Topband: Receivers, Noise Blankers and Key Clicks

[Michael Tope]

On a related topic, I just got done "declicking" the local club's
Yeasu FT-1000 per the procedure on Tom's website. In order
to see just how dramatic the improvement was, I decided to
do a pre and post test. The club has 3 HF rig - 2 FT-1000's
and a Kenwood TS-850. I set up a link simulator using a JFW
300 watt power attenuator followed by a couple of smaller
fixed pads and an HP355C step attenuator. This allowed me
to dial in the signal level going to the test receiver (the
"undeclicked FT-1000"). As an aside, the FT-1000 S-meter
turned out to be surprisingly accurate above S7 (at least on
this particular rig on 160 meters with the preamp "on"). S9
was almost exactly -70dBm, and the S meter reading went
up almost exactly 10dB with every 10 step of the HP
attenuator.

In any case, aside from the loud clicks on the stock FT-1000,
I noticed a lot of noise sidebands on the received signal. Even
at a moderate signal level of -50dBm (S9+20dB), the noise
sidebands of the FT-1000 at 1KHz offset seemed to be
unusually high. Although it didn't click, the clubs TS850 sounded
similarly noisey. At first I thought perhaps I was getting some
stray coupling between the nearby transmitter and the test
receiver, and in fact when I increased the HP attenuator setting
to maximum (120dB), the signal level only dropped to about
S7. I found a couple of those RF inquiry "HI-Q" common mode
filters in the shack and installed them between the transmitter
under test (TUT) and the test receiver. This seemed to cure the
attenuator "blow-by" (the signal level would drop to S0 with the
attenuator max'd out at 120dB), but the noise sideband levels
didn't change appreciably after adding the common mode
filters.

Anyhow, it finally occured to me that I was listening to loud
signals with no band noise to cover up the phase noise
sidebands, so I checked the ARRL test report for the FT-1000
and was shocked to find that the transmitted composite noise
spectrum was only down -100dBc/Hz and 2KHz offset. According
to the reviewer, that was good. If so, I would hate to see bad. In
any case, I ran the numbers, and sure enough, an S-9 +30dB
signal from an FT-1000 would have composite noise sidebands
that would be 20dB above the rigs minimum discernable signal
(pre-amp on, 300 Hz filter bandwidth). I guess we don't hear that
on the air because most of the time the band noise covers
it up. I would sure hate to live close to someone using a rig like
this, however. With the signal level set to S9+50dB (-20dBm),
the noise sidebands from the FT-1000 were measuring S6
(~-85dBm) at 1KHz above the carrier frequency which I guess
is about right when you plug in the numbers (-100dBc/Hz x 300
Hz bandwidth). Still, I was really surprised by how dirty this rig
sounded (contrast this with the Ten-Tec Orion which according
to the ARRL test report has a composite TX noise spectrum
which is down at -140dBc/Hz!!).

In any event, the clicks went away when I installed Tom's
modification (thanks, Tom!!). The composite noise was still
there, however.

73 de Mike, W4EF..............

----- Original Message -----
From: "Tom Rauch" <w8ji at contesting.com>
To: <jimjarvis at ieee.org>; <topband at contesting.com>
Sent: Tuesday, February 03, 2004 2:39 PM
Subject: Re: Topband: Receivers, Noise Blankers and Key Clicks


> Very interesting Jim. That's another good case the ARRL publishing
> close-spaced testing.
>
> > Depending on what signals are present within the 15KHz wide
> > roofing filter, what you hear can be anything from keyclicks
> > to 'thumps and bumps' from lowpass filtered keyclicks, to
> > a 'growling' sound on ssb...to the am detected envelope of
> > your signal of interest, injected after the product detector
> > audio.  Essentially, you're interfering with youself.
>
> The symptoms sound like what the FT1000's (and other Yaesu rigs) do except
> in them it is caused by IM products in a FET amplifier left hanging off
the
> IF system.
>
> If close-spaced dynamic tests were published in QST reviews, many of these
> problems would be corrected. Manufacturers will only test the parameters
> that get headlines in reviews, so the real key is getting the ARRL to
> publish close-spaced test data and occupied bandwidth data for
transmitters.
>
> The tests would have to be at 500Hz spacing on CW, not 2kHz. One new radio
> about to be released has a ~2kHz roofing filter making it have really
> inflated 2kHz test figures, but we really need 500Hz or less spaced
> performance data for CW.
>
> 73 Tom
>
>
> _______________________________________________
> Topband mailing list
> Topband at contesting.com
> http://lists.contesting.com/mailman/listinfo/topband
>