[Elecraft] KPA-1500 Fan Noise

[Fred Jensen]

Yes John, and it turns out to be a fairly simple equation if you make a 
few assumptions.  Around 1958 or so and a student [sophomore], I spent 
the summer with a small crew building KPFK on Mt. Wilson in Los 
Angeles.  The TX was almost totally homebrew [REL serrasoid exciter], 
with 4 4-1000A's in the PA [4 parallel class-C 4-1000A's at 90.7 Mcs was 
a bit of a challenge [:-) ].  About to return to school, I was reviewing 
the physics I'd supposedly learned the previous year and realized I 
should be able to calculate the efficiency of the PA from the smattering 
of thermodynamics I knew.  Tx engineers don't have much to do until 
something breaks.

I knew the pressure difference between the air inlet and the exhaust 
[quite small, there was a differential manometer], the filaments 
dissipated 600 W, nearly all in heat, and I knew the temp difference 
between the gozinta and gozouta air so I could calculate the mass of the 
air moving per unit time or so I thought.  Then, with the PA on and 
stable [3.5 A @ ~5.8 KV], I could do it again only this time calculate 
the power lost to heat the air.  Subtract the 600 W from the filaments, 
and I got 69.4%. I assumed the air was heated at constant pressure, my 
calculus hadn't progressed far enough to do otherwise.

After I went back to school, they finally measured the efficiency on a 
2-stage oil/water cooled dummy load and called me and said they got 
71.3%.  If it doesn't go up the 3 1/8" hardline, it comes out as heat 
... somewhere.  There's only two options.

I'm a little surprised at the obsession with fan noise ... I'd like to 
hear about how much more DX everyone is working with a KPA1500.

73,

Fred ["Skip"] K6DGW
Sparks NV DM09dn
Washoe County

On 6/2/2018 1:06 AM, John Perlick wrote:
> Bill Schmidt is right--you have to get rid of a bunch of heat--no matter what device you are using to amplify (tubes, transistors, or magic). The amplifying device must dissipate considerable heat because we are running Class AB and because of all electronic devices are naturally lossy.
>
>