[Amps] SPE 2K RTTY Duty Cycle / spectral cleanliness

[Joe Subich, W4TV]

On 2015-05-22 9:10 AM, Leigh Turner wrote:
 > It's the higher order IM stuff / crud that has to be curtailed...

It's any of the odd order IM since that falls in the adjacent "channel"
and creates QRM.  Even with low -30 dB IMD, that's a watt or so of
power in each of the IM products for a 1500 W amplifier.

> The modern trend towards Adaptive pre-distortion techniques works
> well to clean up mediocre PA transistors and sort of turn a sow's ear
> into a silk purse.

Adaptive predistortion isn't even necessary.  Fixed predistortion
(set once for the amplifier) will do the majority of the job but
so will proper amplifier design.

The problem with most solid state amplifiers (and the 12V transceivers)
is that the manufacturers use devices rated for pulse or CW service
right up to their full power ratings which means they are in gain
compression above 50W output (or before in the case of some IF stages).
If the manufacturers would stop trying to use the receive IF chain for
transmit (or design the receive IF chain to handle transmit power
levels) and would use more output devices (4 in push-pull parallel)
so the output impedance would be higher and the devices would not be
in gain compression most of the time, TX IMD of even the "12 V" radios
could be in the high -30 dB range even without predistortion.

Add a clean exciter to a solid state PA that used the same concepts
(more devices, higher impedance, farther away from gain compression)
and perhaps include a way to set fixed predistortion in the exciter
to match the amplifier (e.g., a one-time automatic set-up) we would
start to get somewhere.

Right now, many if not all of the solid state amplifier designs are
the equivalent of sweep tube amplifiers, the devices are simply being
pushed too close to the limit of cathode emission, well above the
linear range.  The problem of course is that the additional devices
(transistors, transformers, splitters, combiners, etc.) cost real
money and no amateur wants to pay the added cost.

Broadcast transmitter designers have known this truth for 30 years -
their solid state transmitters have probably a 4:1 ratio of saturated
power output to rated power output so that the devices are kept well
out of saturation where linear operation is required.

73,

   ... Joe, W4TV


On 2015-05-22 9:10 AM, Leigh Turner wrote:
>
> All good valid points that you make here Kevin.
>
> There is also an equal fascination / obsession in some ham radio quarters
> with IM3 and levels!  We are talking here of HF amateur / ham band
> transmitters NOT professional transmitters where spectral cleanliness is
> more important.
>
> It's the higher order IM stuff / crud that has to be curtailed...
>
> The modern trend towards Adaptive pre-distortion techniques works well to
> clean up mediocre PA transistors and sort of turn a sow's ear into a silk
> purse.
>
> Leigh
> VK5KLT
>
> -----Original Message-----
> From: Amps [mailto:amps-bounces at contesting.com] On Behalf Of Kevin Stover
> Sent: Friday, 22 May 2015 10:28 PM
> To: amps at contesting.com
> Subject: Re: [Amps] SPE 2K RTTY Duty Cycle
>
> There are NO 12V final amateur transceivers that make -40dB IM3, none. Never
> have been, never will be. The best are at -33 to -35.
>
> Those transceivers capable of 200 Watts are all usinng 28 or 50V finals and
> can reach -40 depending on the low pass filter used behind the PA.It's time
> for the manu's to get rid of this fascination with running everything off a
> battery.
>
> The most expensive part of any SS amp is NOT the finals themselves it's the
> LPF that will take 1500 watts and keep harmonics legal (-43db HF, -60dB 6m).
> That's not IM3. Name the amp that hits an IM3 of -40. The 8877 will come
> close, the Russian triodes won't, none of the tetrodes will do it.
>
>