I'm missing something here - but what?

[Mick Bailey]

Just spent much of the day troubleshooting a Trace Elliot 100w combo that had blown output transistors. I replaced them and bought the amp up using the light bulb trick with speakers disconnected and it all looked pretty fair, but with a little more DC offset than ideal - 140mv. With the speakers connected the TIP35C conducts heavily and flattens the +30v supply rail (still with a series bulb, so no surprise).

I tested all the transistors OK, but replaced TR3, TR4, TR12 to be on the safe side (as denoted in the attached schematic, but marked differently in the mono version of the amp). I also isolated the preamp.

What I can't see is why the load is causing TR13 to conduct. Any thoughts?

[gbono]

R22 tests okay? Do both supply rails check outwith loading? What are the voltages around the driver and output transistors?

[jmaf]

What I can't see is why the load is causing TR13 to conduct. Any thoughts?

When TR15 is off, there is no path to ground, assuming TR15 is good. So when you add the load it's draining it. That speaker is probably getting the full +35V DC through it... I'd start with a dummy load and go testing from TR13 back...what bias voltage is reaching it off TR12's emitter follower? What bias is reaching the TR12 base?

Use a dummy load and go back from TR13, one of those transistors is bad. What I've learned from the solid state experts here is that they almost never go alone, there's probably another shorted transistor there somewhere.

[gbono]

Double check type and pin out of replacement transistors.

[Steve Conner]

Some amp circuits will do that on the light bulb limiter when a load is connected. If you like, they get stuck halfway through the turn-on thump, and never make it to the intended operating point.

Try bringing up the power and then connecting the speaker. Also check the speaker isn't shorted.

But yes, there may be another blown semiconductor in there.

[J M Fahey]

1) start by doing what Steve suggests, turning amp with bulb limiter on, but no speaker.
Then connect speaker.
2) if still bad, please post and measure the following voltages (with amp on, no speaker, bulb limiter always in circuit until everything cleared)
a) Vbe for Q12/13/14/15
b) offset voltage, no signal, all controls on 0.
c) voltage to ground at Tr1/2 bases
d) end to end voltage from D5 anode to D7 cathode.
I´m asking for 8 voltages in total.
Thanks.
PS: I guess you already checked R22/35/36, plus "mechanical" stuff such as whether TR13/15 collectors are insulated from ground or heatsink, proper soldering, no bridges, no cracked tracks, etc.

[Mick Bailey]

Steve had got it bang-on.

The repair was OK all along, but this amp won't run correctly off a bulb limiter at start-up and with a load connected - it screws up the relative voltages and raises VBE for TR13 to 725mv and it just sits there. Plugging the load in once it's powered up works. I haven't come across this before - lost quite a few hours chasing nothing. I guess learning the hard way makes it stick. Thanks, guys - much appreciated.