From: Paul
Date: Sun, 28 Jan 2001 22:58:40 +0000
Subject: [TSSP] Re: Virtual Secondary Database
Hi All,
Here's an estimator for f1 that you might like to check out,
the variables h, awg, sr, hd and b are as defined in
http://www.abelian.demon.co.uk/tssp/vsd/
f_h = -10.5599*h*h*h + 57.3609*h*h - 88.4737*h + 35.9407
f_awg = -94.763*awg*awg*awg + 9000.55*awg*awg
- 301175*awg + 3.64056e+06
f_sr = 3.49962*sr*sr - 7.90961*sr + 5.81855
f_hd = -0.000209288*hd*hd*hd + 0.00559242*hd*hd
+ 0.0664809*hd - 0.0153714
f_b = log( b/0.2)
t = pow( h, -2.0) * f_awg * f_sr * f_hd
s = -3.83651e-15*t*t*t + 1.17645e-08*t*t + 0.631198*t + 484.407
f1_prediction = s * (1+f_b/100)
Valid for coils without topload.
For frequencies between 20kHz and 1.5MHz, the max error on the
vsd dataset is 3% and the RMS error is 0.9%
For example, my coil,
b = 0.15
h = 1.6
sr = 0.906
awg = 12
hd = 2.76
f1_prediction = 90.35 kHz,
f1_measured = 90.9 kHz.
Cheers,
--
Paul Nicholson,
Manchester, UK.
--
Maintainer Paul Nicholson, paul@abelian.demon.co.uk.