Hi Mark,
That looks more useful and seems in the ball park!
I wonder if you made any measurements at 1500 watts pep output?
Using long sample times with speech you still need narrower bandwidths
than what you are using for that. Otherwise all it shows you is the
bandwidth of the analyzer filter.
Also when you are overlaying the exciter spectrum on the amplifier
spectrum, it is really telling you very little. All you are seeing is
the width of the analyzers filter at different power levels.
If you look the width of the amplifier spectrum at say -40 db and then
drop down 10 db and look at the exciter spectrum, it is just about the
same width. The difference is the slope of the analyzers filter.
If you overlay you should attenuate the amplifiers signal by its power
gain so the peak level is the same as the exciter on the screen. Then
you will be dealing with the same slope point on the filter for
comparison. But again, you need very narrow bandwidths to look at
speech or music. Because your transmitted bandwidth is narrow.
73
Gary K4FMX
Mark Marsden wrote:
> Hi Bill & Amps reflectees
>
> I've remeasured the intermodulation characteristics of my 4CX1000A
> amplifier, more thoroughly, using two transmitters and a Wilkinson
> combiner. These results, and also measurements of the harmonics from the
> amplifier, are on
> http://granta.digital-crocus.com/Linear_design_notes.php3 (half way down
> the page).
>
> 73, Mark, G4AXX
>
> -----Original Message-----
> From: Bill Fuqua [mailto:wlfuqu00@uky.edu]
> Sent: 24 March 2005 17:30
> To: Mark Marsden; amps@contesting.com
> Subject: Re: [Amps] 4CX1000A for SSB??
>
> Ok, I just took a look at this and see where the problem is.
> With a resolution bandwidth of 3kHz the 3 order IMD products are masked
> by the real signal. For example if you should have a 14.2815 MHz signal
> and a
> 14.282 Signal producing 14.2825, you will not see the 14.2825 MHz signal
> because it will be a tiny bit added to the 14.282 signal in the spectrum
> analyzer's detector. What you need to do is to reduce the RBW to say
> 500 Hz or a bit less and sample longer (a lot of talking)until you get
> the full spectrum filled in as you did before. And you should see a
> difference. In fact since the bandwidth of the SSB signal is only 2.5kHz
> or so wide to start with you will most likely not see any IMD products
> using a 3kHz resolution bandwidth. A 2 tone test is good because it is
> easier to compare results but there is nothing wrong with what you are
> doing. You in fact could do a infinite tone test (random noise source)
> and perhaps get results that would be easier to compare to others if
> they should try it.
>
> 73
> Bill wa4lav
> _______________________________________________
> Amps mailing list
> Amps@contesting.com
> http://lists.contesting.com/mailman/listinfo/amps
>
>
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|