Hi all,
This is my first attempt to build something like this, I have only put together some small kits before and nothing to do with audio.
I have put everything together now but I need help fixing it. I checked the power supply and it's ok, get the correct voltage out. I connected some speakers to the amp, the selector works, volume pot is good and I get good audio out from the right channel. The left amp is silent though. If I only connect the left speaker I can hear something very weak coming through only, otherwise it's near silent.
I need help troubleshooting that amp pcb to find out where it goes wrong. I'm not sure how to check the components or what else otherwise to check. The solder joints look ok to me but obviously it goes wrong somewhere.
I have taken some pictures of the whole thing, please let me know what other info you would want me to post. All help is appreciated here. It would be fantastic to be able to finish this great project. I have learned a lot and it would be great if I could get it working as I feel I'm very close now.
One thing I can tell you is when I check DC output on the left out while source is connected I get a steady DC out, something like -5 mV, I think. I can check again.
All pics here
This is my first attempt to build something like this, I have only put together some small kits before and nothing to do with audio.
I have put everything together now but I need help fixing it. I checked the power supply and it's ok, get the correct voltage out. I connected some speakers to the amp, the selector works, volume pot is good and I get good audio out from the right channel. The left amp is silent though. If I only connect the left speaker I can hear something very weak coming through only, otherwise it's near silent.
I need help troubleshooting that amp pcb to find out where it goes wrong. I'm not sure how to check the components or what else otherwise to check. The solder joints look ok to me but obviously it goes wrong somewhere.
I have taken some pictures of the whole thing, please let me know what other info you would want me to post. All help is appreciated here. It would be fantastic to be able to finish this great project. I have learned a lot and it would be great if I could get it working as I feel I'm very close now.
One thing I can tell you is when I check DC output on the left out while source is connected I get a steady DC out, something like -5 mV, I think. I can check again.
All pics here
-5mVdc is very acceptable for an output offset.
It probably indicates that there is no serious error that could damage your chip nor your PSU, nor your speaker.
Remove your speakers insert a dummy load. Power up via a bulb tester.
Let it warm up and while doing so repeatedly check the output AC and DC voltages.
Now apply a quiet music signal. Measure the output voltages again. Are they near zero?
Apply a louder music signal. Check the voltages again. If there is still near zero output, Check around the chip amp to feel if any component is overheating. Do not feel temps around your transformer. It is at mains voltage !!!!
Power off.
Check the underside of the PCB very thoroughly.
Check the component names and values and orientations.
It probably indicates that there is no serious error that could damage your chip nor your PSU, nor your speaker.
Remove your speakers insert a dummy load. Power up via a bulb tester.
Let it warm up and while doing so repeatedly check the output AC and DC voltages.
Now apply a quiet music signal. Measure the output voltages again. Are they near zero?
Apply a louder music signal. Check the voltages again. If there is still near zero output, Check around the chip amp to feel if any component is overheating. Do not feel temps around your transformer. It is at mains voltage !!!!
Power off.
Check the underside of the PCB very thoroughly.
Check the component names and values and orientations.
There is a steady output DC while I have a source and speakers connected, it doesn't change, but the other one plays fine.
Long shot, but make sure none of the RCAs are shorted to the chassis.
Some types of those plastic shoulder washers are very shallow and the barrel might have shifted while tightening the nut.
Unrelated, but I would use the same gauge wire for the outputs as for the DC power.
I would also insulate all the exposed 230V terminals, and re-arrange the RCA/selector wiring to minimize the possibility of noise pickup.
Some types of those plastic shoulder washers are very shallow and the barrel might have shifted while tightening the nut.
Unrelated, but I would use the same gauge wire for the outputs as for the DC power.
I would also insulate all the exposed 230V terminals, and re-arrange the RCA/selector wiring to minimize the possibility of noise pickup.
Since you have one good channel and one that's inoperative, this is a good opportunity to do "A to B" comparisions. With a digital voltmeter, take comparative readings from each of your amp PCBs. It would be interesting to see where they differ (worth reporting back to us on the thread), as that might give a good indication of the problem. Even more worthwhile, if you have an oscilloscope you can use for similar "A to B' measurements while providing a common audio source to both inputs. With the simplicity of the BrianGT boards, the problem should become easily evident. (In addition, you can do a good-board-to-bad-board comparision for resistor values, capacitor orientations--and don't forget a GOOD visual inspection of all solder joints (and potential bridges) on the bad board.
The best troubleshooting tool you have in a two-channel amp, when one channel is giving you problems--is the other (good) channel....!!
The best troubleshooting tool you have in a two-channel amp, when one channel is giving you problems--is the other (good) channel....!!
Looking at your pics, there are a number of PCB pads that look as though they could use more solder. (Might be the source of the problem, if you have a poor/no connection. I'd hit those pads again, and run more solder on them (don't be afraid to fill the pad with solder on each of the PCBs.....but do be careful not to form any solder bridges. (Bridges can be recoverable--but it's easier not to create them in the first place).
As we look at the amp pic that shows one amp module to the left, one to the right, and the PSU and rear apron/RCA connectors in the rear middle, which amp is NOT working--the one on the left, or the right?
As we look at the amp pic that shows one amp module to the left, one to the right, and the PSU and rear apron/RCA connectors in the rear middle, which amp is NOT working--the one on the left, or the right?
The one that's not working is the one on the left. The one that's behind the transformer.
Extra connection?! There's no extra connection, from transformer there's the 4 connections to AC1x2, and AC2x2.
I'll take some pics of the undersides of the amp PCBs, and I'll take note regarding more solder. I'm not sure it's needed but resoldering the faulty amp where it looks sketchy might be good.
Here are some more pics, they are also in the album:
Faulty amp, top
Faulty amp, bottom 1
Faulty amp, bottom 2
Faulty amp, bottom 3
Some pads look quite bad, I will resolder some now, and check again.
Faulty amp, top
Faulty amp, bottom 1
Faulty amp, bottom 2
Faulty amp, bottom 3
Some pads look quite bad, I will resolder some now, and check again.
Firstoff, in the immortal words of Douglas Adams:
DON'T PANIC!!
Keep it simple and methodical. Start at the beginning of the signal chain - the input RCA plug. You obviously have a voltmeter - does it have a "continuity" setting (ie. beeps at you when the probes touch each other)? Use it. Start with one probe on the center if the RCA plug on the outside of the case. Check you get a clean connection to the selector switch. Check that the input signal is not connected/shorted to ground. Verify the selector switch, then the volume potentiometer, etc. Work your way down the chain. Verify your connections are the same on the other channel. Keep checking for shorts to ground.
When you get to the amp boards, check connectivity between components. Do not (if possible) measure between the solder pads. Try to get the probes on the wires as they come out of the components instead.
Take your time and make notes about what you've done as you go along.
One question - I noticed that you have three RCA inputs, but only two connections to the selector switch. It is tough to visually trace all the wires - so what happened to that third channel? Have you verified that you don't have input wires crossed?
Good luck,
-bill
DON'T PANIC!!
Keep it simple and methodical. Start at the beginning of the signal chain - the input RCA plug. You obviously have a voltmeter - does it have a "continuity" setting (ie. beeps at you when the probes touch each other)? Use it. Start with one probe on the center if the RCA plug on the outside of the case. Check you get a clean connection to the selector switch. Check that the input signal is not connected/shorted to ground. Verify the selector switch, then the volume potentiometer, etc. Work your way down the chain. Verify your connections are the same on the other channel. Keep checking for shorts to ground.
When you get to the amp boards, check connectivity between components. Do not (if possible) measure between the solder pads. Try to get the probes on the wires as they come out of the components instead.
Take your time and make notes about what you've done as you go along.
One question - I noticed that you have three RCA inputs, but only two connections to the selector switch. It is tough to visually trace all the wires - so what happened to that third channel? Have you verified that you don't have input wires crossed?
Good luck,
-bill
Last edited:
BTW, and with no offense to the original poster, but this is a prime example of why it's good to try and a) use modular connectors where possible; b) use multiple wire colors and be consistent about their meaning; and c) keep the wire runs neat.
Debugging the rats nest is never fun. We've all had to do it at some point (I'm personally guilt of building some epic examples of "how not to do it" at times in the past). A little up front planning can go a long way in the end.
-bill
Debugging the rats nest is never fun. We've all had to do it at some point (I'm personally guilt of building some epic examples of "how not to do it" at times in the past). A little up front planning can go a long way in the end.
-bill
Hiya!
And thanks everyone for suggestions. So far no progress. I have resoldered some pads on the amp pcb, and rechecked components and things. Still the same. Tomorrow I'll follow signal path and check for good contact everywhere and do some more measurements. I have already done this partly and haven't found anything weird yet.
And thanks everyone for suggestions. So far no progress. I have resoldered some pads on the amp pcb, and rechecked components and things. Still the same. Tomorrow I'll follow signal path and check for good contact everywhere and do some more measurements. I have already done this partly and haven't found anything weird yet.
Last edited:
I'm sure you know this
Use a really hot and clean iron for soldering the pads. Lightly tin the bit and wipe any excess solder off with a damp sponge. The secret is to then firmly apply the iron to the component leg and t the same time "push" a little fresh solder wire onto the bit/part. Don't be to afraid of damaging the parts... if in doubt count 5 seconds or so should allow plenty of heat transfer.
With the amp OFF a few simple resistive checks of signal continuity should hopefully prove something. Measure (and compare) your readings one channel to another and make sure the continuity is as you expect. It's also good practice to keep the meter leads the same way around. Although it doesn't actually matter for pure resistive measurements it does when there are semiconductors and caps around and so it's good practice to keep polarities the same when testing and comparing to "identical" modules.
Be careful if you measure voltages with it on... one slip and the chips will let you know about it, and probably the speakers too.
If you are not sure if the problem is "signal wiring related" could you just swap the inputs to each module to prove it ?
Also (and be careful) if you turn the volume up just a little (so the wiper of the pot is away from ground) and if you touch the "input" side of the pot (not the middle wiper connection) with a metalic screwdriver can you hear a click of any sort ?
But be careful if you have speakers connected. It's sometimes worth just wiring a 47 ohm 0.5 or 1 watt resistor in series with a speaker. It will give total protection and will allow you to hear if the amp is working. The worst that happens is the resistor goes up in smoke.
Use a really hot and clean iron for soldering the pads. Lightly tin the bit and wipe any excess solder off with a damp sponge. The secret is to then firmly apply the iron to the component leg and t the same time "push" a little fresh solder wire onto the bit/part. Don't be to afraid of damaging the parts... if in doubt count 5 seconds or so should allow plenty of heat transfer.
With the amp OFF a few simple resistive checks of signal continuity should hopefully prove something. Measure (and compare) your readings one channel to another and make sure the continuity is as you expect. It's also good practice to keep the meter leads the same way around. Although it doesn't actually matter for pure resistive measurements it does when there are semiconductors and caps around and so it's good practice to keep polarities the same when testing and comparing to "identical" modules.
Be careful if you measure voltages with it on... one slip and the chips will let you know about it, and probably the speakers too.
If you are not sure if the problem is "signal wiring related" could you just swap the inputs to each module to prove it ?
Also (and be careful) if you turn the volume up just a little (so the wiper of the pot is away from ground) and if you touch the "input" side of the pot (not the middle wiper connection) with a metalic screwdriver can you hear a click of any sort ?
But be careful if you have speakers connected. It's sometimes worth just wiring a 47 ohm 0.5 or 1 watt resistor in series with a speaker. It will give total protection and will allow you to hear if the amp is working. The worst that happens is the resistor goes up in smoke.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.