Dear Rik, Pete, Jim, LF Group,
I don't doubt that supply ripple to the PA could produce unwanted sidebands
on a signal, but there are a couple of reasons why I don't think this is so
in the case of G7NKS's signal:
1) The sidebands are spaced at +/-50Hz, 150Hz, while one would expect 100Hz
and multiples due to power supply ripple - this would require a very odd
ripple waveform, and at least several percent ripple, to get sidebands
at -30dB as they are.
2) The sidebands on G7NKS's transmitted signal are mirrored by the apparent
sidebands on signals received at G7NKS, for example:
G7NKS received at M0BMU:
2009-11-15 17:50 G7NKS 0.503802 -27 -1 IO92ub 0.02 M0BMU
IO91vr 38 171
2009-11-15 17:50 G7NKS 0.503852 +5 -1 IO92ub 0.02 M0BMU
IO91vr 38 171
2009-11-15 17:50 G7NKS 0.503902 -25 -1 IO92ub 0.02
M0BMU IO91vr 38 171
M0BMU received at G7NKS:
2009-11-02 08:00 M0BMU 0.503899 -19 -1 IO91vr 2 G7NKS
IO92ub 38 351
2009-11-02 08:00 M0BMU 0.503948 +12 -1 IO91vr 2 G7NKS
IO92ub 38 351
2009-11-02 08:00 M0BMU 0.503998 -18 -1 IO91vr 2 G7NKS
IO92ub 38 351
The apparent sideband levels and frequency offsets are virtually the same,
but the PA can hardly be responsible for the sidebands on the received
signal, and even if supply ripple was causing the problem on receive in some
other way, it is remarkable that the relative levels are virtually the same.
So I think Jim's problem must be being caused by a component of his system
that is common to both TX and RX. Overheating of the PA, etc. is a different
issue.
Cheers, Jim Moritz
73 de M0BMU
----- Original Message -----
From: "James Cowburn" <[email protected]>
To: <[email protected]>
Sent: Wednesday, November 18, 2009 9:33 AM
Subject: RE: LF: Re: G7NKS sidebands
Hi Rik
That's exactly what mine are. I think it's a mixture of stretching the Pa
on the IC735 so its at the edge of its envelope (the radio equivalent of a
stick shaker or stall horn on a plane!) and also RF leakage and mains
leakage into the soundcard system on the Dell lappy. It's only apparent
on
Rx and Tx of strong/local signals where the fundamental is so high (>0db)
that the first side lobes at -30dB are still within WSPR capture range.
|
