I *hope* that this *third* attempt to post on this thread makes it.

As others have said calculating saturation is hard and inaccurate. It changes
with temperature, mchanical strain in the core due to mounting and other factors.

Bsat = Amperes * turns. Electronic circuits which measure saturation in real
time are very popular because of this, such as current mode dc/dc converters. A
very simple circuit which allows one to measure core saturation is a one or two
transistor voltage boost oscillator. The oscillator turns off exactly when the
core saturates so measuring the current will tell one exactly Bsat, given the
number of primary turns.

Modelling Bsat in LTspice is hard as there seems to be built in saturation
modelling but it does not work. Threads can be found where Helmut describes how
saturation can be modeled using atanh(). Atanh() models a soft saturation.

To model a hard saturation one can use a two inductor series model, with a small
inductor representing the air cored inductor and the large inductor the
remainder of the inductance with the unsaturated core. A current controlled
switch shorts the larger inductor when |Il| > Isat. Model the Rdc of the coil by
setting the Rdc of the air cored coil, and model saturation knee 'hardness' by
setting the Ron of the current controlled switch.

The atanh saturation model cannot be used with LTspice coupled coils
(transformers). The current controlled switch based one can be coupled (used as
a transformer model) using two Kxy L1 L2 0.95 Spice statements, one each to
couple the cored and the air cored inductors of the transformer (each winding is
represented as two coils in series as above).

-- Peter


-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist