Semi noisy carvin cx1272

[Sowhat]

I have a Carvin CX1272 mixer in for repair with channels 9 thru 12 being noisy. I figured it was something that is common to those 4 channels like a cracked solder or trace and that the circuitboard was grouped into 3 - 4 channel patterns but this is not the case. It looks like the channels are all discrete circuit patterns and not grouped into 4 channel patterns. I can find nothing physically common to those 4 channels outside of the common connections of Left, Right, Effects and Monitor which is common to all 12 channels. I have started signal tracing this noise with a scope and it is clearly visible when I increase the gain on those 4 channels, but I have not found a place where the noise drops out, not that I have done a lot to this end yet... it's a tightly clustered circuitboard with a lot of traces hidden under parts and a real pain to trace. Has anyone out there come across this or a similar problem and found the culprit?

[Enzo]

Link the schematic.

I don;t know how it is laid out, but it still could be something common. Are all 12 channels on one board? Or is there 8 channels on one board and four more on another? That is common enough when a factory makes both 8 and 12 channel versions of a mixer. Then again, it is likely the last four channels would be part of the master board.

Scope the power supply lines of the affected channels, make sure those are clean. SOmetimes groups of channels are isolated with resistors.

Isolate the problem. What controls affect the noise and which do not? You have a 20db pan button on each channel, does that affect it? The channel volume? Does pan flip the noise side to side? Tone controls affect tone of noise?

And it is entirely possible those four channels saw some action and were damaged, so you might just need four new op amps

Ah, I found the schematic on the Carvin site.
CarvinService.com - Schematic Finder

[Sowhat]

Link the schematic.

I don;t know how it is laid out, but it still could be something common. Are all 12 channels on one board? Or is there 8 channels on one board and four more on another? That is common enough when a factory makes both 8 and 12 channel versions of a mixer. Then again, it is likely the last four channels would be part of the master board.

Scope the power supply lines of the affected channels, make sure those are clean. SOmetimes groups of channels are isolated with resistors.

Isolate the problem. What controls affect the noise and which do not? You have a 20db pan button on each channel, does that affect it? The channel volume? Does pan flip the noise side to side? Tone controls affect tone of noise?

And it is entirely possible those four channels saw some action and were damaged, so you might just need four new op amps

Ah, I found the schematic on the Carvin site.
CarvinService.com - Schematic Finder

It's all one big circuitboard with 12 channels. You can see the identical 12 channel trace patterns all next to one another with a 4 trace buss running across them in the middle for I/O and +-15V/GND traces along the bottom... rather neatly laid out. It takes about 15 minutes to extract the board with all those controls on it but it can be hooked up to it's power supply and the stereo effects unit outside the case so I can get the whole thing running up on the workbench off my variable isolated AC mains supply with good access to both sides.

My current problem is that it's needed for rehearsal tonight and they can work with the 8 good channels with no problem so I've had to reassemble the whole thing again! This is the second time I've had to do this, the first time I got the effects unit functioning and cleaned out the cobwebs and insect cocoons lodged in the connectors. After tonight I get it back then disassemble it again and so on... phew!!!

I wish I could charge for all this time but this belongs to good friends and I'm doing them a favor because they are of course starving musicians who are perpetually broke... It's become kind of an academic challenge to me at this point.

I agree that it's possible that the 4 channels got smacked with the same failure from maybe plugging in some high drive device in one channel then it failed so with the logic of a musician they tried it in subsequent channels popping them until they finally gave up... I don't know at this point but checking the input stage op amps sounds like a smart thing to do right now because I don't really see any 4 channel groups of anything in there.

But that's for tomorrow's adventure, right now it's back together again and off to rehearsal. I'll keep you posted.

[Enzo]

I am not thinking so much a high level signal might have done it in. I am thinking more like some piece of gear with a grouind problem, had some voltage on it, got plugged in.

Do the isolation exercises I described. If the pad button makes the noise louder and softer, it is at or after the noise source. Same with ANY control that affects the sound in ANY way. Look at your schematic. The pad switch is right on the first stage. Look right after the first stage, the monitor send. Is the monitor send noisy too? That branch is before the tone stack. The tone stack appears to be an active stage using the second op amp. So if the gain pad button affects the noise, then that IC is suspect. If thsat has no effect, but the monitor feed is also noisy, that swtill points us at the first op amp. But if hte monitor feed is clean, but the tone stack affects the noise, then that second op amp stage is suspect. And if only the channel loevel control affects the noise, that localizes it to the third op amp stage. Yes, you could pick a channel and scope each stage, but these tests are just a matter of turning knobs on the panel, so it doesn;t even have to be disassembled for it. and it will tell you where the trouble lies.

Pick a channel and fix it. If they truly are all four individually noisy, then at least you might find a clue to the other three. On the other hand if there is some unforseen common element, then fixing one might cause the remaining three to fall into place.

[Sowhat]

If the pad button makes the noise louder and softer, it is at or after the noise source. Same with ANY control that affects the sound in ANY way. Look at your schematic. The pad switch is right on the first stage. Look right after the first stage, the monitor send. Is the monitor send noisy too? That branch is before the tone stack. The tone stack appears to be an active stage using the second op amp. So if the gain pad button affects the noise, then that IC is suspect. If thsat has no effect, but the monitor feed is also noisy, that swtill points us at the first op amp. But if hte monitor feed is clean, but the tone stack affects the noise, then that second op amp stage is suspect. And if only the channel loevel control affects the noise, that localizes it to the third op amp stage. Yes, you could pick a channel and scope each stage, but these tests are just a matter of turning knobs on the panel, so it doesn;t even have to be disassembled for it. and it will tell you where the trouble lies.

Rehearsal was called off so I had some time with it. Looks like the third stage is the source in all 4 channels per your diagnosis procedure. I actually tapped off the signal busses to take my readings for the Left, Right and Monitor signals once I found good access points. Only the channel gain effects the noise ( the gain button only makes it bigger). The monitor is clean as a whistle and the tone stack has no effect. That's a KA5532 which is both the first and third stages, the other amps are half of two MC4558's... bit of waste there. I only have one OPA2134 here and I want to keep that for backup on my personal amplifier... so it's off to the electronics store tomorrow to pick up some audio grade low noise op amps like NE5532's or heaven forbid NTE858's, I'll never find exact replacement KA5532's around here.

[Enzo]

If you have it apart, try some other op amp in there. Got some 4580 or even 4560? 2068? 5532 is a low noise part, yes, but you may find some more common parts as low enough in noise as not to matter. Try it on one channel and see. Might save a trip.

And I certainly would not sweat the difference between NE5532 and KA5532.

[Sowhat]

Well... got busy with some other jobs but finally got back to this Carvin, removed one of the suspect KA5532's and replaced it with an NE5532 and that cured that channels problem. It certainly looks like they blew out all 4 channels with the same problematic input device (whatever that was?)... probably some DI from some flakey amp or something. Now it's onto the other 3 channels. Carvin could not have made extracting those op amps any more difficult than it is, there's no room at all, it's the tightest packed area of the circuitboard... oh well, 1 down and 3 to go.

[Sowhat]

Well... got busy with some other jobs but finally got back to this Carvin, removed one of the suspect KA5532's and replaced it with an NE5532 and that cured that channels problem. It certainly looks like they blew out all 4 channels with the same problematic input device (whatever that was?)... probably some DI from some flakey amp or something. Now it's onto the other 3 channels. Carvin could not have made extracting those op amps any more difficult than it is, there's no room at all, it's the tightest packed area of the circuitboard... oh well, 1 down and 3 to go.

Hold my horses here!!! I told the guy at Marvac I wanted 4 NE5532's and he said sure, and handed me an envelope with 4 8 pin dips in there. I went back to the shop and dutifully installed them never bothering to read the part numbers because my "good seeing" day are long past. I ran the mixer through it's paces and everything worked and sounded great, I could not notice any difference by ear between the untouched channels and the repaired ones, everything was perfect. As a last step I examined my work with a magnifier to make sure no splatter or fragments were on the board before I reassembled things. In doing this I got a look at the part number on the op amps and they were not NE5532's, they were in fact MC1458's. Not what I was expecting and not what you would call an audio grade low noise op amp... high performance but not low noise! Anyway it seems the Carvin don't know the difference so I'll let it ride. Just a cautionary tale here to always double check your supplier, I have been dealing with this guy for years and he has never got my order wrong until now.

[Enzo]

And that lesson serves to put my suggestion in post #6 into context.

[km6xz]

If the only section that was noisy was the line out/channel gain amp, it seems odd that an external signal or input could impact that but not its outside looking mic pre section in the first section of the first IC. The mic pre will be noisier but probably not audible and especially not when used with a live band where even the noisiest mic amp would still be 40 db down from the venue noise floor. The noise trace showed a dominate 500hz sine wave, did you figure out where that was coming from. IC noise is not narrow band, more Gaussian


Next time you are into it, if you are concerned, change them to 5532 not due to the noise alone but due to the 5532 has a few specs that make it more suited to the front end application. If not, no one is going to notice or care. It is an excellent device for many audio applications if its input current is paid attention to. A 4580 would be cheaper and also be quite good for that application.