that you lifted C8.

You say the pan pot doesn't pan it left/right out the mains, but do the mains faders bring it down? If so, is the noise on the two sends as well?

That doesn't make sense if we are assuming a signal path thing. It makes me suspect something in your grounding circuit. Not sure how the pin 3 circuit could do that, or if it is elsewhere, but there it is, food for thought.

[obiwanjacobi]

But your schematic calls for zero ohms, not 10k, and if you look it even says "ground" pointing to pin 3 of the connector, which is connected to R67. Lifting that resistor would open the ground connection so I don't doubt it makes hum. Try shorting across it. Look at one of the other identical channels in that position, is that also 10K or is it zero? I see equivalents on other pages now, also marked zero, but the parts list does say 10k, so ther is a discrepancy.

Yes, they're all 10k on all channel cards. I also noticed the difference earlier and checked. The more cards you plug in, the lower the resistance becomes, because they're all parallel to each other. I hope that is by design ;-)

So this card crackles regardless of which ribbon connector you plug it into?

Yes. Did not try all connectors but 3 or 4 at least. Makes no difference.

do the mains faders bring it down?

Yes. The main/master faders (L/R) bring down the crackling. It is "put on the bus" by this channel card so the

If so, is the noise on the two sends as well?

Not sure- I think I tested that. You go through so many steps and thoughts and theories that I sometimes lose track of what I did or did not do...

[obiwanjacobi]

I am back to suspecting the 48V phantom power to be the cause - or at least is somehow related.
There WAS a noticeable difference when the 48V was connected or when it was not.

I use those bread board wires with dupont connectors (male/female) to extend the bus wires and selectively connect them. I noticed toady that when pin 3 was connected (that virtual ground signal) that just putting my hand/finger close to the wires would induce a hum - so I figure that is what I heard earlier.

Anyway, I measured a resistance of around 400k between pin3 (virtual ground) and pin 4 (48V) on the connector on the bad card. On a good channel card my DMM does not come up with a value - it's open - as one would suspect. Each bus connector is 'enforced'/supported with a blob of some sort of glue over its legs and as I removed the final pieces of this goo I saw the resistance rise. When I could reach the legs I simply clipped all the legs and removed it entirely. I noticed the PCB does not take kindly to being worked on, I already lifted one pad - luckily a not connected pad- so I did not want to take any chances trying to get it out whole.

I have to order a new connector before I can try again (that will be over the weekend I'm afraid)...

I feel confident that I have found the problem but I'm sure Mr. Murphy will see to it that I'd be called a liar... :-)