[sdiy] How to predict a tansformer's current capability?

[Magnus Danielson]

From: patchell <patchell at silcom.com>
Subject: Re: [sdiy] How to predict a tansformer's current capability?
Date: Mon, 24 Dec 2001 07:49:27 -0800

>     Another good predictor for transformer capablility is weight.  The VA
> capacity is proportional to how much the transformer weighs in at.  In order to
> get a certain amount of power, you have to have so much iron.  There are
> variables, such as how the transformer is constructed...but, I don't remember
> the rules of thumb I used to use...

There is indeed more than wire-gauge to transformers. Whatever
material used as a core has saturation (you only avoid this if you use
non-atomic containing vacuum, but any non-fog containing air can do)
and this puts a limit on how much power you can transport
through. Another aspect is the eddy currents which is proportional to
the inverse square of the frequency. This is the reason why higher
frequency requires smaller cores and this is also why switching
supplies requires so low weigth even if they contain
full-transformers.

I've never heard of the actual transformer (the component) containing
a fuse, but you usually build one in as you build a boxed transformer.

For being still such a common component, the transformer is greatly
missunderstood. Many think transformers are magically isolating noise
from the signal path, but there is more than the assymtotic galvanic
isolation. For instance, the inter-coil distributed capacitance most
does not even realise it is there, where I've seen it reprogram DSPs
and take a big PA down on the knees.

A curious usefull use of core saturation is to have a small coil which
one one passes DC current under control of a reostat (really a
variable resistor). The DC current forces the core into more or less
saturation and thus forces the core to allow less or more energy pass
by the other coil. This was in heavy use in theater ligthning and
especially in use for cinema projects arc lamps. These devices was
called transductors and they used a purposefully underdimensioned
core (compared to equalent-sized transformer).

All this migth not have made you able to figure out what loads a
particular unlabled transformer is good for, but hopefull you got a
few more hints on the limits of them.

Merry Christmas,
Magnus