Hi Dave & all,
I didn't say the "tones don't change fast enough", rather that the stabiliser
circuits
tend
to "dc restore" the complex waveform and therefore don't follow the input
correctly.
Incidentally (ignore this part if you are au fait with 2 tone tests) a 2 tone
test
waveform is a 100% modulated AM signal as the peak voltage is 2 x the individual
amplitude and the summation of the signal drops to 0V at the troughs.
Using a sharply rising signal doesn't necessarily give the expected results
either in a
transceiver. You got to remember that there are several parts of the
transceiver which
frequency shape the output signal. The most significant is the ssb filter,
which has a
BW of about 2.5kHz and therefore has a natural damped oscillatory period. Hit
any
high Q narrow band filter with a big pulse of power and it will ring for
several cycles
or
longer depending on the damping factor.
To test an amplifier on its own one has to use 2 rf carriers spaced about 1kHz
apart with
equal amplitude and the correct pep power level, not a simple piece of test
equipment to
make. Oh and I forgot to mention. This test signal must have intermod figures
at least
6dB better than what you are trying to measure. No mean feat when you are
talking about
a 100W pep input signal. I have been that route at work when testing high
power
amplifiers. A typical g-g amp for hf which uses a good pair of tubes (I'm not
going to
get drawn into the argument of what is a "good tube"!) should be capable of
better than
-40dBt (40dB below the peak power of either tone) which is -46dBc (pep). Very
few
modern
solid-state transceivers can get to that level, let alone 6dB below it which is
what you
need to test the amplifier for 3rd order products. When 5th order and higher
products
are considered it is almost impossible to make a clean enough drive sorce to
test such an
amplifier, unless you use a tube running in Class A with about 1kW anode
dissipation!
BTW The spacing of the 2 tones is fairly critical. If using 2 audio tones
into the mic
input
you cannot exceed about 1.5kHz spacing or else one tone will be dropping down
the slope
of the ssb filter and you won't get the correct amplitude ratio. If you use 2
rf
carriers another pitfall awaits the unsuspecting. At close spacing the
difference
frequency, the bit which does the damage is low in frequency. If you were to
use, say,
10kHz spacing the bias decoupling circuits can cause some very wierd effects and
introduce distortion which isn't there for low tone offsets. I know , I have
it happen
to
me. Supporting what I said in my original post that we prefer the
"traditional" test
because it masks some of the funnies!
my 2 cents worth.
John ZS5JF
----------
> From: Dave D'Epagnier <DAVED@ctilidar.com>
> To: 'amps@contesting.com'; 'measures' <measures@vcnet.com>
> Subject: RE: [AMPS] 2 tone intermod test
To: <amps@contesting.com>
> Date: Tuesday, November 23, 1999 10:51 PM
>
>
>
> If John Fielding's comment about the 2 tone levels not changing fast enough
> to adequately exercise the amp are correct then John how about if I
> amplitude modulate the 2 tones (on and off rapidly)? I guess I can run a
> test with voice but it just seems so subjective and would be hard to show
> the results to anyone at the Icom factory in case I decide to send the
> exciter in for a tune up.
>
> Thanks,
>
> Dave
>
> ----------
> From: measures [SMTP:measures@vcnet.com]
> Sent: Tuesday, November 23, 1999 11:39 AM
> To: Dave D'Epagnier; 'amps@contesting.com'
> Subject: Re: [AMPS] 2 tone intermod test
>
>
> >
> >Hello, I want to run a 2 tone intermodulation distortion test on my
> hf
> >transceiver to check for splatter vs. compression level, microphone
> gain
> >settings etc. I know this is not an amplifier question, but this
> list is
> >probably where I can get the best answer. I'm running an Icom-756
> (and Alpha
> >89A). Running this test on an amp would be pretty straight forward,
> but how
> >about on a transceiver? Does anyone know how the test is done on
> >transceivers?
> >
> The 2-tone test does not simulate genuine rotten splatter. For this
> you
> need a complex waveshape. A human or an African gray parrot is such
> a
> source. The splatter test is done by tuning a double-filter Rx
> (such as
> a TS-830) about 3.7KHz above and below the test signal to isolate
> imd
> products from the fundamental) and then comparing the signal
> strength to
> the fundamental. Since S-meters tend to be inaccurate, a 1db step
> attenuator.should be used to measure relative signal strength.
> >
> cheers, Dave
>
> - BTW -- the only splatter that matters to the FCC is what goes
> beyond
> the Amateur Radio Service band.
> ..
>
> - Rich..., 805.386.3734, www.vcnet.com/measures.
>
>
> --
> FAQ on WWW: http://www.contesting.com/ampsfaq.html
> Submissions: amps@contesting.com
> Administrative requests: amps-REQUEST@contesting.com
> Problems: owner-amps@contesting.com
> Search: http://www.contesting.com/km9p/search.htm
>
> --
> FAQ on WWW: http://www.contesting.com/ampsfaq.html
> Submissions: amps@contesting.com
> Administrative requests: amps-REQUEST@contesting.com
> Problems: owner-amps@contesting.com
> Search: http://www.contesting.com/km9p/search.htm
--
FAQ on WWW: http://www.contesting.com/ampsfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|