Idetifying IMD products]

Andre Kesteloot
Sun, 23 Feb 2003 17:30:06 -0500

-------- Original Message --------
Subject: LF: IMd products
Date: Sun, 23 Feb 2003 21:25:53 -0000
From: "Alan Melia" <>
To: "VLF Group" <>

Hi all, I think there could be problems implementing my tests but the usual
check for IMDs in a receiver is to put an passive attenuator in front of the
receiver. Say we use a 6dB attenuator the "real" signals will be attenuated
by 6dB and the 3rd order products by 18dB. This is the easiest way to
identify an IMD problem in the reciever.

It is possible for there to be products from signals way above where you may
be listening. The standard AM station spacing in Europe is 9kHz so there
could be products at around 9kHz. Another example on the 136khz amateur band
is intermod usually now in the front end of the RX between BBC Radio4 from
Droitwich on 198kHz and MSF (Rugby) Standard Frequency transmissions on
60kHz appear on 138kHz. Since reducing either of these signals with a filter
reduces or kills the IMD in most cases, it is within the Rx or pre-amp. It
is quite possible for a strong signal at night to intermod with with a
signal within your groundwave range. The difficulty is trying to disentangle
the modulation to identify it.

If the problem is ICM (Ionospheric Cross Modulation) The two stations are
probably in a line from you. both modulatons will be heard on the product
but the frequency response will be quite restricted. Say one is a BC station
...there will be little modulation above about 1kHz to be heard. ICM usually
require a station with a power of around 200kW, and you may get ICM from
stations further away. It is quite difficult (even for the professionals) to
identify the stations involved. I am not sure whether this can lead to VLF
signals but I cannot see why not as it is a mixing process. The usual effect
is for the distant station to be heard as low level modulation in the
background of the high power station (regardless of the frequency of the
distant station) There is little you can do about this form of interference
unless you can drop it into the null of a loop.

I am afraid that rather amessy explanation as I know but I hope it gives you
a few ideas.

Cheers de Alan G3NYK