Calculations for a choke on the output of a SMPS

[Walrus]

It's been a long time since I was at college, and I've forgotten more than I learnt! So as per the subject, how does one
go about calculating the value of a choke to put on the output of a switched mode PSU to eliminate ripple.

I've been playing around with an XP Power ECM100US15 http://www.xppower.com/pdfs/SF_ECM40-100.pdf
and using a choke I had around (value unknown), followed by a 3300uF cap, can get a good'ish reduction in HF rubbish, but then weirdly seem to end up
with a sawtooth type waveform (5mV p-p) akin to normal analogue psu ripple but at a repitition rate of 100Hz??
Switching frequency is about 70kHz according to datasheet.

 

[Walrus]

Anyone....??

 

[shabtek]

what is the load?

 

[Walrus]

An old API desk.  ;D

16.5 Volts, up to 6 Amps per supply.

 

[audiomixer]

the first thing that happens in the SMPS is the rectification of the mains - that might introduce some 100Hz ripple. while isolated from the primary side you might still get some influence into the secondary. check what this looks like, primary side - but caution, that would be several 100V's and no galvanic isolation to the line! so take extra measures not to kill yourself, and the scope is ground referenced... or double the storage cap test-wise.

- Michael