[Amps] HV MOSFETs for RF

Sat May 6 11:37:14 EDT 2017

I find the work of K1JJ quite interesting.

Adjacent channel suppression in the order of 75 dB for voice signals is very good, and represents
probably the best what reasonably may be realised

In this matter, I have noticed that the noise pedestal around the SSB signal seems to be noticeable for many amateur
transmitters. It would be interesting to find out if more people have studied the actual levels that can be expected, and compared them
to the distorsion products in the same frequency range (+/- 20 kHz or so).

There appears to be quite few measurements made, and the specifications sometimes are quite vague, the 10 kW Collins 208U-10A amplifer is specified to
not degrade the S/N ratio in a 16 kHz bandwidth of an exciter with an inherent S/N of 60 dB at its output with more than 10 dB.

By some calculations, I have been able to estimate the noise floor caused by amplified thermal noise around the centre frequency of the 80's SRT 
TD90 ISB exciter to about -90 dB relative to the nominal PEP within its roofing filter bandwidth of 8 kHz. This may be representative for professional gear,
but any further measurements on amateur equipment seem to be quite absent.

My interest in this question comes from the research for a paper that I and a colleague are in the process of writing for the upcoming Nordic HF Conference.
It is tentatively titled "Performance limits for HF communication systems using realisable hardware solutions "

The main topics are the limits that the different parts of the system chain impose on the total performance; 
transmitter noise and distorsion, co-channel and adjacent channel interference, man-made noise and receiver performance. 

The work has been inspired by the mid-70's work by Marconi's chief scientist B M Sosin and by more recent studies made by Harris and Rockwell-Collins on the performance of wide-band HF systems. Also the publications of Leif Åsbrink, SM5BSZ, that I hope will be able to review the paper have been an inspiration.

73/
Karl-Arne
SM0AOM

----Ursprungligt meddelande----
Från : kd6vxi at gmail.com
Datum : 2017-05-06 - 02:28 (CEDT)
Till : manfred at ludens.cl
Kopia : amps at contesting.com
Ämne : Re: [Amps] HV MOSFETs for RF

It has been done.  - 50 dB or better, from 4 milliwatts.

"It starts with a low level  FT-1000D (4 mW)  output > into a (1 watt)
ZHL-3A lab amplifier > into a homebrew 4CX-350FJ (5 watts) Mr. Clean >
 into an 8877 (50 watts), > into a pair of 8877's (1500 watts).

Today, after a lot of tweaking, using a two-tone test, I am seeing -50dB to
-55dB 3rd order IMD products.  Using voice LSB at 2.8 KHz bandwidth, I see
side crud products down as much as -65 dB within 3khz of center frequency.
This is outstanding and blows me away.  Expand the the pics and count the
5dB vertical grids to get an idea of signal to crud (splatter)."

Tom, K1jj spent a LOT of time perfecting the system.

Note it takes class a and a pair of 8877s driven by a single 8877.  Not a
whole lot of amateurs want 4500 watts of blast furnace to equate to what an
ANAN now accomplishes.. Something Tom has expressed in another
conversation.......

For those that care, the above was plagiarised from
http://amfone.net/Amforum/index.php/topic,33505.0

--Shane
KD6VXI

On Friday, May 5, 2017, Manfred Mornhinweg <manfred at ludens.cl> wrote:

> Cathy,
>
> To your list of other services needing linear amps you can add analog TV
> transmitters (where still used), digital TV transmitters, certain medical
> equipment, marine and aeronautical HF communications, military, various
> scientific uses, satellite transponders, and on and on...
>
> What components do they typically use for their final output,
>> high-power devices?
>>
>
> Whatever they can find that does the job! Tubes in olden times or at
> extremely high power, bipolar transistors after that, then it was VDMOSFETs
> and planar DMOSFETs, then LDMOSFETs, and now MOSFETs made from composite
> semiconductors are being used.
>
> > Could hams use the same approach?
>
> Of course. And they have been doing it for ages.
>
> How do they
>> get around the efficiency and cooling problems we've been discussing?
>>
>
> There are two approaches: Either live with those problems, or use some
> high efficiency scheme to avoid them.
>
> The "live with them" approach typically means using devices with high
> power dissipation rating at quite low power output. For example, check the
> datasheet for the BLF188XR LDMOSFET. It starts giving a table of typical HF
> and VHF applications, most of which are non-linear, highly efficient ones
> with power outputs of 1200 to 1400W along with efficiencies up to 85%. And
> then there is a single, lone linear application mentioned: A digital TV
> transmitter on VHF, delivering 225W output power at 29% efficiency! That's
> how they do it.
>
> And the other group uses high efficiency techniques, like the Doherty
> method, Kahn's EER, or direct baseband PWM.
>
> In this context it's interesting to remember a snippet of ham history: In
> 1974 the amateur satellite AMSAT Oscar 7 was launched. This satellite
> carries a mode B transponder built by Karl Meinzer (DJ4ZC) and Werner Haas
> (DJ5KQ). This transponder uses Kahn's EER, called HELAPS (High Efficiency
> Linear Amplification by Parametric Synthesis) by the builders. Karl made
> this project part of his Ph.D, in 1973. The specs are pretty impressing:
> Operating in the 2m band, with 50 kHz bandwidth, many simultaneous signals
> going through a single transmitter, -40dB IMD, and 10W PEP output at no
> more than 5W average power consumption for the whole transponder! Talk
> about efficiency! Of course this is possible only because of a low
> average-to-peak ratio, but still it's pretty impressive.
>
> The document with schematics is available, and is essential reading on
> this matter, along with Kahn's publications. And AO-7 is still working
> whenever the sun angle permits, 43 years after being launched! Which proves
> that this technology can be quite reliable...
>
> So, don't believe that EER on the ham bands is something newfangled and
> outrageously modern, that needs to be developed from scratch. It has been
> with us for many decades, and the only thing we need to do is come up with
> a practical, inexpensive, good, hopefully reproducible design, that puts
> out legal limit power, using some of the recent high power devices!
>
> Manfred
>
> ========================
> Visit my hobby homepage!
> http://ludens.cl
> ========================
> _______________________________________________
> Amps mailing list
> Amps at contesting.com
> http://lists.contesting.com/mailman/listinfo/amps
>
_______________________________________________
Amps mailing list
Amps at contesting.com
http://lists.contesting.com/mailman/listinfo/amps