[Gllug] OT: Merging UPS outputs

Chris Hunter chrisehunter at blueyonder.co.uk
Sat Sep 3 15:56:58 UTC 2005


John Winters wrote:

> I presume you mean there are no fuses which will *guarantee* to protect
> triacs.  I've certainly used kit where the triacs blew well under half
> the time, so the fuses succeeded at least a bit.

Very rarely, in my experience - it was almost ALWAYS the triac that 
failed, leaving the fuse intact, and the channel permanently powered.

> Having said that, all the serious kit uses (or I should say used,
> because my experience is more than 20 years ago) back to back
> thyristors.  Even those aren't immune to blowing though.  A regular
> maintenance operation on Mini II racks was to replace one from a pair of
> thyristors.  The symptom was that a channel would dim from full to half,
> but no further because one of the thyristors was permanently on.
> Thyristors too need fast-blow fuses to be protected, and even those
> don't always manage it.

More modern thyristors are much more robust, and I've only seen them 
fail when seriously and persistently abused.

> I'm not sure what the current racks use, but they're mightily
> impressive.  The 36 ways of patchable Tempus racks in our local theatre
> were recently replaced with 96 ways of hard wired racks, all in a rather
> smaller space than the Tempus racks had.  When I saw where they were
> proposing to put them I said, "What about the heat?" to which the answer
> was, "There isn't any".  And amazingly, there isn't.  I don't know how
> the new versions work, but where the old racks could be used to make
> toast the new ones seem to stay stone cold all the time.

Many of the modern ones use PWM, which can be amazingly efficient.  One 
design for a PWM controller that I worked on actually had to have the 
waveshape of the output signal "spoilt" to reduce the risetime so as to 
reduce the RF noise!

As an aside, my current PWM modulators (for AM broadcasting) achieve in 
excess of 95% efficiency, and with Class E PAs, the overall transmitter 
efficiency is around 90%.  Heat is much less of a problem in modern 
electronics!

Chris



-- 
Gllug mailing list  -  Gllug at gllug.org.uk
http://lists.gllug.org.uk/mailman/listinfo/gllug




More information about the GLLUG mailing list