Well the new Q4 came in today. Replaced it and sadly it did not fix the problem. I now have -20V instead of -27V so I guess we are going in the right direction.
The only other component that had funny measurements was Q12 but I didn't order any of those. I did unsolder it and remove it from the board and tested it with DMM and it tests good.
I took Q8 out of circuit and it tests fine. I also read that Q8 was compatible with the NTE 129 that I just bought. I had an extra so I swapped it in and got the same measurements as with the original Q8 in place.
Well that's a good call. I've been out there working on it this evening and realized that Q4 was an NPN and I had replaced it with a PNP. That's on me. Stevem recommended nte128 and like an idiot I bought nte129. My old Q4 tested fine now that I have it out of the board so I put it back in and checked all of my measurements again.
Measurements that make me scratch my head are the collector of Q4. The schematic shows -3.7 at Q4s collector but I'm getting -33V. There is a node at that point between Q4 collector, C7 and R18. Should it be -33V on both sides of C7 or could it be a typo on the schematic?
The other one is CR4 where it goes into the collector of Q7. All of the other diodes have the same voltage on both sides but CR4 has -25V on one side and -256mV on the other side and that -256mV is going into the collector of Q7. Should it be like that?
I'll try the Q6 swap tomorrow evening.
Here are my measurements after putting Q4 back
Component In Out
R1 47K 169mV .6V
R2 47K 193mV .6V
R3 82.5K 1% .6V 5.7V
R4 221K 1% .6V -8.5V
R5 2200 5.7V 8.5V
R6 4700 5.7V 5.6V
R7 33K 38mV 0V
R8 39K 42.85V 20.67V
R9 39K 20.68V .6V
R10 4700 -33V -33V
R11 3900 -34V -34V
R12 1820 1% -1.9V -2.2V
R13 33K -1.9V -26V
R14 2200 42V 11.4V
R15 2700 11.4V -19V
R16 97.6K 1% -2.3V -26V
R17 4700 -26V -33V
R18 470 -33V -27V
R19 470 -33V -33V
R20 470 -34V -42V
R21 1K -26V -26V
R22 100 -25V -26V
R23 1K -25V -26V
R24 100 -42V -42V
R25 820 5% -26V -26V
R26 .5 5W -26V -26V
R27 267 1% -26V -26V
R28 .5 5W -26V -26V
R29 267 1% -26V -26V
R30 820 5% -26V -26V
R31 100K -26V 109mV
R32 18K 43V 9.7V
R33 1200 7.6V 8V
R34 4700 8.5V 7.6V
R35 10K 7.6V 0V
R36 1K 9.1V 8.6V
R37 4.7 8.6V 8.5V
R38 10K 1% 8.5V -.6V
R39 8.25K 1% -8.5V -.6V
R40 18K -42V -9V
R41 1K -42V -42V
R42 4.7 -8.5V -8.5V
R43 510 109mV 109mV
C1 0.005 5.7V 0V
C2 0.33 5.7V 38mV
C3 1.0 35V 20.6V 0V
C4 33 15V -1.9V 0V
C5 27 35V 11.4V -26V
C6 27 35V -26V -33V
C7 47 PF -33V -33V
C8 0.005 -33V 0V
C9 47 PF 42V 9.7V
C10 47 PF 9.7V 7V
C11 33 15V 7V 0V
C12 33 15V 8.5V 0V
C13 47 PF -9.2V -.6V
C14 47 PF -9.2V -42V
C15 10 15V -.6V -8.5V
C16 33 15V -8.5V 0V
CR1 1N3193 -25V -25V
CR2 1N3193 -25V -26V
CR3 1N3754 -26V -26V
CR4 FD111 -25V -256mV
CR5 FD111 -26V -27V
Q1 PET8002 B .6V C5.7V E 0V
Q2 2N4249 B 38mV C-33V E .6V
Q3 2N4249 B -2.3V C -34V E .6V
Q4 38735 B -33V C -33V E -34V
Q5 38736 B-25V C 42V E -25V
Q6 38737 B -26V C -42V -26V
Q7 2N3567 B -26V C -256mV -26V
Q8 2N3638 B -26V C -26V E -26V
Q9 PET8002 B 7.6V C 9.2V E 7.5V
Q10 CR10 SZ51218 7V 0V
Q11 2N3567 B -9.7V C 43V E 9.1V
Q12 36892 B 9.1V C 43V E 8.6V
Q13 PET8002 B 8.6V C 9.7V E 8.5V
Q14 2N3638 B .6V C 0V E -9.2V
Q15 2N3638 B -9.2V C -42V E -8.6V
Q16 36892 B -42V C -8.6V E -42V
Q17 2N3638 B -8.6V C -9.2V E -8.5V
Q18 36892 B -20V C 42V E -20V
Q19 36892 B -21V C -42V E -43V
stevem Messages: 4738 Registered: June 2004 Location: NY
Senior Member
How are you testing these Transistors ?
If you are using just a ohm meter test then that does not rule out that they maybe braking down when powered up.
I would try replacing CR5 with a common 1N1007 or even a 1N1004 will cut it.
You may have figured the problem out already. The voltages around Q4 are off, especially at the collector. The collector should have about -3 volts there and not the -33 volts that you have.
That would lead me to believe that the transistor is breaking down when hit with the full voltage of the circuit or that C7 is shorted. As you don't have the correct replacement for Q4, remove C7 and see what happens.
As for the CR4 question, that diode is part of the limiter circuit. That -25 volts is being blocked by the diode and with a negative voltage being applied to Q7 base and emitter, I would expect the transistor to be turned fully off, preventing any voltage from showing up at the collector.
The power amp does not need the limiter circuit there to operate, so you could remove both CR4 and CR5 from the circuit for testing.
Yes, there is a way to reuse the heatsinks, but it is not a simple job.
For testing, you don't need to have the heatsinks on the replacement transistors. The amp will need the heatsinks if you drive the amp hard and for an extended period of time.
So the cap was shorted. Good to know that you found the problem.
So let's see, have you reinstalled all of the original transistors again? Is the amp keeping fairly stable and not going into failure mode? How does the amp sound?
If the amp is sounding good and staying stable, I'd just replace the bad cap and call it good. The cap is a ceramic cap 47 pf (picofarad) and rated at 100 volts or higher.
The fluctuating 1/4 volt on the output is fairly normal and may go away or lessen once C7 is replaced.
stevem Messages: 4738 Registered: June 2004 Location: NY
Senior Member
Mouser electronics part number 532-322505B00.
Once you install these place a dab of silicone sealer down in the hole to keep it held onto the Transistor.
Okay well the new part for C7 came in. I got 47p 100V 10% from newark.
I also reinstalled all of the old components and installed this new capacitor to replace c7. We are indeed stable and it is no longer going into fail mode.
However I just noticed in my meddling that I broke the leads off of CR3. I was warned against this and now I've done it.
It's a 1N3754.
Electronic parts outlet in the Netherlands has some for E7.50 each.
Do not operate the amp without atleast a 1n1004 or 7 in its place .
E7.50, Ouch!
A long time ago I started to test other diodes to use as a replacement for these diodes. Generally little glass diodes like 1N914 or 1N4148s seemed to work the best. The problem comes in mounting them in a way to get them to react to the heat of the output transistors.
If you don't want to buy the expensive NOS one, get a 1N914 type and solder two thin wires to the ends. Heat shrink the connections so that there is no chance of shorting out to the heat sink. Then using as thin a bead of silicon glue as you can, glue the diode to the center of the heat sink near the clip that held the original diode. Then run the wires to the pc board making sure that you keep the correct polarity.
This diode string ties the lower half of the power amp to the upper half. If the diode string breaks circuit, the upper half of the circuit turns fully on and will cause the power amp transistors to fail. So don't run the amp without the diode in place.
Okay I took the 1n914 and added some wire and heat shrunk everything so it wouldn't touch the heat sink. Then I placed the 1n914 in the same bracket that the original CR3 was in. I laid it in there with it kind of looped around it as seen in the attached photo.
Now when I turn it on I don't get the dc hum any more but I get a ton of static.
stevem Messages: 4738 Registered: June 2004 Location: NY
Senior Member
The 2 blue wires on the left end of the output driver board are the audio inputs from each channel / preamp so if you remove them one at a time you can then narrow down if it's one of the channel preamps , or the output driver board
Thank you for the offer. I did find two on ebay for $10. I've paid for them and just waiting on them to ship. I'm hoping to get back to this some this weekend.