Manley Vari-Mu noise issue

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

skidmorebay

Well-known member
Joined
Aug 1, 2011
Messages
134
Hi all,
I have a 1997 Manley Vari-Mu, mastering edition. I've been getting noise on the B channel (crackling, static bursts, elevated low frequency hum). Manley has been wonderful about offering tech support and troubleshoointing, but we haven't tracked down the problem yet. I don't want to send it to them because I'm using the working channel so much right now.

The noise happens usually around 30 minutes after start up. Sometimes the meter on that channel will stop working when the issue is present. The output is about 6db lower also. It will still pass audio.

Steps I have taken include swapping all the tubes, replacing most of the electrolytic caps, replacing the .1, .01, and .027 film caps, and replacing the tube sockets.

Anyone have experience with this and/or have another idea I can try before I box it up and ship it out of state?

thanks much,
JS
 
Were any of the electrolytic caps leaking when you replaced them? I have seen cap leakage in Manley devices and it can and does damage the circuit boards. When changing e-caps in Manley gear you need to A) know to look for the leaking caps, and B) know how to remediate the board damage if it has occurred.
Do you still have the caps you removed? If so, send pics. If not, pics of the board could help.
 
Really good points, thank you. The caps all looked good. No outward signs of deterioration that I noticed. I don't think I still have them. I'll check. I'll take another look at the board on both sides as well, and take pictures. I haven't noticed a damaged trace or similar, but I'll take another look.
 
Check DC resistance of the output transformer primary with it off. output stage B+ while on. I've seen a few going intermittent on their way to death that did similar while I was troubleshooting. Just to take that off the list.

30 minutes also seems like a solder joint or other connection with initial heating. We had similar on an SA-39B that took awhile to find, seemed to be a PCB via that was intermittent. Never measured anything off, but hardwiring around it fixed it.
 
I had similar symptoms with other pieces of gear, and it was the regulator(s) overheating most of the time.. I don't know if the varimu uses regulators, but it's only a suggestion. Good luck putting it back on its feet!
 
30 minutes also seems like a solder joint or other connection with initial heating. We had similar on an SA-39B that took awhile to find, seemed to be a PCB via that was intermittent. Never measured anything off, but hardwiring around it fixed it.
+1 I've seen this behaviour with cracked traces as well. Another test for this sort of issue, albeit a long and painstaking one, is to measure resistance from point-to-point with your DMM. This is easier when you have exposed leads, or, access to the underside of the board.
Example: You have two resistors that connect to a common trace, about an inch apart. Place your meter probes on the resistor leads, close to the body of the resistors. The resistance for an inch of copper, two resistor leads and a couple of solder joints should be very low, not much above what you see when you short the probes together. If there is a cracked trace, cold solder joint, bad via etc. you will see a higher or unstable reading. The idea of measuring between the exposed leads is that you are checking all points in between the two components in one go, without disturbing any intermittent issues. Particularly on older equipment, I find it's helpful to swab the leads with some isopropyl alcohol before testing to get a better contact with your probes.
 
Don't forget that resistors can go bad too - metal film resistors used in HV circuits. The plate resistors are particularly critical. Check the values after allowing a full discharge, or just replace them using HV-rated parts.
 
This seems heat related, for such problems I would try a can of freeze mist. Just start cooling components and solder joints along the circuit path and see if at some point the noise goes away. Freeze mist saved me a lot of time when I was service tech.
 
Thanks for the replies, everyone! All very good ideas.

I had a bit of a breakthrough yesterday. I found that by prodding around with a pencil eraser there is an area near the 5670 tube, where pressing on the pcb makes the noise go away. I re-flowed all the joints in that area and thought I had succeeded, but the noise came back later. I have a hunch it may be related to the limit/compress switch board nearby. Some of the wires from that daughter board connect in the same area.

At first I thought it was just the balance being out of adjustment, as I was able to dial the noise down completely when I re-adjusted it. But I think it may have just been the pressure of the screwdriver on the pcb that did the trick. I patched the output into Pro Tools and watched a spectrum analyzer plugin as I poked around with the eraser. I'm going to keep digging today.
 
Manley has been wonderful. They've given me a lot of tips, sent me a new set of caps, and checked in to see how I'm doing. Way more than I would expect, and more trouble for them than simply tell me to just send it to them for repair.

I've finally found something repeatable:
  • If I press down firmly on the input transformer, the noise goes away, until I remove the pressure.
  • pressing down with a pencil on the main pcb around the transformer does not have the same effect.
  • I removed the transformer (tricky job), since there are some solder pads under it I could not reach to re-flow. I cleaned up the header, re-installed it, and installed a couple jumper wires to be sure I could get a good connection to those pins that I couldn't reach easily to solder well.
  • I re-flowed the points on the input transformer pcb itself
After all that, it makes the noise when it is idling by itself. If I press down firmly on the transformer, the noise disappears. My hunch is that the issue is in the input transformer pcb itself, meaning it's not the pressure that creates on the main pcb that is restoring the proper connection. I've passed this on to Manley to see if it sounds familiar to them.
 
Back
Top