Received-SPF: neutral (google.com: 193.82.116.20 is neither permitted nor denied by best guess record for domain of owner-rsgb_lf_group@blacksheep.org)
Message-ID: <000401c70808$c20cae40$2101a8c0@PCRoelof>
From: "Roelof Bakker" <roelof@ndb.demon.nl>
To: <rsgb_lf_group@blacksheep.org>
References: <1GjjhM-0giu7m0@fwd29.sul.t-online.de>
Date: Tue, 14 Nov 2006 17:20:10 +0100
MIME-Version: 1.0
Subject: Re: LF: RE: LPF coils  and VY1JA to 1W ERP
Content-Type: text/plain;
	format=flowed;
	charset="iso-8859-1";
	reply-type=original
Content-Transfer-Encoding: 7bit
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Status: O

Hello Ha-Jo,

Ten years ago I have been tinkering with passive audio bandpass filters.
In an attemp to reduce the physical size of the inductors, some 9 mm hi-mu 
cores were used. The Q of the inductors was measured using a signal 
generator loosely coupled to the tuned circuit and reading the frequency at 
the - 3 dB points which were measured with a high impedance  probe and 
oscilloscope. I found that the resonance frequency varied with applied AC 
voltage.

In my note book I have the data for a FT-37-43 ferrite core with 200 
windings:
The tuned circuit had a 1 uF capacitor in parallel with the core and was 
coupled to the audio generator by a 10nF capacitor.

Vpp(mV)        F(Hz)        L(mH)
=======================
1                    1214            17.2
2                    1208            17.4
3                    1186            18.0
4                    1176            18.3
5                    1162            18.8
10                  1134            19.7
15                  1103            20.8
20                  1073            22.0
25                  1064            22.4
30                  1026            24.1

A filter was build with 88mH 9 mm cores with a lot of windings.
I still have it in a direct conversion receiver.
The funny thing is that one can use it as a very crude AGC system when the 
signal is tuned to the slope of the filter.

It appeared that larger cores suffered less from this effect and the TOKO 
series 10RB magnitically screened inductors did not have it at all.

Best regards,

Roelfo Bakker, pa0rdt