ESP Project 23

lohk said:
No, I think this is no good choice.

Really? This is not a circuit in the signal path, so "quality" is not really an issue as long as the circuit is still operational. Do you think it would not work correctly?

I should have mentioned that my rails are 42, so the max "normal" voltage the transistors should see is 39v, with a worst case (output shorted to unloaded rail) of 50v. I believe this is well within the 60v safe region. [output shorted to rail means I have bigger fish to fry, but I'd like to avoid secondary explosions in a fault state]

I would email Rod, but because he is so helpful to so many people I like to avoid bothering him where possible.

Thanks for your responses.
No, it is not a question of "quality".
MJE3055/2955s are designated output high current transistors, they need to be driven properly to function in an apropriate way. You can use them for input devices in sensing circuits for example, when high currents have to pass through. They have only low gain, and are quite non-linear at low currents, which can also be a problem here. And they are much bigger and more difficult to implement, and...
I think that they will not work the way you want. But if you insist, you can try it nevertheless ! It is up to you.
You can use - as Rod suggests - any transistor which meets the voltage specs. Current will be low, so possibly BC546/BC556 will work as an alternative. Boy, I have lots of these lying around ! (A zip-loc bag full of them actually...)

best wishes