From: Tony Williams
Subject: Re: 87% All that vector calculus paid off
Date: Tue, 24 Dec 2002 10:09:19 +0000 (GMT)
References: <email@example.com> <firstname.lastname@example.org>
NNTP-Posting-Date: Tue, 24 Dec 2002 10:09:18 +0000 (UTC)
User-Agent: Pluto/1.14i (RISC-OS/3.60)
In article ,
Christopher R. Carlen wrote:
> Tony Williams wrote:
[Ipk = 1.6A, L(effective) = 70uH ?]
> Indeed you are quite right. I am lacking some decent current sense
> resistors. I'll be in a better position to make these measurements once
> I get some of those. Nor did I take pains to cross check my numbers in
> a few different ways.
The WAG'd calcs depend on the accuracy of your numbers
> I think the Al is measured at small signal levels, and at some
> relatively low f like 25kHz, so that at higher frequencies as the
> permeability decreases, it takes down the inductance, and also the
> inductance tapers off a bit with high flux levels.
It looks like 157mT B-pk is too high and causing the
u-effective to roll off.
> Maybe tonight I can check the current with a 1 ohm resistor,
Look out for a slight upward curl at the end of the
It would also be interesting to see if the efficiency
improves if the 2.7uS ON-time is reduced.
An even wilder WAG; If you have 32 turns of (say) 0.8mm
wire, then your I^2*R losses in the winding will be less
than 200mW (if I've done the calcs right).
I don't have data on the core material, but would suspect
that the core loss (swinging over 157mT @ 150000 times/sec)
would be somewhat greater than 200mW.
It looks like there is margin there to swap some of the
core loss for I^2*R loss, and so get a better energy
transfer efficiency from the inductor. ie, increase the gap.
We are just entering the realms of optimising a polarised
inductor.... eg, Hannah curves.