[Amps] Price per Watt Conversation

Sun Apr 23 15:28:28 EDT 2017

Well, I am with Manfred, and I believe the professional transmitter world also is.'
Looking at what is offered to the commercial and military markets, they all use the
distributed amplifier and hybrid tree approach.

I follow the developments closely as a consultant in civilian and naval HF, and it appears
that all high-powered offerings use power modules of maximum 500 or 600 W output, that¨
are summed together. 

Some years ago, I got a preview at the factory in Pomezia of the new generation of SELEX 5 and 10 kW solid-state
transmitters using plastic cased MOS-transistors. Each module uses 2 transistors, and 12 modules are summed to
provide 6 kW from each half in the 10 kW unit,

This is in accordance with previous proven engineering practices. 

The very reliable Rockwell-Collins PA-2250, directly derived from the HF-8023, used the same approach
to reach 3 kW with bipolar transistors as did the, alas short-lived, PA-2224 which used 16 MOSFET modules to reach 4 kW.

The heat dissipation and cooling requirements associated with continuous operation of SSPA:s are often underestimated.
SELEX states that their current 10 kW transmitter dissipates about 28 kW as heat at 10 kW key-down output.

This scales to about 800 W of dissipation from each of the modules with driver stages and power supply efficiency factored in.

A distributed approach is necessary to get rid of these amounts of heat in a reasonable way.

73/
Karl-Arne
SM0AOM

----Ursprungligt meddelande----
Från : manfred at ludens.cl
Datum : 2017-04-23 - 20:04 (CEDT)
Till : amps at contesting.com
Ämne : Re: [Amps] Price per Watt Conversation

Hi Cathy,

> So would it be better to use 4 or or more devices with somewhat lower
> ratings for linear operation?

I do think so. Specifically I would look for devices designed 
specifically for linear service. These have a relatively high power 
dissipation rating, compared to their output power rating. For example 
an ancient Motorola PT9780 transistor, of which I have a bunch in my 
junk box, is rated for 100W RF output, and 350W dissipation. That allows 
it to survive infinite SWR at full continuous power, by the way. The 
very well known and much used MRF150 is rated for 150W output and 300W 
dissipation. That's a bit tighter, so its survival rating is for 
near-infinite SWR at 25°C case temperature, which is impossible to 
maintain in practice. So it's NOT able to withstand high SWR for long, 
without protection circuitry. And among modern LDMOSFETs, one device 
that comes to mind is the MRFE6VP5600H, rated for 600W RF output and 
1667W dissipation.

Instead the BLF188XR and several other modern LDMOSFETs are NOT designed 
for linear use, and you can notice that in the very different ratio 
between output and dissipation: The BLF188XR is rated for 1400W RF 
output but "only" 2000W dissipation. The MRFE6VP61K25H for 1250W output 
and 1333W dissipation, and the new MRFX1K80H for 1800W RF output and 
2222W dissipation.

Given that all these dissipation ratings are based on the case being 
kept below 25°C, which is impossible in practice, they have to be 
adjusted to realistic dissipation values by taking into account the case 
mounting surface, the heatsink/spreader/fan used, and the mounting 
technique (greased or soldered). As a rule of thumb, small devices can 
be made to dissipate 1/3 to 1/2 of the rated amount of power, while with 
large devices this gets worse, like 1/4 of the rating.

So, the 100W output rated PT9780 has a realistic dissipation capability 
of around 120 to 170W, which is comfortable for 50% efficient operation, 
and has some headroom for overloads caused by high SWR. Instead the 
1800W rated MRFX1K80H can achieve perhaps 600W dissipation in the real 
world, so in 50% efficient operation it's good for 600W output - and not 
for 1800W! But in nonlinear, saturated operation, at 75% efficiency or 
better, it can do 1800W output.

> Is there a cheaper option to use simpler heat sinks that don't have
> to be carefully machine, but push them less hard by spreading the
> power over a larger number of devices?

Yes. By using more devices, cooling becomes much easier. Usually it's 
good economy to use more devices, because the higher cost for more 
devices is more than offset by savings in the cooling system. Up to a 
point, of course. This point must be determined for each case.

 > What's the practical upper
> limit of devices that could be used before creating other design
> problems?

I don't think there is an upper limit. It's more a question of what's 
practical. There is an increasing number of hams using homebrew 
amplifiers that use 16, 24, 32, or even 48 cheap 1-dollar MOSFETs. Those 
FETs are much like your IRF510, only for higher voltage. The IRF710 is a 
typical one used in those amps.

Manfred

========================
Visit my hobby homepage!
http://ludens.cl
========================
_______________________________________________
Amps mailing list
Amps at contesting.com
http://lists.contesting.com/mailman/listinfo/amps