Marshall 8080 sticks to -ve rail on startup

[Alex R]

Hi all, long time no see. I have a VS8080 in the shop and the output sticks, humming, to the negative rail for a few seconds on startup, then with a pop it settles and works fine. I put two new main filter caps on it and checked all the semiconductors and they give me beeps, can't find any open resistors. Before I start checking component by component is there anything you can tell me to help focus the search? Am I actually likely to be looking at a leaky TR9 or 10, for instance? TR10 sees most of the -ve rail voltage on its base whilst it's sticking, as I guess you'd expect, I'm wondering whether the poor thing is just strained. If so, any ideas for a replacement pair? I stock 2STA1962/2STC5242, these should be ok?

Schematic attached. Thanks very much in advance.8080___8100_Power_stage.pdf

[Jazz P Bass]

While TR10 'may' be strained, try to narrow it down.
C47, if leaking could put the -V voltage on TR10 base.
Temporarily remove the cap & see if that helps.
A bad TR4 (MPSA06) could also do it.

[g1]

How about the +/-15V rails. Are they coming up evenly?

[Alex R]

Thanks for that g-one - what are you thinking about the opamp rails?

Jazz P, I lifted the cap and it seemed to stop it so I changed C47 and TR4... but it didn't cure it, I think I fooled myself by forgetting to discharge the caps fully, as if they're not totally discharged it doesn't stick when you turn it back on.

I now have at last noticed that the outputs are darlingtons.

[Alex R]

I changed the outputs too, as well as C47 and TR4 - still no change. Any more ideas why this should stick to the -ve rail for a few seconds on startup?

[Steve Conner]

Have you tried replacing C20 and/or C55? If the amp develops a negative DC offset, these get reverse biased.

And if they were leaky, it could upset the DC conditions and make the DC offset worse, which would make the leakage worse and so on in a vicious circle.

I mention this because you're stuck to the negative rail, which is an extreme example of a negative DC offset.

I'd also check any of the small signal transistors that run hot, as I've seen these develop intermittent opens before. Also check R62 and any other large resistors for bad solder joints.

[Enzo]

I too would like you to WATCH the 15v rails while this happens.

You don;t have leaky outputs, they won;t heal themselves. And TR9 is the bias transistor. It it shorts or leaks, then the amp will run cooler with more crossover distortion. It won;t make DC on the output. If TR9 opens or can't conduct enough, then the amp will turn both sides of the outputs on and it will blow fuses, not go to DC.

I'd say your amp is being told to go to DC. Note the TL072 runs on +/-15vDC. If those rails do not come up together, it will be off to one side. SO check to see if both supplies come up instantly or if one side lags. Then too, the 15v rails are used to bias TR5,6. Again, any imbalance there will cause offset.

Steve might have a point with the small transistors. MPSA06 and MPSA56 are about 22 cents each, so for a dollar you can replace all four of them.

[Alex R]

Thanks Enzo and Steve.

I replaced c20 and c55, the result is that it sticks for a shorter time now, just under a second. Still a noisy buzz whilst it does so, though.

The opamp supply rails settle at just over 14v each. The +ve rail comes straight up, whilst the negative rail sits at around 12.5v whilst it sticks to the rail, then pops up instantly when the buzz stops. I'm wondering whether this is a cause or an effect?

The amp only does this when the caps are fully discharged, so if I try again within a few minutes without discharging the caps it doesn't stick.

OK gonna swap a few transistors. Thanks again for the help guys.

[Alex R]

replaced tr11 and c40; no change... lunchtime!

[Steve Conner]

When feedback loops are involved, it gets hard to tell the difference between cause and effect.

Did you check C40 and C41, the bypass caps for the 15V rails? Maybe one of them is bad and the difference in capacitance is causing the imbalance.

[Jazz P Bass]

Can you monitor the bases of TR5 & 6 while it latches.
My take is one is coming on early.
Something is turning on TR 4, which is turning on TR 10.

[Enzo]

When the negative rail is low, the amp slams over to DC. MY prime suspect just intantly became the -15v zener.

[Alex R]

When the negative rail is low, the amp slams over to DC. MY prime suspect just intantly became the -15v zener.

YAY new pair of zeners cured it! Many thanks to all and to Enzo!

[Enzo]

Great! Happy birthday then.