On Tue, 6 Nov 2007 10:09:34 -0000
"hamilton mal" <[email protected]> wrote:

> John
> Your beacon was peaking S5 at 1732z then qsa 0 for periods and up
> again, characteristic of insufficient power for the freq at that time.
> It is the same on HF with qrp stns, in and out of the noise, whereas
> qro stns are solid copy.
> 73 de Mal/G3KEV
> 
>

Thanks Mal,

This was the purpose of the test and you've confirmed my own thoughts
on the matter. 

A few dB makes a huge difference when signals are close to the noise -
the difference between "workable" and "not there". I'm speaking only of
ear-copied CW.

QRP, in and of itself, is fine. I'm a G-QRP Club member (#2377) and
rarely exceed 10W on HF. I know 5W is the QRP "limit" and I do often
work at 5W or less. I find an extra 3dB, up to 10W, makes a massive
difference at times. It's almost never necessary to exceed 10W for most
purposes.

On 500 it looks like really low ERPs can occasionally be used
there is a practical limit to getting workable S/N ratios. For
reasonable path lengths this seems to be at the very least 25mW
ERP and up. Any lower and it's a very, very hit and miss affair. All
depends of course on the receiving station's noise levels. 

Trying to run at 10W or less is not really going to net many longer
range QSOs except under very good conditions.

Thanks to all who sent reports. It was a useful exercise. 

Cheers,

John