Theo Markettos wrote:
Thanks, I think I'll do that when I've had a chance to buy the caps (there
are some odd values like 180uF 400V and 1200uF 35V). There's some singeing
of the PSU PCB on the primary side but all the joints look OK... but I can't
measure the resistors near the singeing until I get my multimeter sorted
(see Fluke 87 thread).

I've now replaced all the caps and it's on soak test - seems to be working
fine (this post hereby constitutes my 'tempting fate' test). There wasn't
anything obviously wrong with any of them - I ought to make myself an ESR
meter sometime to test this sort of thing.

One thing I wondered - there were two 1200uF 35V 105degC electrolytics in
parallel after the coil acting as smoothing of the +5V output. Since they
were in parallel and that value is a bit tricky to get (I've not heard of
electrolytics in E12 values before) I replaced both with a single 3300uF 35V
105degC which fits (just about) in the same space. But I wondered if there
was a thermal reason for this - would having two caps in parallel cause them
to heat up less and so last longer?

Theo