From: John Tserkezis 
Reply-To: Fidonet AVtech Echo 

Bob Lawrence wrote:
> JT> Actually, increasing turns would probably make it worse unless
> JT> it's the extra winding impedance you're relying on?
> 
> JB> That had me wondering too, but I don;t really understand power
> JB> transformer design, turn ratios are the limit of ability to
> JB> design transformers. 

  Do'h.  A higher inductance might have something to do with it as well...

>  Iron saturates rather quickly. For a given core size, more turns
> means less flux density and less iron losses (less saturation). The
> field coil in a synchronous motor is the same.  Of course, the power
> is reduced, but iron-loss drops a lot faster than power. For 10% less
> power you can drop losses, and motor-noise, by half.

>  Most motors (and transformers) are run in iron-saturation using
> temperature rise as the limit (60C max). It's an art to screw the
> maximum output for minimum cost, and a beginner often crosses the
> line.

  That's what I was going it say- it's a fine line.

>  A transformer is not easy to design. First you pick a core, then the
> turns, the turns ratio, then wind it up to see that the wire fits in
> the hole, load it up to make sure the volts are right, and run it to
> measure temperature rise...

>  It's too big a fiddle... but it is possible to design the whole
> thing on paper, temperature rise and all, and get it right first time.
> If you try to fiddle it, you'll never get it right.
> 
>  SMPS are even worse, because wire-losses are almost unpredictable.  

  They're scary.  The cheapest supplies are designed by geniuses.  Since the 
highest single component cost is the transformer, they make that as cheap as 
possible, and design around it.
  They could do away with all the protection, spike and noise suppression and 
quite a lot of other problems if they used a good transformer.  You'd end up 
with a text-book example power supply that way- but it would cost a mint- all 
due to the transformer.

-- 
       -o)
       /\\    Message void if penguin violated
      _\_V    Don't mess with the penguin

Linux Registered User # 302622                         http://counter.li.org>
Fido: 3:712/610  BBS/FAX: +61-2-9716-8310  Internet: jt{at}techniciansyndrome.org

--- ifmail v.2.15