Re: [Elecraft_K3] filters
toggle quoted messageShow quoted text
From: Elecraft_K3@... [mailto:Elecraft_K3@...] On Behalf Of Rob Sherwood.
Comments on my K3 data vs. the three quoted roofing filter bandwidths.
Once the dynamic range values are up in the mid 90s or higher, several factors affect the results. Passive intermodulation in the roofing filter is often a issue, and there is sample variation if one tests several different filters with the same filter bandwidth. The variation in dynamic range due to the slight intermodulation in the filter itself can scatter the results a few dB. It is hard to know whether the Intermod in the measurement is the result of distortion in the filter or the mixer, not that it really matters.
Also the K3 receiver with the 200-Hz filter (and 200-Hz DSP bandwidth) has a lower noise floor than with the 500-Hz filter and it integrates less phase noise. Thus the dynamic range is modestly higher with the narrower filter. In the case of the data with the 200-Hz filter, there was no Intermod in the filter or the mixer compared to the level of phase noise. The League now calls that Reciprocal Mixing Dynamic Range (RMDR).
Nit picking between a 95 dB number and a 101 dB number is rather pointless. If one is really trying to work weak DX in a pileup, keying sidebands of the undesired stations will often be the limiting factor if they are within 1 to 2 kHz of the DX.
On SSB the problem of the intermodulation sidebands of an interfering signal is even worse. Most rigs with a 13.8 volt power amp are not very clean compared to a tube rig like a Collins 32S-3. Let's say the DX is S3 and there is QRM 3 to 5 kHz away that is S9 +30 dB, a 60 dB difference. The splatter from the adjacent signal will likely make copy of the DX difficult. The dynamic range of the radio at 90+ dB is way better than the splatter problem.
A 2.1-kHz roofing filter (or a 1.8-kHz) and a 1.8 DSP bandwidth will eliminate the offending signal, but it cannot get rid of the in-band splatter. The Intermod products are on frequency.
It is unfortunate that while receivers have improved dramatically in the last 10 years, transmitters (particularly on SSB) have not.
73, Rob Sherwood, NC0B.
On Apr 30, 2014, at 9:37 PM, "Joe Subich, W4TV" <lists@...> wrote: