Date: Thu, 15 Oct 2015 23:44:36 +0100
From: Chris Wilson <chris@chriswilson.tv>
Organization: Gatesgarth Developments
Message-ID: <323462263.20151015234436@chriswilson.tv>
To: Alan Melia <rsgb_lf_group@blacksheep.org>
In-Reply-To: <9E19F2EA072B4AE3A7A112534A29B460@gnat>
References: <1541636554.20151013163425@chriswilson.tv> <43464DACBFAA49C08A73FB93FFF52105@gnat> <603371935.20151014123307@chriswilson.tv> <9E19F2EA072B4AE3A7A112534A29B460@gnat>
MIME-Version: 1.0
Subject: Re[2]: LF: Re: Should I be aiming for a better match than this?
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Status: O

Hello Alan,

Wednesday, October 14, 2015


Hi again Alan. Thanks for the info. I have got a better match by
making a new tap on the loading coil between 2 existing ones and now
the variometer will shift the phase side to side as expected. You were
spot on there!

Today I have finally got what I think is a handle on what's happening.
If I TX into my dummy load, from lowest power to highest there is no
odd waveform displayed. As one would expect the V and I are in perfect
phase, and they look like a single trace.

If I connect my aerial set up sometimes I can get a near perfect phase
and amplitude, but after a while it starts to go awry, for no apparent
reason. The voltage trace gets a dog leg in it, that gets worse.
Sometimes I can actually alter the shape of the traces with the output
level control on my TS-590 exciter. Similarly sometimes I can tweak
things right again with the audio level from the sound card, in either
WSPR or OPERA modes.

However, when TX'ing into a dummy load the RF level from the TS-590
exciter and the audio level have very little, if any, impact on the
scopematch waveforms save when the level drops below about -10 dBm.
Then of course drive is lost and the amp trips out.

The circa 0 dBm maximum DRV out from the 590 goes to a pre amp inside
the amplifier, then to a frequency doubler, then into the input side
of the G3YXM amp circuitry.

What I have discovered, and I am pretty sure this is what is
happening, is that at higher power levels RF is getting back into the
input of the amp, or the CMOS stuff or low volatage power supply, and
it seems to sort of "run away with itself".

I placed a BNC T at the input of the amp, where a short RG-58 co-ax
cable feeds it from the DRV (exciter) output from the TS-590. A -10dBm
to 0 dBm signal from the 590, say generated bu the TUNE function in
WSPR is immediately modulated, even running low power, and at high
power from the amp there's a lot of signal imposed. If I disconnect
my antenna co-ax at the amp and run a long co-ax to my dummy load,
there's no noticeable modulation of the input. Am I right in thinking
that would mean there is no INTERNAL cross talk within the amp itself,
or have I got that wrong?

I would welcome tips as to how to test how the RF is getting in, and
where, please. Another poster, John Rabson, talked of isolating mains
ground and RF ground. As I have had occasional RCD trips at the house
distribution panel when TX'ing at high power with this setup, I am
wondering about this. He spoke of an isolation transformer, but I am
not sure just what he meant by that. Apart from the normal house mains
grounding of the amp case my only other ground is a six foot spike
near the loading and matching coil, about 10 feet away and connected
to the matching coil earthy end and the braid of my coa-ax feed from
the  amp  to the matching coil. As my shack is upstairs I have not had
much  luck using it as an RF ground as the cable length to it radiates
on  HF,  so  I have tended not to use it at all, and with a horizontal
loop  as  an  HF aerial I didn't need to feed it against ground like a
ling wire for example.


> Hi Chris well it may depend where abouts in its rotation the movable coil
> is. If you think of tuning a coil with a ferrite slug you can get a peak
> when the core is in the middle of the coil. However to get it right you need
> two peaks, one with the core at the top of the coil and one where it is at
> the bottom and a dip in the middle.  Now its different with the variometer I
> know because you have the phase indication which should indicate too 
> much/not enough reactance. I am thinking if you have say too much 
> inductance, as you rotate the varible coil the antenna will come closer to
> resonance then as the turning continues it will go away again. However I
> would expect the phase shown on the scope to stay the same side showing say
> always below resonance. (then you would tap up a bit from the bottom and try
> again). The current and voltage traces should sweep past each other but not
> jump one side to other. That is rather odd.

> I'm sure you can find a few 3m dishes if you really want to look like GCHQ
> :-))  Of course the higher you put your antenas the less obvious the are and
> the smaller they look
>  :-))

> Alan
> G3NYK


-- 
Best regards,
 Chris                            mailto:chris@chriswilson.tv