[Prev][Next][Index][Thread]
Re: k - Alternative Method
-
To: tesla-at-grendel.objinc-dot-com
-
Subject: Re: k - Alternative Method
-
From: EDHARRIS-at-MPS.OHIO-STATE.EDU
-
Date: Wed, 06 Mar 1996 11:42:06 -0500 (EST)
-
>Received: from phyas1.mps.ohio-state.edu (phyas1.mps.ohio-state.edu [128.146.37.10]) by uucp-1.csn-dot-net (8.6.12/8.6.12) with ESMTP id JAA27393 for <tesla-at-grendel.objinc.COM>; Wed, 6 Mar 1996 09:42:13 -0700
Hi Malcolm,
Terrribly sorry for not corresponding lately. I've been in rush to
finish my thesis since I might be in line for a job at Bell Lab's new wafer
fab. I've been a bit more active on the list lately, however, since I've
been home the last 5 dayes with the flu!
Anyway I just saw your post - amazing - I was just thinking about
the exact same method! Doing a quick and dirty error anaylsis suggests that
one needs perhaps 4digits for k of about .1, though 3 1/2 digits might
suffice for the higher k values. I have also noticed that handheld LCR
meters tend to be thrown off in the L values if the winding resistance is
above a few ohms....
The plus sign in your M=(Lx+Ly)/4 formula bugs me though. If I may:
Ls = Secondary Inductance
Lp = Primary Inductance
then
Lx = Lp + Ls +2M
Ly = Lp + Ls -2M
So
M=(Lx-Ly)/4
I hope I haven't goofed (if so I'll blame the flue ;). In the limiting case
of no coupling Lx=Ly=(Lp+Ls) and this gives M=0.
Take care,
-Ed