Larger power supplies will have diodes in parallel, and they will even put
the output trannies in parallel. You can get away with it with diodes but
you have to be carefull to use the same type, preferably from the same
batch. Otherwise you end up with some saturated and others barely turned
on,
and you get a high failure rate.

During 50 years 'in electronics', I can't immediately recall seeing any
power supplies with diodes in parallel. I'm not saying it's never done
but, if it is, there should also be some current-balancing resistance in
each path. This could be low-value resistors, or even the resistance of
the secondary windings of paralleled transformers which feed each set of
diodes. I don't think that having to use diodes from the same batch is a
very good design criterion!
--
Ian

Having diodes from the same batch is something we do at home when we make
our own klugey experimental stuff. Of course they aren't going to do this in
production equipment!

I only have 25 years against your 50, but I've seen so many things I can't
immediately recall seeing a lot of the things I've seen . At one time I
was doing the work on our power supplies, and we had a lot. And of course
there were resistors in series with each diode, the purpose of which is to
minimize variations in the operating parameters of the diodes themselves. I
never bothered to test them to see how equally they shared the load, but the
things worked. I've never had to replace one of the diodes themselves. It's
possible that the diodes would turn on such that one didn't turn on until
another approached saturation. So long as they don't burn out I guess it
would work OK. Hmm...Let's use big doides mounted in a huge heat sink
because some are going to be near saturation while others are idling. I've
seen enough klugey designes that I would not be surprised if that really was
how it worked