Randall RG80 popping noise

[nevetslab]

With your supplies at +/- 36VDC (stated early on as the buss voltages), your constant current source isn't working. The base of Q3 needs to be roughly 10V less negative than the -36V supply buss. Q3 is controlled by Q5, which has an RC network between ground and it's base C5 and R5. Check to see if Q4, Q5, D1, ZD1 & ZD2 are ok, as well as if Q3 is ok.

If the constant current source circuit was working, you'd have 1.38mA flowing thru Q3 to the diff pair Q1 & Q2. I get that by subtracting the Vbe of Q3 from the 10V zener voltage of ZD2, and divide by R8. With 1.38mA flowing up to the diff pair Q1 & Q2, each would be running on half that current, or 691uA. We've seen there's a lot more current flowing in the collector resistors of Q1 & Q2, so first, we need to get the front end correct. All else should follow once that's set up.

In a pinch, you could disconnect the R6 from the collector of Q5 and take it to ground instead. Then, if ZD2 and Q3 are ok, that will light up Q3 for the correct current to feed Q1 & Q2. We'll see what happens from there.

[pontiacpete]

ZD2 ,Q3, and Q4 are shorted. I have replacements for Q3 and ZD2 but I don't have Q4, 2SA1015. Is there a good replacement for this that I can sub in or should I order some? The Towers's International doesn't offer any suggestions.
D1 Q5 and ZD1 all look good.
The voltage on the base of Q3 is now about -37v

EDIT: I found some 1015, strike the request above

[pontiacpete]

now I'm seeing 1.6v across R10 and it's steady. To ground there's 35v and it's steady.
Should I check anything else before hooking up the output?

[nevetslab]

now I'm seeing 1.6v across R10 and it's steady. To ground there's 35v and it's steady.
Should I check anything else before hooking up the output?

Ah, now we're getting somewhere! So, you're seeing 1mA flowing thru R10/Q1. See what the voltage is across R8 (cathode resistor of Q3). Next, see what the voltage is across R11 (for the current thru Q2), then voltage across R13 & R14 (for the current thru Q6), then voltage across R15 & R16 (for the current thru Q7/Q8/Q9). And, the voltage at the collectors of Q7 & Q9. If they're still joined with R12 as the temporary output of the amp (minus the output stage). Hopefully that's near 0VDC.

[pontiacpete]

Across R10 2.2v
Across R8 there's 9.3v
Across R11 there's 0.37
Across R13 and R14 , R15 and R16 there's 0v
the buss is also at 0v

[nevetslab]

Across R10 2.2v
Across R8 there's 9.3v
Across R11 there's 0.37
Across R13 and R14 , R15 and R16 there's 0v
the buss is also at 0v

With 9.3V across R8, there's 1.38mA flowing thru Q3 as the constant current. With 2.2V across R10, virtually all the current is flowing thru Q1, and only a trickle thru Q2. Q2 is suspect. Also, with 2.2V across R10, there should be about 1.6V across R15 (34mA), but you have no voltage across R15 nor R16. R16 had burned when R12 burned. I suspect both R15 and R16 are open. There's not enough current flowing thru Q2 to turn on Q6. There should be around 1.38mA flowing thru Q6, R13 7 R14. For there being 0V on the output buss, it suggests Q2's b-e junction may still be there. Q1's b-e junction established the collector of Q3 is at -0.6V, so you'd have 0V at the buss (tied to collectors of Q7 & Q9 at the moment).

Anyway, we're making progress.

[pontiacpete]

Had a little set back here. Somehow something touch something it wasn't suppose to and R15 & 16 burned along with Q7 &9. It's tight on the board. Replaced them all and now moving on I have voltages across these resistors
R10 1.1v
R8 9.3v
R11 1v
R13 1.1v
R14 1v
R15 .5v
R16 .5v

[nevetslab]

Had a little set back here. Somehow something touch something it wasn't suppose to and R15 & 16 burned along with Q7 &9. It's tight on the board. Replaced them all and now moving on I have voltages across these resistors
R10 1.1v
R8 9.3v
R11 1v
R13 1.1v
R14 1v
R15 .5v
R16 .5v

Ah, now looking normal. 9.3V across R8 (6.8k) yields 1.37mA as constant current to feed Q1 & Q2. 1.1V across R10 (1.6k) yields 688uA, and 1V thru R11 yields 6.25uA, so the front end now looks normal. 1.1V across R13 (1.6k) yields 688uA, and 1V across R14 yields 625uA, so we have steady current thru the Q6 branch, and is sourcing base current to Q7 & Q9. With 0.5V across both R15 and R16 (47 ohms) yields a steady 10.6mA thru that string.

NOW, I think you can put the output stage back together. But first, as I believe you've restored the delay-turn-on circuit that controls Q3 constant current source, power this up and down a few times and watch the voltage across R8. The RC network C5/R5 has a time constant of 2.2 sec, so there should be a short delay as that network ramps up, turning on Q3 where you'll see the 9.3V appear, and all the rest light up and hold steady. With that power-up circuit working, I think we'll see steady current thru the output stage, and the popping should all be gone.

[pontiacpete]

That worked out exactly as you said. I monitored the voltage across R8 and watch it ramp up to 9.3v. I just counted one thousand one , one thousand.... and it came up between 2 and 3 seconds. Mind my asking how you figured the time constant for that RC network?
Anyway, no popping with the output connected. Ran the amp for more than an hour and it's holding.
Thank you so very much for the help, Nevetslab!! That was a great learning experience. Hope others find this useful.

[nevetslab]

The time constant is R x C in seconds. Though I never stopped to try and figure out just where Q3 would finally turn on. That type of constant current source is widely used, and I'm a little surprised Randall didn't add a fast-off action to it, unless it's there and I just didn't look close. The amp will normally turn on quiet, and if shutting down the current source, will turn off quietly as well.

Yes, that was a pleasure and it taught me a good bit as well...many thanks for being my 'hands and eyes' in the mechanics of thinking it all thru. My pleasure!!