> Is this totally correct with some of the new generation of
> FFT based spectrum analysers, Tom?
I have a newer Agilent FFT spectrum analyzer, and it is
certainly correct on it. Mine has the optional fast ADC
(option 101) high bandwidth FFT and the FFT function only
works to 10MHz. The FFT is primarily for analyzing signals
already within the passband of the analyzer IF, like
modulation characteristics. It can zoom in on a modulated
2GHz carrier and analyze the close-spaced frequency
components. It can't find a few cycle long burst of energy
on an unknown frequency.
http://cp.literature.agilent.com/litweb/pdf/5091-4049E.pdf
A spectrum analyzer sweeps frequency while looking for any
signal or noise and gives an amplitude display. It's great
for looking at the frequency content of a reasonably steady
signal. It's useless for identifying a dampened wave or
frequency of an unknown transient, especially when someone
just sticks a wire near a circuit and calls that a probe.
> Admittedly, it would seem to me that a better tool for
> this would be a sampling oscilloscope, such as the LeFroy,
> which will capture single cycles of 2.4GHz, and do FFTs,
> thus acting as a spectrum analyser (although without the
> resolution of the conventional spec an.).
A very good very fast FFT oscilloscope would be the only
tool for the job, but the claim is a spectrum analyzer was
used.
> Now if I'd gone into RF integrated circuit design and gone
> to California, rather than into applications and systems
> engineering, I could confidently expect to be having to
> scrape by on about $200k a year and could afford
> one.......And even then, it's very doubtful if any VHF
> ringing would be seen.
It's especially doubtful any VHF ringing would ever be seen
with the limited rise and fall times of the anode. Any bias
change is limited to the frequency response of the path back
to the cathode as well as the anode response
characteristics, and that limit is well below VHF.
What Rich is proposing is a small plate current level change
(from 200mA or less to near zero mA) at a rate below 1 MHz
(longer than 1 microsecond) will ping a circuit at 100MHz
and cause damage while a signal changing plate current from
a few amperes to zero amperes at 30MHz (.0333 uS long) will
not.
It's worse than the melted glass theory and backed only by
an instrument that can't be used to measure what was claimed
to have been measured.
73 Tom
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|