This looks better, what you had in your first post was a C-E shorted VAS transistor pulling
the drive to the output close to the negative rail. By taking it out we allow the resistors R718
and R720 to pull it to the positive rail suggesting that nothing is shorted/failed in the drivers
or output stage. I've added some comments along side your voltages above.
The driver B-E voltages look high, can you measure across rather than with respect to
ground as a sanity check. Obviously with respect to ground should work.
Left diff transistor Vbe = .9 - .284 = .6 which is a normal Vbe drop. The other is reverse
biased and current limited by the 33K resistor.
The output is not loaded correct?
Do you have a test oscillator and scope - not essential wondering?
Last edited:
The next test to get the output more stable would be to put 3.9K+5.6K = 9.5K tacked
to the C - E pins of Q704 location - this should bring the output close to zero volts but
without feedback so just ballpark.
The drivers might be blown but this is another sanity check.
Voltages ?
to the C - E pins of Q704 location - this should bring the output close to zero volts but
without feedback so just ballpark.
The drivers might be blown but this is another sanity check.
Voltages ?
Hi,
Ok, so next thing would be to solder 9,5 kohms between the collector and emitter-connections of the now empty Q704-location?
And measure all the voltages of the two drivers?
And still short C-E on the pre-driver as before?
// Lasse
Ok, so next thing would be to solder 9,5 kohms between the collector and emitter-connections of the now empty Q704-location?
And measure all the voltages of the two drivers?
And still short C-E on the pre-driver as before?
// Lasse
Seems that Pete has things under control.
For some time I've wondered about RCA replacement transistors. Seems you may need at least one.
I guess your PCB uses TO-5 (TO-39) metal cans for all but the diff pair and output? These seem to be going out of fashion. Some devices that are still around last time I looked were the 2N5680/2N5682 which may work for the drivers and VAS.
If suitable TO-5 cans cannot be found then the alternatives might be BD139-BD140.
Originally they had a BVceo of 80 and BVcbo of 100 so should work for 70W amp, but it would be a good idea to check the BVceo of any devices if you go that route as not all manufacturers show the 100V spec.
Otherwise it is probably a case of having to use TO-220 drivers, but there are several types which might be suitable.
The output types if still working should be used as they won't be seen again, but good alternatives are the MJ15003/4 and you could make the output fully complementary.
The diff pair could, at worst, be made from two discretes without too much problem.
For some time I've wondered about RCA replacement transistors. Seems you may need at least one.
I guess your PCB uses TO-5 (TO-39) metal cans for all but the diff pair and output? These seem to be going out of fashion. Some devices that are still around last time I looked were the 2N5680/2N5682 which may work for the drivers and VAS.
If suitable TO-5 cans cannot be found then the alternatives might be BD139-BD140.
Originally they had a BVceo of 80 and BVcbo of 100 so should work for 70W amp, but it would be a good idea to check the BVceo of any devices if you go that route as not all manufacturers show the 100V spec.
Otherwise it is probably a case of having to use TO-220 drivers, but there are several types which might be suitable.
The output types if still working should be used as they won't be seen again, but good alternatives are the MJ15003/4 and you could make the output fully complementary.
The diff pair could, at worst, be made from two discretes without too much problem.
Well, now it has a light bulb in series with the mains.
And in order to make 9;5 kohms,
I bought a box with assorted resistors. 1460 pieces!
Should
I proceed as of post #25?
Regards// Lasse
And in order to make 9;5 kohms,
I bought a box with assorted resistors. 1460 pieces!
Should
I proceed as of post #25?
Regards// Lasse
Thought you might have them or a 10K in your junk box, sorry about that.
Yes sure go ahead.
Does your amp have the single can diff pair, they switched to individuals at
some point?
The output is not loaded correct?
Do you have a test oscillator and scope - not essential wondering?
Do you have dummy 8 ohm loads or any at all?
A high bias amp might have issues with the lamp, this is certainly not a high bias amp.
Yes sure go ahead.
Does your amp have the single can diff pair, they switched to individuals at
some point?
The output is not loaded correct?
Do you have a test oscillator and scope - not essential wondering?
Do you have dummy 8 ohm loads or any at all?
A high bias amp might have issues with the lamp, this is certainly not a high bias amp.
Last edited:
The C-E short goes to the transistor Q708 with the Bias designation in the schematic,
today it is called a Vbe multiplier.
Yes 9.5K at the empty Q704 location.
Enjoy, you are making good progress.
A 25W resistor would be fine as long as you only bring it up to 60W for a few seconds.
75 - 100 is better if you want to see long term 60W out.
A 25W resistor would be fine as long as you only bring it up to 60W for a few seconds.
75 - 100 is better if you want to see long term 60W out.
Something strange is that the voltages did not look right in the last
test, but they do more or less now.
The output should be close to zero, 10s of mV is good.
Emitters of drivers should be close to the output as they are.
The top driver base should be .5-.6V above the output as it is,
but the lower driver Q706 should be negative relative to ground,
did you drop the sign?
I'll guess that the diff pair emitters are at .5 ish volts - actual?
Voltage across 390 collector load?
Have you worked before with a signal generator and scope on an amp?
Do you want to try it on the good channel.
First warning, on an old amp like this with slow outputs do not drive with
full power sine waves above 5 KHz. Do not try full power square waves.
test, but they do more or less now.
The output should be close to zero, 10s of mV is good.
Emitters of drivers should be close to the output as they are.
The top driver base should be .5-.6V above the output as it is,
but the lower driver Q706 should be negative relative to ground,
did you drop the sign?
I'll guess that the diff pair emitters are at .5 ish volts - actual?
Voltage across 390 collector load?
Have you worked before with a signal generator and scope on an amp?
Do you want to try it on the good channel.
First warning, on an old amp like this with slow outputs do not drive with
full power sine waves above 5 KHz. Do not try full power square waves.
Last edited:
Current to diff pair emitters is (36.1 - .76)/8.2K = 4.3 mA
Current through 390 is ,94/390 = 2.4 mA or about half of the tail current, this looks normal.
Things looks reasonable, I'd remove the C-E jumper on the Bias transistor and see if
you can adjust the output stage bias. The lamp will help to protect things if you bias
it too high.
Current through 390 is ,94/390 = 2.4 mA or about half of the tail current, this looks normal.
Things looks reasonable, I'd remove the C-E jumper on the Bias transistor and see if
you can adjust the output stage bias. The lamp will help to protect things if you bias
it too high.
Note the test points TP-1 and 2 on your schematic, with no input adjust
the pot for .008V between the test points.
I = .008/.27 = 29.6 mA
the pot for .008V between the test points.
I = .008/.27 = 29.6 mA
- Status
- Not open for further replies.