Recently I bought an old Akai AM-2200 amplifier for cheap, which is distortion on one output. First order of business was to replace the capacitors, didn't help the problem but I would have done that anyway due to the age of the amp.
I have tested each individual transistor in the faulty side of the amp using the diode voltage drop test, and all checked out okay.
I used a 1K sine wave for testing, and put a scope on the speaker output to see what the distorted sine wave looked like. With no load on the amp, it looked fine, just the same as the good side (see first photo). But when I connected a load (8 ohm speaker) the voltage drop on the output decreased significantly, and in the second photo you can see what the distorted waveform looks like.
The only thing that was changed between the two photos, is the first photo is with no load connected, the second is with the load connected. On the good side of the amp, connecting the speaker made very little difference to the wavefrom.
Another observation I made, is that if I increase the bias on the faulty side, then I can turn the volume up higher before it starts distorting. That leads me to think that one of the power transistors or drivers is not switching properly, so that it works okay while in class A mode but when switching to class B, is only half working. The only thing I dont understand is how that could happen if all the transistors tested okay, unless the e-b-c diode test is can make a transistor that is still faulty appear okay. Any help is greatly appreciated, all comments and ideas are welcomed. Thanks in advance
I have tested each individual transistor in the faulty side of the amp using the diode voltage drop test, and all checked out okay.
I used a 1K sine wave for testing, and put a scope on the speaker output to see what the distorted sine wave looked like. With no load on the amp, it looked fine, just the same as the good side (see first photo). But when I connected a load (8 ohm speaker) the voltage drop on the output decreased significantly, and in the second photo you can see what the distorted waveform looks like.
The only thing that was changed between the two photos, is the first photo is with no load connected, the second is with the load connected. On the good side of the amp, connecting the speaker made very little difference to the wavefrom.
Another observation I made, is that if I increase the bias on the faulty side, then I can turn the volume up higher before it starts distorting. That leads me to think that one of the power transistors or drivers is not switching properly, so that it works okay while in class A mode but when switching to class B, is only half working. The only thing I dont understand is how that could happen if all the transistors tested okay, unless the e-b-c diode test is can make a transistor that is still faulty appear okay. Any help is greatly appreciated, all comments and ideas are welcomed. Thanks in advance
Attachments
Last edited:
Check emitter resistors and such. Some of them seem to be fusible types (if I'm interpreting "F.S" correctly), which can apparently be unreliable.
The "PROTECTION" labels in the schematic are hilarious. They should've placed those at TR8 and TR9. 😛
The "PROTECTION" labels in the schematic are hilarious. They should've placed those at TR8 and TR9. 😛
I agree with sgrossklass about fusible resistors. Replace with plain resistors of similar value and retest the loaded output. As a caution, I don't find Vbe measurement is fully reliable in checking power BJTs. Some faiures go through stages that don't always affect the Vbe.
Last edited:
I found the fault in the amp, and by accident. I had failed to notice tr4 and tr5 on the pcb, so hadn't removed and done Vbe test on them. Anyway, I had the power amp board removed and had it supported and made sure it wasn't touching anything so that I could take measurements with it powered up etc, and I accidentally slipped, and something live must have touched the chassis because it blew a fuse and caused the output to go DC (not good). I found the only damaged transistor was TR4, which was an npn 2SD438. I replaced it with a BD139 and now the amp works great! It must have been faulty earlier, and when I accidentally shorted the amp it must have failed completely because that's the only transistor I replaced and now both channels work correctly.
Can someone please explain to me the purpose of TR4&5? I have uploaded the full schematic now which shows tr4&5
Can someone please explain to me the purpose of TR4&5? I have uploaded the full schematic now which shows tr4&5
Attachments
TR4,5
It looks to me like each is a regulator for the power rails of the input/VAS stages of both L & R amplifiers. It seems an unusually refined technique for such a simple amplifier but there it is - separate power supply for lower noise, one hopes.
It's ironic that you had to make sure it really was blown before you identified the fault 😀 but I guess that's fine now, as you say. Something tells me that the symptoms you described didn't match up to this fault, however. You were talking about load variation flattening the output level and the waveform suggests the input/VAS is functional but the output stage has a current problem. If you have full output with no load, the fault is not in the input/VAS stages, so I suspect that a problem is still lurking in the output stage.
BTW, thanks for posting schematics - it makes help so much easier 🙂
It looks to me like each is a regulator for the power rails of the input/VAS stages of both L & R amplifiers. It seems an unusually refined technique for such a simple amplifier but there it is - separate power supply for lower noise, one hopes.
It's ironic that you had to make sure it really was blown before you identified the fault 😀 but I guess that's fine now, as you say. Something tells me that the symptoms you described didn't match up to this fault, however. You were talking about load variation flattening the output level and the waveform suggests the input/VAS is functional but the output stage has a current problem. If you have full output with no load, the fault is not in the input/VAS stages, so I suspect that a problem is still lurking in the output stage.
BTW, thanks for posting schematics - it makes help so much easier 🙂
I should have said that a clue lies in the fault being in only one channel but these transistors supply both, so you can understand the difficulty.
Yes that's what I thought, I even swapped the output transistors from the good side to the bad side and it had no effect on the output behaviour. After replacing that one transistor I cannot get the amplifier to fault and it sounds great, so at that point I have declared it fixed 🙂
- Status
- Not open for further replies.