Author Topic: Audio amp step-by-step repair  (Read 5788 times)

0 Members and 1 Guest are viewing this topic.

Offline okwTopic starter

  • Regular Contributor
  • *
  • Posts: 244
  • Country: no
Audio amp step-by-step repair
« on: September 07, 2021, 04:05:22 pm »
Hi! I've searched the forum but couldn't find any beginners "step-by-step" guide for audio repair. Of course, all repair would greatly depend on the symptoms, but any help would help me. Specifically I have a Ampeg BA115HP (bass guitar amp) which makes a huge pop at power up and does not output any sound when played.
Here is the schematic:
https://ampeg.com/support/files/Schematics/B%20Series/BA115HP/457SCH_0.pdf

I have a Fluke 183, UNI-T UTG962E signal generator, TDS 1012B oscilloscope, soon a 200w 8ohms dummy load (and other equipment) at my disposal.
I'm decent in power and digital repair, but have not yet tried audio, so please, any help on where to start and how to move through the circuit would be appreciated :)
 

Offline keland_uk

  • Contributor
  • Posts: 21
  • Country: gb
Re: Audio amp step-by-step repair
« Reply #1 on: September 07, 2021, 04:18:28 pm »
Ah, you have MOSFET's, they should call them fuses. :-) (Sounds like it's not blowing your fuses tho, unlike my Sony whose ouput MOSFETS had shortened.

I'm in same boat ref starting out repairing audio amps.

I'm fixing a Drawmer 1960 compressor for a friend. Valve heater power supply suspect, BUT not whats causing a problem on CH2 which doesn't work in bypass

I have a Flair1 infrared camera, and it appears that one of the NE5532's has failed, certainlty very hot..... (watch the voltages out there....)

Good luck... I'll also be interested in any hints et al
 

Offline Smith

  • Frequent Contributor
  • **
  • Posts: 376
  • Country: 00
Re: Audio amp step-by-step repair
« Reply #2 on: September 07, 2021, 04:36:19 pm »
Ah, you have MOSFET's, they should call them fuses. :-) (Sounds like it's not blowing your fuses tho, unlike my Sony whose ouput MOSFETS had shortened.

Not quite, the MOSFET's blow your fuses.

I assume you are talking about amplifiers in particular. It's basically just like any other repair.

If it blows fuses, you start at the power supplies. Most amplifiers quite a lot of different voltages. It may be a good idea to find the reason you lost the power first, most times just replacing fuses and blowing hem again does more damage than you started with.

A common problem the safety kicks in. The output relays will be disabled, and if there is a display it will tend to say something like protection. Some amplifiers have LED's or will blink some LED. This commonly means there is DC on the output, and the amplifier is pretending you to blow up your speakers. Mostly means one or more output transistors (and probably the resistors) are blown, or one of the voltages is gone.

If it doesn't start at all check all voltages, and find out why they where gone. Sometimes caps, regulators, bad connections (incl solder joints).

If anything is not working, or not properly, check switches, potentiometers and relays first. Specially with older gear. Specifically relays are difficult to troubleshoot the first time. Normally the sound will work, and at low volumes a channel will distort, of stop outputting audio. You can clean potentiometers or switches, relays should be replaced. If it is not your gear check ALL switches and connections. You won't believe how many amps I "fixed" by enabling the right input, plug the input to the right connection or switching the speaker selection switch.

If you still have problems, check ALL soldering connections, specifically ones that tend to get hot (output transistors, voltage regulators)  Re-solder them is necessary. A plastic stick is great for finding bad solder joints. Apply voltage to the device and just poke around carefully.

Sticky brown goo that was used to hold connectors and caps in their place can become conductive after any years.

Maybe this helps as a basic guide  ;)

In your case I would do the following:

1. Do NOT power the device with speakers connected. It has NO protection, and DC will kill speakers fast.
2. Check all the voltages
3. If they are OK, put some audio in the amp. Like a sine wave from a PC or function generator. Measure down the schematic, follow the path of the audio, until it stops (check AC AND DC on BOTH channels). Use a sine at it is continuous and easy to measure with your multimeter. Audio will go on and off and is terrible to troubleshoot. It's best not to use a load, as a sine is a quite heavy load for any amp.
« Last Edit: September 07, 2021, 04:54:04 pm by Smith »
Trying is the first step towards failure
 

Offline keland_uk

  • Contributor
  • Posts: 21
  • Country: gb
Re: Audio amp step-by-step repair
« Reply #3 on: September 07, 2021, 05:35:53 pm »
Not quite, the MOSFET's blow your fuses.

I meant fuse destroyers  :-DD

but thanks, thats helped me as well.

Just to be clear ref load, if the o/p is via valves then you need a load? ie fine for semimiconductors, not for valves in the o/p stage.
 

Offline keland_uk

  • Contributor
  • Posts: 21
  • Country: gb
Re: Audio amp step-by-step repair
« Reply #4 on: September 07, 2021, 05:46:36 pm »

 soon a 200w 8ohms dummy load (and other equipment) at my disposal.


I've done a dummy load based on Pete Milletts design, using an old BOSE SA2 case, was a bit of fun.
https://www.diyaudio.com/forums/tubes-valves/284351-audio-dummy-load-2.html
Thinking of doing a Reactive load based on this design to fit same said Bose SA2 Shell
(SA3 are useful as standalone amplifiers so I wouldn't iuse one of those, seem to be a lot of people throwing these Bose amps out, the B&O ones are silly money)
https://www.thegearpage.net/board/index.php?threads/aikens-reactive-dummy-load.1072793/

Is your one commercial load, and is it reactive load? (Since Guiter Amp)

 

Offline Smith

  • Frequent Contributor
  • **
  • Posts: 376
  • Country: 00
Re: Audio amp step-by-step repair
« Reply #5 on: September 07, 2021, 07:14:48 pm »
Just to be clear ref load, if the o/p is via valves then you need a load? ie fine for semimiconductors, not for valves in the o/p stage.

You don't need a load on most amplifiers.  Amplifiers MAY become unstable without load. Never happened to me though.

The only thing you need to remember when using a load, is that an amp will play audio, and the average power to the output will therefore be quite low. When using a load with a function generator you can generate a lot of heat within minutes depending on the volume. This will stress both the output stage AND the power supply. Most power supplies will have reasonable capacitance as a buffer, the power supply might not be able to generate this continuous load.

Not to mention the load resistors can become very hot.
Trying is the first step towards failure
 

Offline keland_uk

  • Contributor
  • Posts: 21
  • Country: gb
Re: Audio amp step-by-step repair
« Reply #6 on: September 07, 2021, 10:35:01 pm »

Not to mention the load resistors can become very hot.

Yep, basically think of them as an oven  :-+

Had to blow on mine to keep temperature reasonable.....Albeit I'd never run the dummy load at that wattage, and anyway I was overegging these, should have limited the total to 100W, since 50W resistors in series, hence current goes through both  |O

Mine is so I can test the THD et al after I've "fixed" said amplifier, my ears aren't what they used to be...

 

Offline okwTopic starter

  • Regular Contributor
  • *
  • Posts: 244
  • Country: no
Re: Audio amp step-by-step repair
« Reply #7 on: September 27, 2021, 11:24:12 pm »

I've gone through the whole circuit now, and all voltage rails are ok, signal levels (test points) are more or less within the values from the service manual. Except this:
TP10 (NE5532 out2) should be 2.0VRMS, which is 10mV. I suspected the op-amp and replaced it, still 10mV. I measured in2- and it is also around 0V. Even though TP12 was 950mVRMS (should be 800mVRMS), which is connected with a 1k between them (and a lot of other stuff).
TP11 should be 26.3V, but i 31.6V.

The schematic is here:
https://ampeg.com/support/files/Schematics/B%20Series/BA115HP/457SCH_0.pdf

Any help would be appreciated.
 

Offline Grandchuck

  • Frequent Contributor
  • **
  • Posts: 646
  • Country: us
Re: Audio amp step-by-step repair
« Reply #8 on: September 28, 2021, 11:19:36 am »
R22 open?  Or joint or PCB problem?
 

Offline xavier60

  • Super Contributor
  • ***
  • Posts: 2823
  • Country: au
Re: Audio amp step-by-step repair
« Reply #9 on: September 28, 2021, 11:42:02 am »
TP10 signal will be low if the amplifier's output isn't loaded.
What is happening at the output?
TP 11 should be 0V DC.
« Last Edit: September 28, 2021, 11:43:42 am by xavier60 »
HP 54645A dso, Fluke 87V dmm,  Agilent U8002A psu,  FY6600 function gen,  Brymen BM857S, HAKKO FM-204, New! HAKKO FX-971.
 

Offline okwTopic starter

  • Regular Contributor
  • *
  • Posts: 244
  • Country: no
Re: Audio amp step-by-step repair
« Reply #10 on: September 28, 2021, 02:52:48 pm »
R22 open?  Or joint or PCB problem?

R22 is 1k (0.998k ish) . I've resoldered most of the solder joints.
 

Offline okwTopic starter

  • Regular Contributor
  • *
  • Posts: 244
  • Country: no
Re: Audio amp step-by-step repair
« Reply #11 on: September 28, 2021, 02:56:13 pm »
TP10 signal will be low if the amplifier's output isn't loaded.
What is happening at the output?
TP 11 should be 0V DC.

I left the output open, to not damage the speaker. Should I connect an oscilloscope on the output? No risk of damaging it (it is a TDS 1012B with x10/x1 probe)?

In the schematics TP11 says 26.3V. Do you mean it should be 0V without load. And 26.3V with output loaded?
 

Offline xavier60

  • Super Contributor
  • ***
  • Posts: 2823
  • Country: au
Re: Audio amp step-by-step repair
« Reply #12 on: September 28, 2021, 03:36:30 pm »
TP10 signal will be low if the amplifier's output isn't loaded.
What is happening at the output?
TP 11 should be 0V DC.

I left the output open, to not damage the speaker. Should I connect an oscilloscope on the output? No risk of damaging it (it is a TDS 1012B with x10/x1 probe)?

In the schematics TP11 says 26.3V. Do you mean it should be 0V without load. And 26.3V with output loaded?
Probing the output will not damage your DSO. The schematic shows TP11 as being 26.3V RMS. Im certain that it means 26.3V RMS AC.
Don't connect the speaker if there is significant DC voltage at the output.
« Last Edit: September 28, 2021, 03:38:08 pm by xavier60 »
HP 54645A dso, Fluke 87V dmm,  Agilent U8002A psu,  FY6600 function gen,  Brymen BM857S, HAKKO FM-204, New! HAKKO FX-971.
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 14192
  • Country: de
Re: Audio amp step-by-step repair
« Reply #13 on: September 28, 2021, 03:48:11 pm »
The ouput should be close to 0 V DC level, but of cause with the music / tone, which is AC.

Similar TP12 should be near zero ( < 10  mV) DC.

With a 10:1 probe the scope is safe. Most scopes should also be OK even without the probe, but no need to try.


The first part should be up to TP12 - that is the low power part. So try with some signal from the function generator and check with the scope.
A slightly higher amplitude like 950 mV instead of 800 mV should be OK. For easier comparision maybe reduce the amplitde at the input to get back to the nomal 800 mV.

 

Offline okwTopic starter

  • Regular Contributor
  • *
  • Posts: 244
  • Country: no
Re: Audio amp step-by-step repair
« Reply #14 on: September 28, 2021, 05:08:30 pm »
TP10 signal will be low if the amplifier's output isn't loaded.
What is happening at the output?
TP 11 should be 0V DC.

I left the output open, to not damage the speaker. Should I connect an oscilloscope on the output? No risk of damaging it (it is a TDS 1012B with x10/x1 probe)?

In the schematics TP11 says 26.3V. Do you mean it should be 0V without load. And 26.3V with output loaded?
Probing the output will not damage your DSO. The schematic shows TP11 as being 26.3V RMS. Im certain that it means 26.3V RMS AC.
Don't connect the speaker if there is significant DC voltage at the output.

Yes, at all test points (TP) I've only measured AC. I'll go through the circuit again and measure DC, and see where DC is introduced. If this is the correct way?
 

Offline okwTopic starter

  • Regular Contributor
  • *
  • Posts: 244
  • Country: no
Re: Audio amp step-by-step repair
« Reply #15 on: September 28, 2021, 05:10:25 pm »
The ouput should be close to 0 V DC level, but of cause with the music / tone, which is AC.

Similar TP12 should be near zero ( < 10  mV) DC.

With a 10:1 probe the scope is safe. Most scopes should also be OK even without the probe, but no need to try.


The first part should be up to TP12 - that is the low power part. So try with some signal from the function generator and check with the scope.
A slightly higher amplitude like 950 mV instead of 800 mV should be OK. For easier comparision maybe reduce the amplitde at the input to get back to the nomal 800 mV.


I reduced the amplitude to 700mVrms (sine 440Hz).
I'm measuring 37.5V DC on the output (J23) with my multimeter.
Oscilloscope on same output: Pk-Pk 17.6V, Period 2.264ms?, Cyc RMS: 38V, Freq 440Hz.
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 14192
  • Country: de
Re: Audio amp step-by-step repair
« Reply #16 on: September 28, 2021, 05:14:27 pm »
38 V DC and some 17 V_PP would suggest that one side never turns on. If the FETs get really hot it may be 1 side never turns off.
A next point to check is the gate voltages / BJT collectors.
 

Offline okwTopic starter

  • Regular Contributor
  • *
  • Posts: 244
  • Country: no
Re: Audio amp step-by-step repair
« Reply #17 on: September 28, 2021, 05:32:44 pm »
38 V DC and some 17 V_PP would suggest that one side never turns on. If the FETs get really hot it may be 1 side never turns off.
A next point to check is the gate voltages / BJT collectors.
I'll do that. What voltages am I looking for?
I used a thermal camera and the FETs stay cold. Seems the two 330R (R66 / R68) gets really hot (~90'C), fast. Along with R6/R7, but they're not nearly as hot.

I DC measured every test point, all within 0-4mV DC. Except:
TP4: -1.4mV.
TP6: 158mV
TP7: 127mV.
TP10: -262mV.
TP11: 2.25V.

I also discovered, the amplitude on the oscilloscope jumped from 16-18V pk-pk (output J23) to almost 90V immediately when I touched TP12 (pin 1 of U5 / NE5532). And it stayed 78-90V after i let go of TP12.
« Last Edit: September 28, 2021, 06:40:22 pm by okw »
 

Offline okwTopic starter

  • Regular Contributor
  • *
  • Posts: 244
  • Country: no
Re: Audio amp step-by-step repair
« Reply #18 on: September 28, 2021, 05:47:30 pm »
38 V DC and some 17 V_PP would suggest that one side never turns on. If the FETs get really hot it may be 1 side never turns off.
A next point to check is the gate voltages / BJT collectors.

I DC measured the FETs relative to ground.
G: +/-48V
D: 37V
S: +/-51V
The N-channels (Q5/Q8) G and S are negative, the P-channels (Q2/Q7) G and S are positive.

The other transistors relative to ground:
Q1 gate: -3.614V
Q3 gate: -2.891V
Q4 gate: 3.282V
« Last Edit: September 28, 2021, 06:06:20 pm by okw »
 

Online Audiorepair

  • Frequent Contributor
  • **
  • Posts: 700
  • Country: gb
Re: Audio amp step-by-step repair
« Reply #19 on: September 28, 2021, 06:30:50 pm »

TP4: -1.4mV.
TP6: 158mV
TP7: 127mV.
TP10: -262mV.
TP11: 2.25V.
TP12: 37.5V.

I also discovered, the amplitude on the oscilloscope jumped from 16-18V pk-pk to almost 90V immediately when I touched TP12 (pin 1 of U5 / NE5532). And it stayed 78-90V after i let go of TP12.



It doesn't make sense that you are reading 37.5v at TP12, the output of op-amp U5A, when this op-amp has a supply voltage of +/- 16v.
Maybe you have a ground missing somewhere or your measurements/references are flawed.

Providing a temporary ground reference to this part of the circuit by touching it and then stuff happens might be a clue.


Edit:  Perhaps the +/-50v supply does not share the same ground as the low voltage stuff.
« Last Edit: September 28, 2021, 06:44:28 pm by Audiorepair »
 

Offline okwTopic starter

  • Regular Contributor
  • *
  • Posts: 244
  • Country: no
Re: Audio amp step-by-step repair
« Reply #20 on: September 28, 2021, 06:42:14 pm »

TP4: -1.4mV.
TP6: 158mV
TP7: 127mV.
TP10: -262mV.
TP11: 2.25V.
TP12: 37.5V.

I also discovered, the amplitude on the oscilloscope jumped from 16-18V pk-pk to almost 90V immediately when I touched TP12 (pin 1 of U5 / NE5532). And it stayed 78-90V after i let go of TP12.



It doesn't make sense that you are reading 37.5v at TP12, the output of op-amp U5A, when this op-amp has a supply voltage of +/- 16v.
Maybe you have a ground missing somewhere or your measurements/references are flawed.

Providing a temporary ground reference to this part of the circuit by touching it and then stuff happens might be a clue.

Sorry, my bad. I meant J23 (output), was 37.5V DC.
TP12 is 940mV AC, 4.0mV DC (but fluctuating quite a lot between 3.5-4.3mV DC).
 

Offline xavier60

  • Super Contributor
  • ***
  • Posts: 2823
  • Country: au
Re: Audio amp step-by-step repair
« Reply #21 on: September 28, 2021, 07:22:51 pm »
While there is 37V DC at the output, the waveform must also be distorted?
When you said
 "I also discovered, the amplitude on the oscilloscope jumped from 16-18V pk-pk (output J23) to almost 90V immediately when I touched TP12 (pin 1 of U5 / NE5532). And it stayed 78-90V after i let go of TP12."
Sounds like something is going intermittent open, Q1 possibly. Are Q1 and Q4 on a heatsink?
Can you show us the output waveform when the 37V DC is present? A small load might better show up the problem, like a household light bulb.
HP 54645A dso, Fluke 87V dmm,  Agilent U8002A psu,  FY6600 function gen,  Brymen BM857S, HAKKO FM-204, New! HAKKO FX-971.
 

Online Audiorepair

  • Frequent Contributor
  • **
  • Posts: 700
  • Country: gb
Re: Audio amp step-by-step repair
« Reply #22 on: September 28, 2021, 07:32:16 pm »
Remove the mute JFets Q6 and Q9, and bridge D-S of Q9 to maintain the signal to the op-amp, and put them back later.

These are so often faulty, but partially, or intermittently, or just don't help when you are fault finding as  they mute stuff because there is a fault.




The Opto-coupler OC1 also seems to be a limiter, which would also reduce the AC in DC fault conditions.
You could replace the resistive part with a similar fixed resistor to eliminate it while testing.

These aren't nearly as unreliable as a J112 however.
« Last Edit: September 28, 2021, 07:54:43 pm by Audiorepair »
 

Offline xavier60

  • Super Contributor
  • ***
  • Posts: 2823
  • Country: au
Re: Audio amp step-by-step repair
« Reply #23 on: September 28, 2021, 07:55:38 pm »
TP12 should normally be within a few mV of 0V DC.
HP 54645A dso, Fluke 87V dmm,  Agilent U8002A psu,  FY6600 function gen,  Brymen BM857S, HAKKO FM-204, New! HAKKO FX-971.
 

Offline okwTopic starter

  • Regular Contributor
  • *
  • Posts: 244
  • Country: no
Re: Audio amp step-by-step repair
« Reply #24 on: September 28, 2021, 07:56:34 pm »
While there is 37V DC at the output, the waveform must also be distorted?
When you said
 "I also discovered, the amplitude on the oscilloscope jumped from 16-18V pk-pk (output J23) to almost 90V immediately when I touched TP12 (pin 1 of U5 / NE5532). And it stayed 78-90V after i let go of TP12."
Sounds like something is going intermittent open, Q1 possibly. Are Q1 and Q4 on a heatsink?
Can you show us the output waveform when the 37V DC is present? A small load might better show up the problem, like a household light bulb.

Waveform is a beautiful sine at 440Hz, same as going in just very amplified.

Q1 and A4 are on a heat sink. Those clip on, mickey mouse heat sinks.

I'll see if I can find a normal light bulb. I mostly have LEDs.

See images for input signal (721mVrms, 440Hz sine with probe set to 1x). Oscilloscope probe is set to 10x.
And one output image untouched, and one output image where I touch TP12. The crazy amplification did not keep after I let go of TP12 (like the last time, last time it was also slighly distorted).
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf