Hi Infinia,
If the really big amps use 2R2 and up, I don't think that is much of an issue. Never had one go in normal use. Mind you, they are 10W and 20W parts.
When the secondary DC comes up enough to pull in the relay, the surge resistor is shorted and can cool down. Remember that your max inrush current will be LINE VOLTAGE / SURGE RESISTOR + XFORMER DCR. The higher resistance is nicer to everything, including other units on the same circuit.
-Chris
If the really big amps use 2R2 and up, I don't think that is much of an issue. Never had one go in normal use. Mind you, they are 10W and 20W parts.
When the secondary DC comes up enough to pull in the relay, the surge resistor is shorted and can cool down. Remember that your max inrush current will be LINE VOLTAGE / SURGE RESISTOR + XFORMER DCR. The higher resistance is nicer to everything, including other units on the same circuit.
-Chris
Adcom GFA 555 bias adj
Great thread here. I'm pretty active in some of the other forums, but found this thread because I'm in the middle of fixing a GFA 555 with a dead channel. So far looks like output transistor failure from the last owner pushing it hard in bridge mode. I've got an open/shorted Q-18 and an open reistor at R-25. Driver checks out fine.
I'm looking for the bias setting for this amp. It is wonderful that the schematic and parts list was posted, but what about the Sm settings, which I think consist only of bias.
Thanks a bunch!
Great thread here. I'm pretty active in some of the other forums, but found this thread because I'm in the middle of fixing a GFA 555 with a dead channel. So far looks like output transistor failure from the last owner pushing it hard in bridge mode. I've got an open/shorted Q-18 and an open reistor at R-25. Driver checks out fine.
I'm looking for the bias setting for this amp. It is wonderful that the schematic and parts list was posted, but what about the Sm settings, which I think consist only of bias.
Thanks a bunch!
DJK, thanks for your reply and I agree, the drivers should be replaced as well.
Before reading your post, I briefly ran it up with my variac and the formerly dead channel now works but is not right, outputs on the repaired board are getting warm very fast. I haven't checked but I'm sure the bias is out of control.
Are the drivers the TO-126 units Q 9 and Q 10?
Before reading your post, I briefly ran it up with my variac and the formerly dead channel now works but is not right, outputs on the repaired board are getting warm very fast. I haven't checked but I'm sure the bias is out of control.
Are the drivers the TO-126 units Q 9 and Q 10?
GFA 555 channel repair
I have updated info with new issues if anyone is following along.
This all pertains to my failed left channel, the right channel is working fine and the bias is stable at 20 mV and the DC is a bit high at .45 mV, but acceptable for now.
I replaced both driver transistors, Q12 and Q11 which call for 2SD525 and 2SB595, with Onsemi MJE15030/31. After reading a few threads here, looks like MJE15032/33 may have been better choices, which I also have in stock. Any comment appreciated.
Anyway, I tested Q4 and Q7 out of circuit and they passed the diode test, so I reinstalled. I do not have any equipt. to test beyond a simple diode test.
Powered up again via dimbulb tester and the repaired channel is not right, runaway bias over 6 vdc and I get about 6.5 DC at the speaker terminals. The amp outputs on the repaired channel, but obviously things are not right.
I'm a great parts replacer, but not so good with theory. So if anyone wants to jump in with suggestions for my next move, I'd really appreciate it.
Thanks!
I have updated info with new issues if anyone is following along.
This all pertains to my failed left channel, the right channel is working fine and the bias is stable at 20 mV and the DC is a bit high at .45 mV, but acceptable for now.
I replaced both driver transistors, Q12 and Q11 which call for 2SD525 and 2SB595, with Onsemi MJE15030/31. After reading a few threads here, looks like MJE15032/33 may have been better choices, which I also have in stock. Any comment appreciated.
Anyway, I tested Q4 and Q7 out of circuit and they passed the diode test, so I reinstalled. I do not have any equipt. to test beyond a simple diode test.
Powered up again via dimbulb tester and the repaired channel is not right, runaway bias over 6 vdc and I get about 6.5 DC at the speaker terminals. The amp outputs on the repaired channel, but obviously things are not right.
I'm a great parts replacer, but not so good with theory. So if anyone wants to jump in with suggestions for my next move, I'd really appreciate it.
Thanks!
Hi:
As I understand, you replaced only *one* output transistor Q18 and the others Q17, Q19, Q20 are still "originals"? Ideally when one output transistor goes bad, one should replace all four. Did you find a replacement Toshiba 2SD424 for the Q18? I had once repaired a GFA-555 with bad outputs by using the Onsemi MJ15024/25 devices.
This may be tedious, but could help diagnose the source of trouble: Check the bias and DC at the speaker terminals by using only 1 pair of outputs (say Q17 and Q13) in the bad channel and temporarily desoldering the rest of the output transistors.
Rgds
Mayank
As I understand, you replaced only *one* output transistor Q18 and the others Q17, Q19, Q20 are still "originals"? Ideally when one output transistor goes bad, one should replace all four. Did you find a replacement Toshiba 2SD424 for the Q18? I had once repaired a GFA-555 with bad outputs by using the Onsemi MJ15024/25 devices.
This may be tedious, but could help diagnose the source of trouble: Check the bias and DC at the speaker terminals by using only 1 pair of outputs (say Q17 and Q13) in the bad channel and temporarily desoldering the rest of the output transistors.
Rgds
Mayank
Mayank, yes, I know the outputs should be replaced in sets, but my work at this point is more in the diagnostic mode and once I make some progress, I will decide if I go with new Onsemi's or a set of original Toshiba's.
I do a lot of work on vintage receivers and have a decent supply of genuine Toshiba 2SB554/2SC424 outputs, so what I replaced was a known good original and not a Chinese fake.
Along the lines of your suggestion, I swapped the output board that I repaired with board 1 from the other channel. With the repaired board in the left channel, I get DC of .40 mV and I can run the Bias at spec, no problem. The output sounds fine and this has led me to believe that the problem is in the input/driver PC board.
I suspect the Q4-Q7 transistors, they did pass the diode test, so I'm not sure. I checked the resistors and diodes incircuit and they test the same as the other side. I'm guessing I have a bad transistor, but at this point I'm not sure how to proceed.
I do a lot of work on vintage receivers and have a decent supply of genuine Toshiba 2SB554/2SC424 outputs, so what I replaced was a known good original and not a Chinese fake.
Along the lines of your suggestion, I swapped the output board that I repaired with board 1 from the other channel. With the repaired board in the left channel, I get DC of .40 mV and I can run the Bias at spec, no problem. The output sounds fine and this has led me to believe that the problem is in the input/driver PC board.
I suspect the Q4-Q7 transistors, they did pass the diode test, so I'm not sure. I checked the resistors and diodes incircuit and they test the same as the other side. I'm guessing I have a bad transistor, but at this point I'm not sure how to proceed.
Good to eliminate the variable of possible fake Chinese Toshiba devices.
To be certain "beyond-a-reasonable-doubt" that the output boards are now OK, I would recommend a final step (continuing along the lines of your diagnosis):
Swap (right/good -> left/bad) the two sets of 3 single strand wires for each channel connecting the input board to the output boards. These are light brown/brown/green & orange/yellow/red representing V+/+D/+B and V-/-D/-B.
If the repaired output board is OK, you should have normal Bias and DC measurements in your "bad channel". Also, the good output board should have high Bias/DC.
If this is what you observe, then the Output transistors *and* drivers are OK and the problem is definitely in the input board.
Rgds
Mayank
To be certain "beyond-a-reasonable-doubt" that the output boards are now OK, I would recommend a final step (continuing along the lines of your diagnosis):
Swap (right/good -> left/bad) the two sets of 3 single strand wires for each channel connecting the input board to the output boards. These are light brown/brown/green & orange/yellow/red representing V+/+D/+B and V-/-D/-B.
If the repaired output board is OK, you should have normal Bias and DC measurements in your "bad channel". Also, the good output board should have high Bias/DC.
If this is what you observe, then the Output transistors *and* drivers are OK and the problem is definitely in the input board.
Rgds
Mayank
Mayank, I did as you suggest and the problem stays on the driver board, regardless of ouput board location. Thanks for the suggestions.
At this point, I'm going to pull Q4 and Q7, which passed the diode test, and swap them side to side and see if the problems stay with the predrivers or stay on the input/driver board. If that is inconclusive, I'll test all the diodes and TO-92 transistors out of circuit and hopefully find something suspect there. I already tested the diodes and resistors incircuit and all was good.
I have no problem throwing parts at this problem, but I'd like to learn exactly what has failed.
Any other suggestions?
At this point, I'm going to pull Q4 and Q7, which passed the diode test, and swap them side to side and see if the problems stay with the predrivers or stay on the input/driver board. If that is inconclusive, I'll test all the diodes and TO-92 transistors out of circuit and hopefully find something suspect there. I already tested the diodes and resistors incircuit and all was good.
I have no problem throwing parts at this problem, but I'd like to learn exactly what has failed.
Any other suggestions?
Okay, swapped the suspect predrivers, Q 4 and Q 7 to the "good' side and it works fine.
I ended checking all the resistors, electrolytic caps and diodes incircuit and they were fine. I pulled all the T0-92 transistors and found that Q 1 had a weak short. I did not test the silver mica or the film caps, perhaps I should test them for a short?
I don't know what Q 1 does in the ciricuit but I'm not sure it is responsible for both the runaway bias and the big DC at the speaker terminals for that channel.
So...I need to find a good non-NTE subsitute for a 2SC2240. Looks like the amp uses 4 per side, so I may just replace them all.
Also, it safe to power up the amp with parts missing from one channel of the driver board? I wanted to check DC on the channel that had the suspect predrivers.
Thanks
I ended checking all the resistors, electrolytic caps and diodes incircuit and they were fine. I pulled all the T0-92 transistors and found that Q 1 had a weak short. I did not test the silver mica or the film caps, perhaps I should test them for a short?
I don't know what Q 1 does in the ciricuit but I'm not sure it is responsible for both the runaway bias and the big DC at the speaker terminals for that channel.
So...I need to find a good non-NTE subsitute for a 2SC2240. Looks like the amp uses 4 per side, so I may just replace them all.
Also, it safe to power up the amp with parts missing from one channel of the driver board? I wanted to check DC on the channel that had the suspect predrivers.
Thanks
I used to be an audiphile.
I hung out at CES with the reviewers and people with mega systems.
Because of the obscene pricing in the high end (milionaires club), I am taking refuge in DIY and the people here are fantastic.
I never liked the sound of an amp bridged. It also sounded more closed in. It never breathes.
I think if you want to improve the sound of an amplifier driving a heavy load you improve the power supply.
What I have done is convert a stereo amp to a monblock and they use 2 of them. If the amp uses a common power supply you now have twice the energy storage and the amp now has more depth and breathes. It is true that the wattage stays the same unless you have a low impendance speaker, in which case the amp will put out more power.
Just my thoughts.
I hung out at CES with the reviewers and people with mega systems.
Because of the obscene pricing in the high end (milionaires club), I am taking refuge in DIY and the people here are fantastic.
I never liked the sound of an amp bridged. It also sounded more closed in. It never breathes.
I think if you want to improve the sound of an amplifier driving a heavy load you improve the power supply.
What I have done is convert a stereo amp to a monblock and they use 2 of them. If the amp uses a common power supply you now have twice the energy storage and the amp now has more depth and breathes. It is true that the wattage stays the same unless you have a low impendance speaker, in which case the amp will put out more power.
Just my thoughts.
You wrote: "Is there any reason that you preferred the mje15030/31 drivers to the MJE15034 and 35s that Rod Elliot thinks so highly of?"
I have the 15030/31's in stock, so it was an easy swap in, however it was not an issue. I'm lazy and want immediate results, but don't mind making subsequent parts swaps once the problem is figured out.
My plan on this amp is to figure out what is wrong with it and then recap/change failed parts with upgrades and learn from the experience. My system doesn't depend on getting this GFA-555 running again luckily!
I have the 15030/31's in stock, so it was an easy swap in, however it was not an issue. I'm lazy and want immediate results, but don't mind making subsequent parts swaps once the problem is figured out.
My plan on this amp is to figure out what is wrong with it and then recap/change failed parts with upgrades and learn from the experience. My system doesn't depend on getting this GFA-555 running again luckily!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.