Content-Transfer-Encoding: 8bit
Message-ID: <5.1.0.14.0.20020611163249.00aa4208@gemini.herts.ac.uk>
Date: Tue, 11 Jun 2002 17:55:50 +0100
To: rsgb_lf_group@blacksheep.org
From: "James Moritz" <j.r.moritz@herts.ac.uk>
Subject: RE: LF: DJ
In-reply-to: <01C21150.F2C33350.g4jnt@thersgb.net>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii; format=flowed
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>

Dear Andy, LF Group,

At 14:04 11/06/2002 +0100, you wrote:
>...Anyway, isn't 2 mix - the red coloured one -  the wrong material for 
>this frequency ?  I thought that was
>meant for the 2 - 20 MHz region.
It is the power materials catalogue that I have here; -2 mix is the lowest 
permeability material listed, and is recommended for resonant inductor 
applications. The curves give core loss data from 50kHz to 2.5MHz. All the 
higher permeability materials have higher loss at a given flux and 
frequency - some are better than others. Using a lower permeability means 
more turns for a given inductance, which in turn means lower flux for a 
given applied voltage and therefore still lower core loss. The limiting 
factor is when the winding copper losses dominate the core losses - I 
suppose if you want 100s of uH or mH the higher permeability stuff wins out.

Micrometals say that inductors where the core losses exceed the copper loss 
are a bad idea because the heating of the core is a vicious circle where 
the losses increase as the temperature rises - It doesn't say if actual 
thermal runaway is possible, although it does say that prolonged operation 
at high temperature causes irreversible increase in loss as well.

The Philips data for 3C85 says at 100kHz and 100mT, loss is about 
100mW/cm^3, and effective volume is 17.8cm^3 for an ETD 44 core, so your 
transformer core loss should be about 1.8W, plus winding losses.

Cheers, Jim Moritz
73 de M0BMU