On 28/01/2015 19:54, Ralph Mowery wrote:
"Peter Able" stuck@home wrote in message
o.uk...
On 27/01/2015 17:41, gareth wrote:
.

Computer power supplies become dated so quickly - change of specs / change
of connector styles - that these caps are likely to have a fair bit of
life left in them - although they may over-smooth in a design meant for
the old combined 8+32uF - red spot for the 8 - that used to be the bedrock
of valve receiver PSUs.


When not using the origional value of capacitor in the power supply it is
important not to go too large. If the supply has a tube rectifier, the tube
will have a maximum capacitor rating. Too large of a value is bad for the
tube. If way too large and simiconductors are used for rectifiers, they can
become a problem also. Mainlly blown fuses when the supply is turned on
unless a soft start is included. That is usually a resistor in the primary
line that is shorted out after a second or so.


Hence my warning about a design specifying 8uF, Ralph. What will narrow
the gap will be that several ex-SMPSU capacitors will be used in series
- so the effective will be less than the individual capacitance.

As for limiting inrush, the transformer secondary and rectifier
impedances will, hopefully, take care of that - but as you write, these
things should be considered.

I wonder how much carnage was caused by those rectifier replacements for
valves which were a set of semiconductor rectifiers potted in an
International Octal body plug? Plug-in replacements - like hell !!

As for fusing - I guess that one can still get slow-blow fuses?

PA