I'm sorry I lost the mail about the two tone test.
Here is why it is done in amateur linear PA's.
SSB (and some other) forms of modulation contain multiple
frequencies at the same time. If you remember the description of
mixers, if a system is non-linear (power output does not track the
power input change by a uniform amount) two frequencies become mixed.
You get the two original frequencies, the sum of those two
frequencies, and the difference of those two frequencies.
You also get higher order products of two times one frequency minus
the other. Three times one minus the other, etc.
Let's say you have a 7.200 MHz signal mixed with a 7.203 MHz signal
(this would be two tones, 3 KHz apart). You would have 14.400 -7.203
= 7.197 as one byproduct. And 14.406-7.200= 7.206 as another. This
is the third order product.
Another would be three times one minus two times the other. That
would give 7.194 and 7.209 as byproducts, these are the fifth order
products.
It goes up to the highest odd order product you care to take.
If the PA does not track the input signal levels equally and linear,
mixing occurs.
Linearity is tested by feeding two equal pure sine wave tones
(normally from two separate CW transmitters) into a PA, and measuring
these distortion products. They can be measured with a selective
level meter (like a receiver with a calibrated meter) or a spectrum
analyzer.
If the amplifier is changing gain as the tones add and subtract,
extra frequencies will be produced. The PA will act like a mixer as
well as an amplifier. With a two tone test, you can measure the
effects of gain distortion or non-linearity.
(Notice this has everything to do with gain changes as the input
level changes, and nothing to do with grid current or anode current.)
The only problem with a two tone test is it does NOT change power
supply loading, except at a VERY fast rate (the rate is F1 minus F2,
or 3 KHz in this case.
Almost any amount of capacitance will filter out the 3 KHz
change in power supply loading, so a two tone test NEVER
represents the true SSB performance (although it is better than a
wild guess made by looking at meter currents or tube tables). That's
why I suggest varying the level of the two tones at a speech syllable
rate, or a three tone test.
I hope this helps.
73, Tom W8JI
--
FAQ on WWW: http://www.contesting.com/ampfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|