Gerwyn gave me the filter because it seemed, "broken," and he wanted me to test it.  He built it 50 odd years ago as part of his HND course.


The thing went on the slab for investigation.


First, what is it?  It is a fairly well made, workshop built filter and not a kitchen table lash up.  The opened up casing revealed quality silver plated inductors and polypropylene / polystyrene capacitors.  The voltage ratings could not be found on the cans.  The PCB seemed sound and free from corrosion and all solder joints were good:

The back engineered schematic trace-out reveals a capacitive input, 9th order, low-pass Elliptic filter. The circuit, with both measured and extracted values, was transferred to the AADE Filter Design programme.  (Ignore the "R" value - it refers to the working impedance, ie, Z=50 Ohms):

The AADE calculated response, from 3 MHz to 400 MHz, was:

The filter was then re-assembled and tested on the Marconi 2382 Spectrum  Analyser:

The actual responses, as opposed to calculated responses, were as follows.  From 0 to 400 MHz, the vertical dotted line was the ref freq of 200 MHz and each vertical line is set at 40 MHz per division.  An analyser is normally referenced to -10 dBm as the zero loss reference line - it's just the way it is done.  On the left, just before the 40 MHz line, the nose can be seen and the rapid drop off of the skirt.  The noise floor appears at roughly -65 dBm, until it starts to rise after 200 MHz:

Opening the plot out to a centre, ref freq of 25 MHz and a vertical division of 5 MHz per div shows the marker on the nose at 31.9 MHz and the steep drop of the skirt to a floor of about -70 dBm.

And at 50 MHz ref, just for comparison:

The noise  floors are always a little compressed until we expand in their region, shooting the upper part of the response off screen at the top.  The floor is revealed at approx -90 dBm and, given the starting ref 0f -10 dBm, is hence [-90 dBm -(-10 dBm)]= - 80 dBm:

A very nice *SERVICEABLE* low-pass HF filter.  In general, given the capacitors, it is doubtful if they are up to high power HF use.

Certainly the filter is very good for receiving and/or Tx power at the more QRP end of things.

Given your lack of success, recommend you check your other cables and connections just in case anything has been disturbed around the shack

Thanks to GW0OAJ for this article

A Broken Filter

Copyright 2015  ©  Cwmbran and District Amateur Radio Society

Back CADARS Logo
Home About Us GB3RT Repeater News and Events Education Local Information  Local Conditions Useful Info

Cwmbrân & District Amateur Radio Society

For access to this area email webmaster


Members Only