On my LM3886 amp, which was pre-assembled, there is intermittent static which sometimes results in the entire channel dropping out. It is not isolated to left or right, and seems random as to which channel will make noise or fail. What seems to "fix" it is turning the amp off and on again, and then it will work properly until I turn it off again. Sometimes I have to turn it on and off several times before it works properly.
My first guess was that there was an issue with the volume pot, so I cleaned it with some deoxit and that seemed to help but it did not eliminate the issue. I did replace the volume pot with a linear attenuator (awesome upgrade, sounds so much better and I did not expect that) BUT the channel static/cut-out issue persists. So I can say for sure the pot was part of the problem, but before I start replacing other parts I was hoping that someone else could suggest what this could be:
1) The power switch seems like the next logical thing to replace but it's a small switch and it appears to be wired downstream from the mains, meaning: Mains > Fuse > Transformer > Switch (on secondaries) rather than Mains > Fuse > Switch > Transformer Primary. If I had built the amp myself I would have placed the switch between the fused mains and transformer primary...and if I have to switch the switch, I'll likely rewire it to be that way unless there's a reason not to.
2) The volume pre-amp board, where the attenuator connects, also has two opamps and is powered by a secondary low voltage winding off of the transformer. Is the preamp board itself bad? Any way to test this without replacing it?
3) Something else? What?
My first guess was that there was an issue with the volume pot, so I cleaned it with some deoxit and that seemed to help but it did not eliminate the issue. I did replace the volume pot with a linear attenuator (awesome upgrade, sounds so much better and I did not expect that) BUT the channel static/cut-out issue persists. So I can say for sure the pot was part of the problem, but before I start replacing other parts I was hoping that someone else could suggest what this could be:
1) The power switch seems like the next logical thing to replace but it's a small switch and it appears to be wired downstream from the mains, meaning: Mains > Fuse > Transformer > Switch (on secondaries) rather than Mains > Fuse > Switch > Transformer Primary. If I had built the amp myself I would have placed the switch between the fused mains and transformer primary...and if I have to switch the switch, I'll likely rewire it to be that way unless there's a reason not to.
2) The volume pre-amp board, where the attenuator connects, also has two opamps and is powered by a secondary low voltage winding off of the transformer. Is the preamp board itself bad? Any way to test this without replacing it?
3) Something else? What?
Intermittent issues are fun, eh? 🙂
If the amp wakes up one day with the left channel dead and wakes up a different day with the right channel dead, then I'd say whatever is failing is NOT common to the two channels. If it was a power supply or power switch issue, I would expect it to take out both channels.
Can you do anything to provoke the issue? One way to do this is to apply pressure to the circuit boards with the amp powered up and connected to a pair of cheap test speakers. See if you can provoke the issue by pressing down on the circuit boards with the handle of a small screwdriver.
Also wiggle the wires connecting to the circuit boards to see if you can provoke the issue. It wouldn't surprise me if the issue turns out to be a bad connection on a power supply wire or signal wire.
Does the issue happen when the volume control is turned all the way down? If not, the issue is with the preamp.
Tom
If the amp wakes up one day with the left channel dead and wakes up a different day with the right channel dead, then I'd say whatever is failing is NOT common to the two channels. If it was a power supply or power switch issue, I would expect it to take out both channels.
Can you do anything to provoke the issue? One way to do this is to apply pressure to the circuit boards with the amp powered up and connected to a pair of cheap test speakers. See if you can provoke the issue by pressing down on the circuit boards with the handle of a small screwdriver.
Also wiggle the wires connecting to the circuit boards to see if you can provoke the issue. It wouldn't surprise me if the issue turns out to be a bad connection on a power supply wire or signal wire.
Does the issue happen when the volume control is turned all the way down? If not, the issue is with the preamp.
Tom
I tried pressing on the various components using a pen and didn't observe any problems. All of the wires were secure. Then, while playing sound, I use a small screwdriver to tighten the wires to and from the preamp. It goes from the RCA inputs to the preamp, and from the preamp to the amplifier board. What I noticed is that when I would touch the screwdriver to the signal wires coming OUT of the preamp or the other ends going into the amp, it would induce static. If I touched the signal wires coming into the preamp, or the RCA jacks themselves there was nothing at all. I also touched the grounding point for the signal wires and nothing happened. When the amp wasn't playing right, touching the signal wires with the screwdriver seemed to fix it. Maybe the wires are lonely and just want attention? I don't know if it is normal or not but it seems like there is something amiss with the preamp board. It's not completely dead but it has a gremlin. The preamp board is clean and all of the things soldered to it are solid. No cracks in the PCB or delamination. I don't know why touching the signal wires with the screw driver would cause static there but nowhere else. If it matters, the preamp is powered directly by the transformer AC and has its own little rectifier / power supply.
If you have the preamp connected to the power amp and both are powered up, I would expect a touch of a screwdriver to the connections between the two to do very little. It's normal that you'll pick up some hum if you touch the input to an amp if it's floating (i.e., not connected to anything).
Tom
Tom
Just for good measure, make sure that the LM3886 are well-seated on the heat sink. You can provoke thermal cycling if the devices aren't making good contact.
Is there a speaker protection relay between the LM3886 and the speaker?
Several times I've seen poor / worn relay contacts cause this kind of random trouble.
Similarly, is there a relay in the preamp output for power on delay? Same trouble.
Some relays have a removable cover and the contacts can be cleaned. Others are sealed and need replacement.
Relays are high on my list of supects that cause random behavior based on 40+ years of experience.
Several times I've seen poor / worn relay contacts cause this kind of random trouble.
Similarly, is there a relay in the preamp output for power on delay? Same trouble.
Some relays have a removable cover and the contacts can be cleaned. Others are sealed and need replacement.
Relays are high on my list of supects that cause random behavior based on 40+ years of experience.
@maudio Which part(s) to spray and what would be the reason for freeze spray? Testing for failure or is it a way to repair potentially failing components?
@jackinnj The chips are secured by a screw and have a thermal pad but I'll double-check to make sure nothing made its way loose.
@peranders Is there any way to readily identify this particular resistor? The amp PCB isn't labeled in detail and I don't have the original schematic.
@techtool Yes, on the amp board there is a relay and I've had my eye on it as a possible culprit. It has a clear cover and looks clean/new but I may spray some deoxit on it and see if that helps. There is no relay on the preamp board though.
Thanks for all the suggestions, guys!
@jackinnj The chips are secured by a screw and have a thermal pad but I'll double-check to make sure nothing made its way loose.
@peranders Is there any way to readily identify this particular resistor? The amp PCB isn't labeled in detail and I don't have the original schematic.
@techtool Yes, on the amp board there is a relay and I've had my eye on it as a possible culprit. It has a clear cover and looks clean/new but I may spray some deoxit on it and see if that helps. There is no relay on the preamp board though.
Thanks for all the suggestions, guys!
Freeze spray causes a thermal shock. That'll sometimes provoke intermittent failures to happen. By freeing parts of the circuit you can sometimes get lucky and tease out which area of the board that's causing the issue.
You can find the mute resistor by following the trace from the mute pin (pin 8) to the resistor.
If the relay is a speaker protection relay, you could try shorting it out with a piece of wire soldered to the bottom of the board. If the amp works reliably with the wire in place, replace the relay. Obviously you won't have any speaker protection when you short out the relay.
Tom
You can find the mute resistor by following the trace from the mute pin (pin 8) to the resistor.
If the relay is a speaker protection relay, you could try shorting it out with a piece of wire soldered to the bottom of the board. If the amp works reliably with the wire in place, replace the relay. Obviously you won't have any speaker protection when you short out the relay.
Tom
And I would connect the power switch on the mains side.
Better practice, so the entire unit is out of circuit when switched off.
Right now the primary side is always connected to mains until you switch off the mains, which is a makeshift job really.
Check or re do the signal input to pre amp to power amp side first, and use the biggest rated relay that will fit.
Their contacts may me worn out after many years of service.
MOV across secondary side of power supply will reduce contact wear, if you choose to use one.
Better practice, so the entire unit is out of circuit when switched off.
Right now the primary side is always connected to mains until you switch off the mains, which is a makeshift job really.
Check or re do the signal input to pre amp to power amp side first, and use the biggest rated relay that will fit.
Their contacts may me worn out after many years of service.
MOV across secondary side of power supply will reduce contact wear, if you choose to use one.
Last edited:
If you take a picture of the amplifier pub, then I can try to point it out but before you are doing too much. Do you have a voltmeter and/or an oscilloscope? It's essential to know the voltage in interesting places when it doesn't work.
OK, looks like the problem was the speaker protection relay. After spending more than 30 minutes trying to remove the cover of the relay intact, I decided to resort to an "ah, fcck it" method and cut the tabs with an x-acto knife. So annoying how parts that should be serviceable are so troublesome to access. Anyway, I got it open and dripped some deoxit on the contacts. I then gently scraped them with the x-acto knife because I don't have fine grit sand paper handy, nor did I want to use sandpaper. Long story short, it works really good now...don't know if it's going to stay or not but at least I know which part to replace if it starts acting up again.
For those who may be wondering, the particular relay is an Omron G2R-2 2577H4 that is rated for 5A @ 30 VDC and 250 VAC and it's about 4 years old, roughly daily use in that time. It isn't sparking or arcing so I guess the contacts just oxidized enough that the contact was unreliable. That also explains why turning the amp off and on would temporarily fix the issue.
Sidenote, I did gently tighten the screws on the tabs that secure the LM3886 chips to the heatsink. They were a bit loose. I haven't had any thermal shutdown issues yet and hopefully it stays that way.
Thanks @techtool and everyone else.
For those who may be wondering, the particular relay is an Omron G2R-2 2577H4 that is rated for 5A @ 30 VDC and 250 VAC and it's about 4 years old, roughly daily use in that time. It isn't sparking or arcing so I guess the contacts just oxidized enough that the contact was unreliable. That also explains why turning the amp off and on would temporarily fix the issue.
Sidenote, I did gently tighten the screws on the tabs that secure the LM3886 chips to the heatsink. They were a bit loose. I haven't had any thermal shutdown issues yet and hopefully it stays that way.
Thanks @techtool and everyone else.
Just drill a hole in the case next time.
Heat works to make holes as wello .
Then spray through the opening.
That is enough at times.
Heat works to make holes as wello .
Then spray through the opening.
That is enough at times.
You're welcome and glad it worked out. It was fresh on my mind as I just did the same repair last week on an Onkyo TX-26 receiver from 1985. Spooky....
Might want to put a lock-washer on the screws so it doesn't happen again, or perhaps one of those Loctite "Threadlocker" products.
Omron G2R-2 is a pretty common relay. If you know the coil voltage you can find a replacement on Mouser/Digikey. Just search for G2R-2. I'd replace it or you'll be digging into that amp again a few months from now.
You don't need a lot of torque on the mounting bolt. I go with 0.5 Nm, which is what Keratherm recommends for their thermal pads for that package size. That's barely beyond finger tight.
Tom
Used a higher rated relay, if possible, the GR2 is only about 5 amps.
It is generic, others also make relays with the same foot print.
Panasonic, among others, also make good relays, so if you can't find what you need in Omron, other makers do offer similar and higher rated relays.
Same footprint should be available, so less issue of mounting.
Parallel the contacts, if not done already. So less load on the contacts.
It is generic, others also make relays with the same foot print.
Panasonic, among others, also make good relays, so if you can't find what you need in Omron, other makers do offer similar and higher rated relays.
Same footprint should be available, so less issue of mounting.
Parallel the contacts, if not done already. So less load on the contacts.