[Elecraft] TX IMD

[Geoffrey Mackenzie-Kennedy]

David,

Sorry for being slow to reply.

You wrote:

> Is there a "sweet spot" to which one can adjust the power output for 
> minimum IMD,

In general terms the short answer is that there might be. If the two tone 
signal driving an amplifier under test is very 'clean' in terms of odd order 
IMD products, the ratio between the amplitude of each odd order product 
generated by the amplifier and the desired tones should gradually decrease 
as power output increases if the amplifier is 'well behaved', the limit 
being near and below compression. In practice some amplifiers are not well 
'behaved' which can result in either an increase or decrease from the 
expected amplitude of odd order IMD products as the power output is 
increased. Please note that any misbehaviour of the third order products vs 
power output can be quite different from any misbehaviour of the fifth, 
seventh, ninth etc.

If the two tone signal from the driver is not squeaky 'clean' then it will 
of course influence the outcome, which is one reason why in Type Approval 
specifications for commercial use SSB transmitters used to transmit speech, 
the two tone input interface is the mike connector, also the audio line 
input if the latter exists.

>  if so, how would you do it easily?  I presume you would need another rx 
> or spectrum analyser.

My choice is to use a spectrum analyser because all significant odd order 
products can be seen simultaneously while varying the transmitter's power 
output. Situations do arise for example in which the third order products 
might dip at some power output but the fifths increase.  A receiver can be 
used, but because only one product can be watched at a time the task becomes 
tedious. The input applied to a spectrum analyser or receiver must be of 
such a level that neither is overloaded.

> And I bet it changes with antenna reflections, ie from one end of the band 
> to the other.

How much will depend on the design of the amplifier and the load presented. 
One approach used is to specify the load VSWR limits between which the 
transmitter's performance must meet the power output, IMD, spurious output 
requirements and so on, usually using a 50 +j0 load as the reference.

Be assured that I am strongly opposed to any move to introduce any form of 
mandatory specifications covering equipment used in the Amateur Service in 
addition to those that already exist.

73,
Geoff
GM4ESD