[Amps] Price per Watt Conversation

[Roger (K8RI)]

Couple of points:

I hope I didn't make any of my usual math mistakes.
The SS amps I'm familiar with do use a modular approach.  Quadra and I 
think the Tokyo Hi-Power (no longer available)
The quadra uses 8 MRF151s (Sorry, I had been calling them 150s) in 4 pallets

MRF151G Is a different animal than the the MRF151s. It's a flat pack 
rated at 300 W out or double the MRF151s and multiple units are 
relatively easy to cool. It's basically 2 151s in a single package. 
Although more expensive than the standard MRF151 it's still an 
economical way to a linear, legal limit amp that is tolerant of 
relatively high SWR.

Yes there is a limit on how many you can parallel, but it's dependent on 
the frequency and circuit design. Just like the new LDMOSFETs are rated 
to 470 MHz, but a practical circuit doesn't exist for a single amp to 
cover that entire range. Sure you can parallel 16 cheap devices, or 
more, but a point is reached  in the circuit design where it just 
doesn't do well over a wide range. There are individual exceptions. OTOH 
with good transistors like the MRF150G, or the new LDMOSFETs, 4 or even 
8 can cover the 1.8 through 50 MHz range. Although there is no specific 
number limit because of the wide variance in devices and circuit design, 
there is a practical limit for a specific circuit.

Temperature.   Although ratings are given at 25 C, typically air cooled 
devices run between 40 C and 70 C (based on experience with computer 
design and cooling) The MRFX1K80H is derated at 11.11W / deg C, so at 40 
C (40-25 = 15 X 11.11 = 166.5) or 1800-166.5, or a MAX rating of 1633.5, 
but for 70C we have (70-25= 45 45 X 11.11= 499. 95, or 500W  1800 - 500 
= 1300W.    Again, this is compared to class C, but 4 devices should 
loaf along at 375W output ea in linear service at 40C.  With water 
cooling and chilled water, the case temp really can be close to 25 C  
(77F) and the much lower voltages used in SS amps makes water cooling 
much easier than in tube amps.

Running linear at 60% of 1300W is 780W gets us into only half the 
derated power which "should be OK for SSB, but digital?

This brings up the question of duty cycle. Just what are we referencing 
against.  Deep class C could be 50% or even less.  If so, then SSB at 
20% should be OK, BUT digital at 100% in linear (AB1) is likely to 
overheat the devices.  Just divide the figures by 2. so we are looking 
at 750W per device which is respectable and SSB is comfortably at half 
the dissipation on SSB peaks, but digital is now running pretty much AT 
the MAX rating which pretty much guarantees a short device life.
If that class C is less than 50% then these devices are nowhere near the 
advancement they appear to be.

I'd be more comfortable with 8 of the 150Gs in PP / parallel in a manner 
similar to the 8 standard 151s in the Quadra as 8 X a real 300 W = 
2400.  Not cheap, but at $107 ea they look pretty good compared to the 
newer MRFX1K80HR5. Actually they look mighty good! certainly robust 
enough to handle any mode CCS with the total rating for 2400W output. We 
are looking at a real rating of 2400 W for $856 compared to an 1800W 
output in deep class C for $250 ea, or $1000 for a questionable legal 
limit on all modes.

So with the 50V MRF151G compared to the 65V MRFX1K80HR5, we are trading 
off some questionable advantages and more simple circuit for 4 pallets 
using 8 MRF151Gs in PP with a more complex cooling arrangement and a 
little more than twice the derating above 25 C @ 2.85W/ C. OTOH the 151G 
has a very low thermal resistance (junction to case) of 0.35 deg C / W. 
HOWEVER the MRFX1K80HR5 has only 0.09 C / W

Dollar per watt the 151G wins, hands down.  About $150 cheaper for about 
twice the real power.

    73, Roger (K8RI)

On 4/23/2017 2:04 PM, Manfred Mornhinweg wrote:
> 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. 


---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus