Adcom GFA-555 Help Needed

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello,

I picked up an Adcom GFA-555 with a blown left channel. As you can see, I've circled in red the burnt resistors (R1 and R7) and transistor (Q8). I'm assuming the problem is a blown output transistor, but I wanted to see if anyone has experience with this particular problem. If you agree that this is the most likely cause, I plan to update the left channel with the following parts (and later upgrade the right if it works as planned):

Q1-Q3, Q10: 2SC2240 (need 4 per channel) - Replace with KSC1845-FTA
Q4: 2SC2912 - Replace with KSC3503 DSTU
Q5: 2SA1016 - Replace with KSA992-FATA
Q6: 2SA1207 - Replace with KSA992-FATA
Q7: 2SA1210 - Replace with KSA1381 ESTU
Q8,Q9: 2SA970 - Replace with KSA992-FATA
Output Transistors
Q11: 2SB595 - Replace with MJE15031
Q12: 2SD525 - Replace with MJE15030
Q13-Q16: 2SB554 (PNP) - Replace with MJ21193G
Q17-Q20: 2SD424 (NPN) - Replace with MJ21194G

Does this seem right? Does anyone see any problem with these choices? I will do the caps as well.

Your help and advice is much appreciated.
 

Attachments

  • Adcom 555.jpg
    Adcom 555.jpg
    406 KB · Views: 380
Have you examined the schematic to determine what those resistors do? I would recommend against wholesale changing out of parts until you really understand what's wrong with the amp.

R1 is the INPUT series resistor. For that to blow means that someone must have either applied some sort of high voltage to the input, or when Q1/Q8 shorted (presumably), the short applied rail voltage to the input (presumably frying the guy's source/preamp output stage). Q8 is a bias transistor that is in parallel with R7, so whatever killed Q8, also probably killed R7. I'd also check Q7 and Q3 before lighting up the amp, since they are in the current path for the parts that got fried.

At any rate, if these are only parts that are fried, I'd start with replacing Q1 and Q2 and the resistors. I'd also read the posts on the GFA 565, which is the big brother of the GFA 555. The circuits are similar. See my thread on the Adcom 565. Yet Another Adcom GFA-565 Thread

In that thread there is a matcher circuit used for finding matched pairs of differential transistors. You will need to do that.

I'd carefully check the Q1 transistor that you remove. I'll bet it is shorted collector to base, and that's the problem.

Unless you are certain there is a problem in the output stage, I'd leave that alone. It is expensive and technically difficult to rebuild (AMHIK).

Here is the schematic:

AdcomGFA555.jpg


And here is the manual that sort of explains the circuit.

https://www.vintageshifi.com/repertoire-pdf/pdf/telecharge.php?pdf=Adcom-GFA-555-Service-Manual.pdf

The 555 is a great amp! So, good luck getting yours going!

Scott
 
Last edited:
Q1 Measurement

Using a multimeter on the Q1 transistor, appears to reveal the problem. Connecting the negative lead to the base and the positive lead to the collector, it measures .061V nstead of OL. Reversing the leads reveals the same measurement.

I'll report back after testing Q2, Q7, and Q8.

Your help is MUCH appreciated.
 
I would check R17 and Q4 to assure that whatever happened didn't propagate into the output stage. It is probably OK, but out of an abundance of caution...

I'd go ahead and try it with replacement transistors. You will eventually want to replace Q1 and Q2 with a matched pair. See the thread I suggested above for the matcher circuit. You should be able to breadboard that easily, and then test a handful of transistors to find two with similar characteristics.

I am curious. The description says: "Input amplification and feedback occur in Ql and Q2, a standard differential pair input stage, with R3 and R4 providing local feedback for open loop gain control."

However, in the diagram, I can't find R4. One would expect that it would be in the emitter leg of Q2, since differential pairs are usually symmetrical. Wondering if there is a typo in the schematic.
 
I went ahead and checked R17 and Q4 and they were fine. I also checked most of the other components on the board and could not find anything wrong.

So I went ahead and powered it up with a dim bulb tester and there was no indication of a short. I hooked it up to a preamp and speakers, still no sound in left channel.

So I checked the fuses for the left channel and one of them was blown (from before my repair). I replaced it and hoped that would solve the problem, but powered it up again and immediately blew the same fuse.

Checked my work on the input board again and all the other connections and I couldn't find anything wrong.

So after reading a few other threads, I pulled out two of the output transistors on that channel to check them. They are Toshiba 2SB554. The strange thing is that according to the spec sheet, these are PNP transistors, but when I measure them with a standard multimeter diode test, they test like NPN transistors. I checked multiple times to make sure I am measuring correctly based on the original toshiba spec sheet. And the two I pulled measure exactly the same.

I'm hesitant to conclude that they are damaged and order new ones until I'm sure I am not missing something here. It seems odd they measure the same and don't appear shorted other than measuring in reverse of what I expected.

Any ideas or other suggestions on how to proceed?
 
So I've hit a wall. It turns out that all the output transistors test fine. The problem I mentioned earlier was that my multimeter was setup wrong, and once I corrected it I retested literally every component in the output stage and they all tested perfectly.

But I'm at a loss as to what could be blowing the fuse.

I've checked all the transistors, replaced the electrolyte capacitors, and checked the key resistors at the beginning and end of the input stage.

I also tested the voltage coming into that channel and it is the same as the right channel.

Only thing else I can think of is to pull one leg of every resistor, but I just have this sinking suspicion that that is not the problem.

ANY HELP AT ALL WOULD BE SO APPRECIATED!!!!
 
The schematic shows the + of C3 going to R5 and the + of C5 going to the collector of Q9. If C5 is backwards it might affect the bias but honestly I'm not positive. I'm still learning. Worth getting them in the same way as the schematic anyway. :)


You didn't say which fuse is blowing.


TJ
 

PRR

Member
Joined 2003
Paid Member
> Does the polarity of C3 and C5 matter?

C4 does not really matter; but is correct as drawn.

C5 backward will short-out the bias. The amp will "work" (after all other faults are fixed) but idle stone-cold and be hoarse on small signals.
 
Now I'm really at my wit's end. I've replaced all the transistors in the left channel (apart from the output transistors, all of which I tested with a multimeter out of circuit with the diode test).

I tested all resistors and diodes and replaced the two that had problems.

The whole PCB has basically been rebuilt.

But now when I fire it up without a speaker load and connected to a dim bulb tester, R9 fries. The good news is that for the 30 seconds before it fries, there is only about 50mV across the speaker terminals as opposed to the full voltage that was there previously.

Can anyone speculate as to what could be frying R9? I checked all the transistors again and they have not shorted as far as I can tell.

Is it likely a problem with the output transistors that a diode test would not catch? I checked the emitter resistors as well.

Really could use some ideas before I drop $50 on new output transistors for no reason.
 

PRR

Member
Joined 2003
Paid Member
> R9 fries

That can't possibly happen. Something very strange is wrong.

If Q5 is remotely alive, R9 can't get more than 0.7V.

Even if Q5 is dud, Q3 R14 won't pass much over 2mA, and even all of that flowed up Q2 to R9 won't begin to warm R9.

So you need to stop assuming it is mostly right, or wholesale replacement, and figure out what is wrong.

Do you have a clear (un-fuzzy) image of the schematic? (My eyes have had all they can take.)

Leave R9 toasted. Mark-up a copy of the schematic with measured voltages. Is Q3 Base really 1.2V above -Supply? Are any B or E of Q1 Q2 within a Volt of zero?

While in there with a bright light.... are you sure some previous repair person has not left scraps or blobs of solder, stray wire, metal chips?
 
I'm attaching a scan of the service manual with the schematic. I'm 99% sure those transistors are working. They are brand new and not shorted, but I am going to recheck everything.

I don't know how this can be happening either, which is why I'm hoping someone can help :)

I appreciate your thoughts and efforts!
 

Attachments

  • hfe_adcom_gfa-555_service.pdf
    439.8 KB · Views: 78
So after removing R9, I fired up the amp last night and measured the voltage across each restistor. I didn't have time to measure voltage to ground, but I can do so tonight if helpful.

Here are the values:

R1 - 0
R2 - 216mv
R3 - 250mv
R5 - 425mv
R6 - 66V
R7 - 1.2V
R8 - 620mv
R10 - 1.2V
R11 - .7V
R12 - .6mV
R13 - 76V
R14 - 600mv
R15 - 21mV
R16 - 482mV
R17 - 4.9mv
R18 - 4mV
R19 - 295mV
R20 - 617mV
R21 - 2.3V
R22 - 0mV
Emitter Resistors - 1.38V
Positive and Negative Rails to Ground - ~72V
Across Caps - 72V

I have not tested the transistors again, but will do so tonight. It appears to me that Q7 might have gotten fried. There is some sort of glossy liquid underneath it. It is strange because that is not from a capacitor. It is definitely from Q7. The eletrolyte capacitors are new and intact.

Any initial thoughts are appreciated. I will check the transistors tonight and report on the voltages to ground of the resistors if helpful.

Thank you for your help.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.