The transistor is not the problem.

Replacement necessary of both? No!

But both of them must be ok. The amplifier factor must fit together. There is no use of a transistor with factor 20 fitted with another one with 100. This will not work. They must form a pair with the same amplifier factors.

Just for testing and find the fault? Just replace it... and replace it later by the real one.

Next problem: did you adjust the no-load-current? A potentiometer, adjusting the current when running in standby?

No? OK, then the next transistor will fail also.

Power amps are always problematic. Because of that, use a light bulb in the powersupply... better 2x in the positiv and negativ supply. When the light goes on, you have a problem. Or you have found the problem.

Use a 5W 12V or even 2-3 in line depending on the size of the amp ... at least it will limit the smoke.

If the output stage transistors are held by a long bar with threaded brass inserts, the output
stage transistors are special in that  they have a matching, relatively large, "hole" for the mounting bar.
If that's the case PM me and I'll have a look in the dungeons.

- And check everything else in that circuit before powering up again.
Of course also check the idle current setting. Replace the trimmer, adjust and monitor the current while slowly powering up on a variac.

Martin

The output transistors are screwed into an aluminum heatsink, the normal way (see picture).

I need replacement transistors to repair this Beomaster.

The Beomaster was working normally. I was going to measure the idle current and one output transistor suddenly failed. Before this happened, the heatsink did not get warm, so I think the idle current was actually too low and definitely too high.

You might want to consider using modern, readily available transistors while you are trying to sort out your failing transistor issue. This could save you a significant amount of money versus installing rare and expensive vintage transistors. This is especially true if you use the trial and error methodology.

I keep a fairly large supply of cheap Chinese parts on hand for this purpose. I'll swap them in as part of the debugging process. Once I have the circuit working, I'll replace them with either high quality modern parts or (expensive) original vintage parts. What I do depends on my goals for the particular piece of equipment.

Glitch

Yes, but which modern transistor is a working equivalent? These are Darlington transistors, if you weren't aware of that.

A quick google search finds one person claiming success using a TIP132/TIP137 combo. I recall seeing other stories where the person used something else on a BM901.

Another tactic is to use the published cross reference charts. That transistor might be old enough that you will have to cross reference the cross references. Verify that the specs (voltage range, gain, etc.) are similar to that of the original part.

A "close enough" part won't necessarily perform exactly like the original.

I've only had to do this with final stage transistors a couple of times. Hopefully, someone with more experience, or experience with this exact situation, can chime in with better advice.

Of course, Martin's advice of replacement using the correct parts with a procedure that won't damage them is the best way (if you have the necessary equipment).

Glitch

 

Thanks for the TIP, Glitch. I compared the data sheets and these seem promising and are available at decent prices.