From: G0MRF@aol.com
Message-ID: <b4840ce9.2471c628@aol.com>
Date: Mon, 17 May 1999 15:21:12 EDT
Subject: LF: Rx converter performance.
To: rsgb_lf_group@blacksheep.org
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 8bit
Sender: <owner-rsgb_lf_group@blacksheep.org>
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org

I was pleased to hear Roger's 73kHz beacon on air earlier in the week.   
Although a little down on its usual level, I was suspicious that my converter 
wasn't performing as well as it does on 136k.

Using Roger's beacon and some transmissions on 136 as test sources, I 
discovered that my receiver was being "de-sensed" by the level of local 
oscillator feedthrough.

The converter upconverts the LF band to an IF of 10MHz. Unfortunately the 
closer you get to "10MHZ" the lower the performance.  136 is OK but 73 was 
poor in comparison.

As I have already fitted a multi turn pot to optimise carrier balance on the 
MC1496, I appeared to need a very narrow filter on the output.  normal coils 
/ caps were no use, but the answer came in the form of an 86 pence 10MHz 
ceramic resonator from RS components.

I fitted two small ferrite transformers with a 1:9 and 9:1 ratio back to back 
on the output of the converter. From the centre point (at 350 Ohms) I 
connected the resonator in series with a 65pF variable to ground.  Listening 
to the 10MHz breakthrough and tuning the capacitor I managed to reduce the 
level by 27 - 28dB.

The performance on both bands has been improved.  MDS on 136 has been 
improved by about 1 -2 dB,   but 73k  is 10dB better than before!

The technique could be useful to anyone trying to build a robust RX (DL) 
using high level diode ring mixers.

resonant frequency of the "notch filter" could be set anywhere from 9.6 MHz 
to 10.05MHz.

73


David