> >Mixers are always non-linear devices, if they aren't they won't
> >"mix". Why doesn't the mixer in your receiver create close-in IMD
> >or cross modulate?
>
> Mixers do generate intermod products. However, we have to define what is
> an "acceptable" level of intermodulation distortion. A mixer driven at
> proper levels with a proper LO applied will produce negligible distortion.
> If IM performance improvements are desired, a mixer with a higher LO
> drive capability can be used.
That's right. It's an imperfect world.
The mixers in my FT-1000, however, are cleaner than the class A
push-pull FET's used in the preamplifier to drive the mixers.
The mixers in your SSB transmitter are also likely very much
cleaner than the various "linear" stages in the transmitter.
> The mixer in your receiver WILL generate IM in the presense of strong
> signals (such as in a contest weekend!). That's why on almost all rigs
> you have a front end attenuation setting. On the fancier rigs like the
> FT-1000 series, you have a preamp that you can switch in and out and
> several attenuation levels you can switch in and out as well.
Cross modulation and IMD also occurs in nickle plated coax
connectors. The problem is what the rest of the system does with
IMD, not how many angels (or devils) we can fit on the head of a
pin.
> As to Tom's comment about 1/2 cycle single stage amps producing more
> distortion than a push-pull amp, I agree with him. Distortion (IM
> products) are dependent upon bias conditions (conduction angle) not how
> many tubes you use. If that were the case, you could build a class C
> linear amplifier with many tubes! Two tubes each running in opposite half
> cycles will produce just as much IM products as a single tube. Perhaps
> this myth got started from the fact that perhaps at one time someone's
> exciter when driven hard produced a lot of IM. So, the guy built an amp
> with two tubes, cut his drive level in half an viola! reduced his IM. But
> that had everything to do with a crappy exciter and not the amp.
I think this all got started because it is easy (and a natural
mistake) to confuse harmonic distortion that does NOT affect
signal quality with IMD distortion that does affect signal quality.
People often take performance results of an audio amplifier out of
context and apply them to resonant PA's. Obviously, if we think
about it, harmonic distortion (which is measured at a cyclic rate) is
a non-issue. The closest harmonic distortion product that would
affect our 14 MHz rig appears on ten meters.
The closest IMD product would occur at one time the lowest
frequency tone separation away from each tone, but that distortion
occurs because of the shape and amount of envelope distortion.
Even non-linear stages can provide excellent suppression of these
undesired in-band products, if the shape of the non-linearity is
correct. Some class C amplifiers, when operated at levels outside
the crossover distortion point, can even produce acceptable IMD
performance in non-critical applications.
Hopefully all the time on this thread is worth it, and some old wive's
tales will die. It isn't what happens over a single RF cycle that
causes problems in narrow band PA's. It is what happens as the
envelope level is changed, and the slope of the response to that
change.
73, Tom W8JI
w8ji@contesting.com
--
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
|