The Cyber-Spy.Com Usenet Archive Feeds Directly From The Open And Publicly Available Newsgroup Sci.Electronics.Design
This Group And Thousands Of Others Are Available On Most IS NNTP News Servers On Port 119.
Cyber-Spy.Com Is NOT Responsible For Any Topic, Opinions Or Content Posted To This Or Any Other Newsgroup. This Web Archive Of The Newsgroup And Posts Are For Informational Purposes Only.
From: "John Jardine"
Newsgroups: sci.electronics.design
Subject: Re: 2.5kW-3kW-SMPS-(cheap)-suggestions?
Date: Wed, 18 Dec 2002 18:52:05 -0000
Message-ID:
References: <20021216172355.11044.00000073@mb-mc.aol.com><20021216173636.11044.00000074@mb-mc.aol.com>
NNTP-Posting-Host: 217.135.181.139
NNTP-Posting-Date: 18 Dec 2002 20:39:30 GMT
X-Newsreader: Microsoft Outlook Express 5.00.2615.200
John Woodgate wrote in message
news:mjXsOWALlBA+Ew2S@jmwa.demon.co.uk...
> I read in sci.electronics.design that John Jardine
> wrote (in k>) about '2.5kW-3kW-SMPS-(cheap)-suggestions?', on Wed, 18 Dec 2002:
>
> >Mixing and matching (design!) of these 10 variables (there are more) for
> >each product worldwide would be very, very, difficult without the
massively
> >helpful extra degree of freedom that a variable input voltage (say) can
> >give. About 3 or 4 of the constraints are sortable as a matrix of
> >simultaneous equations but that's as far as it goes. The process
variables
> >have only recently been able to be simulated by FE methods.
>
> Sorry, I don't buy that explanation. There are enough independent
> variables to optimise design with a fixed voltage.
> --
> Regards, John Woodgate, OOO - Own Opinions Only.
http://www.jmwa.demon.co.uk
> Interested in professional sound reinforcement and distribution? Then go
to
> http://www.isce.org.uk
> PLEASE do NOT copy news posts to me by E-MAIL!
Yes indeed!, there are enough. I even had fun writing a simple prog' to
optimise on any one electrical aspect.
But ... The physical and electrical variables are *not* independant and have
also wildly different 'weights' attached to them. These weights are the cost
penalties for use.
The best solution is (usually) aimed for as the cheapest solution. So we're
looking for lowest cost as a minima with all (say) 10 variables contributing
*non-linear* cost penalties. Each variable effects each other variable. (I
suppose in a way the process has similarities to that involved in a mixer
diode).
We're now on D.O.E. Ranking, 'Variation & Co-Variation' turf. For example
... Assuming some known power is wanted for a known time (ah! but how is
this 'known' in the first place?). The electrical variables are juggled and
we optimise on a decent 110V element based on 5 metres of Nickel/Manganese
resistance wire. This nails down the electrical supply aspects and also a
few of the big physical ones such as adequate heating area coverage and
chemical resistance. So a couple of miles of it are bought and a test phase
begins.
But ... NiMn has a +5000 ppm tempco. Such that as the heating cycle
proceeds, the wire is taking less and less power as it gets hotter. This has
serious knock-on effects with the plastic's dynamic heat-flows sufficient
that this wire is useless. Mechanical re-engineering is now needed to make
use of this particular wire. Cost of wire 50p, cost of
(re)design/engineering based on using that optimum 110V wire, £50,000.
Unless spec' standards dictate otherwise, the wire voltage cannot be
allowed to be the boss factor. The reality is that this wire's tempco power
drop off is allowed for, by heating it at a higher voltage and a different
time. The electrical re-spec' and test can be established at a fraction of
the mechanical engineering costs. This results in oddball electrical
heating equipments.
Did you buy this one? ;-)
regards
john
Go Back To The Cyber-Spy.Com Usenet Web Archive Index Of The sci.electronics.design Newsgroup
|