This IS a valid question.
I am not sure about others, In my case it was 90% Cost. With side effect
benefits.
I knew that for high duty cycle modes you should run this amp at reduced
power.
I was using the SB-220 at 500 watts output in a RTTY contest. But I
guess even that low was not low enough for about 4 hours into the
contest the magic smoke that makes this stuff work escaped from the HV
transformer.
Shopping around for a replacement transformer gave me a major case of
sticker shock. I bet the cost of all the replacements I was finding was
close to the cost of the AMP when it was bought new! I was ready to
either shelf the amp or sell it.
Then thought, well voltage is voltage, is there anything else out there
that can make the same or similar voltages and would work.
And there is, Yes it is considerably larger than the original
transformer, so it required some modifications of the chassis of the amp
to allow it to fit. But it does fit. Barely!
And it cost 25% of what all the replacement transformers were wanting.
And yes with the raised voltage and the cost savings I decided to make
changes to the whole HV section too. Changed to safely handle the new
parameters. Whole new caps and resistors, New Harbach HV board etc. I do
not know if it is the new cap bank, or the transformer itself, but it is
incredibly stiff compared to the stock one that was in there. The HV
voltage drop from no load to max out is not even 100 volts! like 3700 no
load to 3600 at 2Kw.
The AMP runs stone cold, it is HARD I mean HARD to get the plates to
even turn red, never mind the orange that people say it should often do
to "Getter". At 1500 Watts OUT, Even with it transmitting a dead
carrier, after minutes of making RF, the plates were still gray. At 2Kw+
out, during a normal like Sweepstakes exchange of length, you may get
some Red plates. Orange needs the same 2Kw AT A SOLID CARRIER for about
a solid minute to get to the orange state.
Yes the output network needs work. Unless I run 1400+ watts all the
time. Then the network runs fine!
I have learned that I need more "L"
But that is a work in progress.
Trying to find good new taps was a pain. So in the amp now is a 18uH
roller inductor.
So now the only thing the output bandswitch is now doing is add that cap
when on 80 meters.
Now I am still lost when it comes to this "Q" thing.
And what are the optimal settings for this network. That roller changes
everything.
I can get a perfect 1 to 1 match, with the Plate cap at absolute minimum
meshed state, and I also can get a perfect match, with the cap like 50%
meshed, like where it would be normally on 40 meters, but still on ten.
By the roller being in a different place. And of course anywhere in-between.
I am learning this combo of different values of plate cap and the
inductor, is what determines the level of that mysterious "Q" but then
again I may be wrong here.
The AMP is like 99% ready to be put into service,. Just need the help
with this "Q" Stuff, and to discover what to do with the plate cap and
"L" values for optimal performance.
Joe WB9SBD
On 10/11/2021 4:49 AM, Rob Atkinson wrote:
Why do some of you guys want to jack up the plate v. on your 3-500s?
The power difference won't be noticed and you'll create a cascade of
problems. You still have the same plate dissipation and plate current
limits. The pi network won't work, the DC blocking caps may become
marginal, the power supply filter cap bank may get stressed and the
IMD spec. gets worse. If you want to jack up the B+ get a 3-1000Z.
73
Rob
K5UJ
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|