Ampeg SVT-6 Pro No Output

[Tom Phillips]

It's not just that the insulators can fail by voltage punch-through. If they are reused (especially multiple times) they will no longer efficiently transfer the thermal energy between the FET and the heat sink.

[g1]

So are you using a limiter lamp or not?

If you have burnt a source resistor and some fets, you are not measuring the source resistor voltages correctly during slow power up. (post #231)
It's not possible to blow a source resistor till there is significant voltage across it. Definitely much more than 25 millivolts (.025V) DC.

[ca7922303]

So are you using a limiter lamp or not?

If you have burnt a source resistor and some fets, you are not measuring the source resistor voltages correctly during slow power up. (post #231)
It's not possible to blow a source resistor till there is significant voltage across it. Definitely much more than 25 millivolts (.025V) DC.

Using a power analyzer along with the variac. I've been using the same lugs where the heat sink attaches to the upper and lower boards. Previously was getting +/--65v at all the drains at full power when tested before. Just got 9v on all drains. Not sure how ground was different.

[ca7922303]

It's not just that the insulators can fail by voltage punch-through. If they are reused (especially multiple times) they will no longer efficiently transfer the thermal energy between the FET and the heat sink.

Thanks, seems like it could be causing some of the issues occurring.

[ca7922303]

So while I'm waiting on insulator strips and two(2) mosfets, what do you think of me disconnecting the two(2) mosfet boards and making sure that correct voltages are going to mosfet boards when parts arrive? I suspect I have some more parts that are bad somewhere.

[g1]

Yes, if you can completely disconnect the power amp boards, do that and check the voltages in the power supply.

[ca7922303]

Yes, if you can completely disconnect the power amp boards, do that and check the voltages in the power supply.

Removed J30, 37, 40, 31, 38, 41, 47, 48. Ch 1 - completely disconnected. With variac@55vac, Ch1+ still reading P Channel-Source(7mv)Drain(1mv)Gate(-0.5v)
N Channel-Source(7mv) Drain (7mv)Gate(3.8v).
Went ahead and measured voltages on the power supply and all test points were correct taking into account I was at 55vac(doubled readings).

[ca7922303]

Yes, if you can completely disconnect the power amp boards, do that and check the voltages in the power supply.

What do you make of this? looked at the website where you bought the MosFET's. The only thing I found there, which had to be toggled to get additional information on what they had to offer....they made NO MENTION of the parts being Matched for Transconduction...which translates to yielding equal Current sharing which is needed on these Ampeg SVT-series MosFET Output Stage amps. So, we have NO idea what to expect. That's why I wanted to have you slowly turn up the Bias so we could start seeing them conduct, and get readings without any going way high in current (voltage across the 0.47 ohm Source resistors) while waiting for those not yet turning on to see where we are. Not dissimilar to setting bias on power tubes, and matching them.
I've bought 28 mosfets from same source and I'm wondering if buying two more because they claim to be closely matched is going to resolve issues?

[ca7922303]

Guy I have bought these from says he followed this to make tester for matching mosfets: https://www.passdiy.com/project/arti...tching-devices

[g1]

The matching is something to check after everything else is up and running normally. With bias all the way down, a properly running amp shouldn't get damaged by unmatched fets, and the fets should not get harmed either.

You had no power supply getting to the drains when the amps were all connected. Or the supplies were shorted out by the amp boards, but I would expect that to blow a fuse or cause other issues, or show up on your power analyzer.
If you were getting half voltage at supply rails with 55VAC line voltage, that same half supply voltage should show up at drains when amps are hooked up.
But start up slow again, say 20VAC. Drains should have same voltage as you measure at +/-65 rails in supply, whatever that may be.
If it's not the same at the drains as the supplies, find out why.

If drains are same as supply rails, check each source resistor. Slowly ramp up voltage in steps, checking all source resistors each time you increase variac. Do not allow any source resistor to get to more than 10mVDC for now.

[ca7922303]

The matching is something to check after everything else is up and running normally. With bias all the way down, a properly running amp shouldn't get damaged by unmatched fets, and the fets should not get harmed either.

You had no power supply getting to the drains when the amps were all connected. Or the supplies were shorted out by the amp boards, but I would expect that to blow a fuse or cause other issues, or show up on your power analyzer.
If you were getting half voltage at supply rails with 55VAC line voltage, that same half supply voltage should show up at drains when amps are hooked up.
But start up slow again, say 20VAC. Drains should have same voltage as you measure at +/-65 rails in supply, whatever that may be.
If it's not the same at the drains as the supplies, find out why.

If drains are same as supply rails, check each source resistor. Slowly ramp up voltage in steps, checking all source resistors each time you increase variac. Do not allow any source resistor to get to more than 10mVDC for now.

Tested Ch 1- board only. It has all components and reads consistent before voltage applied.
Variac @ 20vac N Channel Source(-152mv) Drain(-11.6v) Gates(-2.9v).
P Channel Source(1@-90mv, 4@-152mv)Drain(+11.6v)Gates(+0.9v)

Variac @ 30vac N Channel Source(-1v) Drains (-17v) Gates (-4v)
P Channel Source(-1v) Drains(+17v)Gates(+150mv).

[ca7922303]

Variac @ 55vac N Ch S(-3.5v) D(-32v)G(-6.3v)
P Ch S(-3.5v) D(+32v)G(-2v)

[ca7922303]

Found a previous 6 pro post with suggested test points. With variac @ 55vac Q3/Q4 C/B/E(+3.7v/-0.5v)(+42v/+42v)(-42.7v/-42.7v)
Q18/Q19 C(+42v/-42v)
IC1 Pin #7(0.5v)
Q22 B/C/E(-6v/-5v/-4v)
R16(10mv/3.8v) R17(-0.5v/10mv)
D4(3.8v/-75mv) D5(-75mv/6mv) D6(6mv/-293mv) D7(-309mv/-0.5v)
Q16(0.5v/-5mv/0.5mv) Q17(1mv/56mv/33v)
Q2(-2mv/37mv/1.5v)
R14(43v/43v) D2(45v/43v) R10(1.5v/3.8v)
R22(42v/32v) D33(-3v/-6v) R148(-3v/-3v)
I believe post suggested checking current on R10,(0.5vac/20vac). The 20vac I'm getting is going from 0 to 20vac and back to 0.

[g1]

Found a previous 6 pro post with suggested test points. With variac @ 55vac Q3/Q4 C/B/E(+3.7v/-0.5v)(+42v/+42v)(-42.7v/-42.7v)
Q18/Q19 C(+42v/-42v)
IC1 Pin #7(0.5v)
Q22 B/C/E(-6v/-5v/-4v)
R16(10mv/3.8v) R17(-0.5v/10mv)
D4(3.8v/-75mv) D5(-75mv/6mv) D6(6mv/-293mv) D7(-309mv/-0.5v)
Q16(0.5v/-5mv/0.5mv) Q17(1mv/56mv/33v)
Q2(-2mv/37mv/1.5v)
R14(43v/43v) D2(45v/43v) R10(1.5v/3.8v)
R22(42v/32v) D33(-3v/-6v) R148(-3v/-3v)
I believe post suggested checking current on R10,(0.5vac/20vac). The 20vac I'm getting is going from 0 to 20vac and back to 0.

Can't follow anything the way this is written. Sometimes CBE, sometimes BCE, sometimes not even stated.
Put each transistor on it's own line with e, b, and c and polarities noted each time, like this:
Q22 B(-6V) C(-5V) E(-4V)

[ca7922303]

Found a previous 6 pro post with suggested test points. With variac @ 55vac Q3/Q4 C/B/E(+3.7v/-0.5v)(+42v/+42v)(-42.7v/-42.7v)
Q18/Q19 C(+42v/-42v)
IC1 Pin #7(0.5v)
Q22 B/C/E(-6v/-5v/-4v)
R16(10mv/3.8v) R17(-0.5v/10mv)
D4(3.8v/-75mv) D5(-75mv/6mv) D6(6mv/-293mv) D7(-309mv/-0.5v)
Q16(0.5v/-5mv/0.5mv) Q17(1mv/56mv/33v)
Q2(-2mv/37mv/1.5v)
R14(43v/43v) D2(45v/43v) R10(1.5v/3.8v)
R22(42v/32v) D33(-3v/-6v) R148(-3v/-3v)
I believe post suggested checking current on R10,(0.5vac/20vac). The 20vac I'm getting is going from 0 to 20vac and back to 0.

Q3 C(+3.7v) B(+42v) E(-42.7v)
Q4 C(-0.5v) B(+42v) E(-42.7v)
Q18 C(+42v)
Q19 C(-42v)
Q2 E(-2mv) B(37mv) C (1.5v)
Q22 E(-6v) B(-5v) C(-4v)
Q16 B(0.5v) C(-5mv) E(0.5mv)
Q17 E(1mv) B(56mv) C(33v)