[Elecraft] IMD

[Mark J. Dulcey]

Sean Baxter wrote:
> Jerry:
> Assuming there is not a conspiracy of radio manufacturers trying to pawn off
> cheap radios on unsuspecting amateurs, I have the following questions
> concerning IMD.
> 
> How bad is the difference between -21 verse -30 dB IMD?  Can someone
> actually hear a difference on the signal?  Is it wider or more distorted
> at -21 dB?  Maybe it costs many dollars more to engineer and manufacture a
> radio with -30 dB IMD specifications?  If no one can hear the difference but
> only seen on a scope, then who cares?  I do not know maybe Gary or Wayne
> would know the answer.

It's not the people listening to you that notice the difference (the 
effect in audible quality of your own signal is minor, and besides, the 
crystal filters in the receiver will get rid of a lot of those IMD 
products); it's the people trying to carry on a QSO a couple of kHz up 
the band that notice it. A 9dB reduction in IMD products equals 1.5 S 
units less QRM; a real difference in crowded band conditions.

Yes, it costs more to design a radio with better IMD performance; you 
usually have to use the output devices more conservatively (meaning more 
expense for transistors or tubes) and you might need to design more 
negative feedback into your design (meaning more overall open-loop gain, 
which may require higher-gain drivers and output transistors and/or an 
additional circuit stage). You're also likely to need to operate the 
final output stage in a more linear mode (class A like the optional mode 
of the FT1000, or at least higher quiescent bias in class AB mode), 
which means higher power consumption, a real drawback for radios that 
may be used for mobile or portable operation.

So far, there has been little market pressure for better transmitter 
performance; unlike receiver IMD specs, hams haven't shown any interest 
in paying extra to get it, or in choosing rigs based on transmitter IMD. 
Unfortunately, unlike receiver IMD, investing in improved transmitter 
IMD performance doesn't improve your own on-air experience much; the 
improvement is in the experiences of other hams. Still, if hams start 
paying more attention to transmitter specs, the equipment manufacturers 
will start to give us better radios.

Meanwhile, there is a way to improve the performance of most existing 
designs - just turn down the output power a bit. That does, in effect, 
two of the things I mentioned; operating the output devices more 
conservatively, and increasing the quiescent bias (not the absolute 
amount, but in relation to the output power). Cutting back to half power 
probably cuts the IMD level of a typical ham rig by about 3dB (that is, 
the ratio between signal and IMD products; the actual level of the IMD 
goes down by more than that).