[Amps] Centurion cooling improvement ideas for AM service

[Will Matney <craxd1@ezwv.com>]

Hal,
When I spec a transformer for any amp and assuming it's for AM-CW 
service along with SSB, I figure the current and voltage that will be 
needed to sustain the peak power when tuned correctly at 50% duty cycle 
minimum. Some don't do this but I do and is what I recommended on the 
transformer design webpage. Actually, for ICAS the transformer should be 
rated for 50% duty cycle which is the same as the max peak output power 
running at class AB1 less the lost power (the other 50%). This is 
generally always greater than the power needed to sustain the carrier at 
100% in any case. If it is not, it should be re-figured to do so. So to 
figure what power output from the transformer is needed, take the 
maximum peak power output and figure the voltage and current from it and 
not the input power. If the amp is going to run at 100% duty cycle or 
CCS, the you will need to figure it from the maximum input power.

An example would be an amp with an output of 1500 watts PEP and a 
carrier of  375 watts. A plate voltage of 2500 Vdc and 600 mA would be 
needed to produce this at 50% duty cycle and or ICAS. For CCS at 100% 
duty cycle, and assuming a 50% efficiency for class AB1, it would need 
2500 Vdc and 1.2 amps. One popular transformer manufacturer, I reverse 
engineered some of their transformers and found some were around the 50% 
rating the way I do it, but found some less, especially in some smaller 
ones. Some rate them only for the 375 watt portion which would only 
yield 150 mA. I have also seen some rate them at 100% CCS only for the 
carrier which would then be 300 mA. However, some long winded guys would 
tax a transformer and power supply a good bit at either of these. At 
least, it should be 300 mA, but I recommend 600 mA. A manufacturer would 
also look at somebody getting inebriated one night, lock the key down 
and walk away from one. Then come looking for a replacement transformer 
under warranty. To me, I think I would want the 600 mA model here. It 
more than likely will fail over a period of time too because its 50% of 
the full input. The transformer does get a rest though between the 
voltage peaks of each cycle, but I'm really not sure how much and if it 
has ever been researched. My guess is about 15-25%, but that's me 
guessing too. Another thing a manufacturer will look at is a greenhorn 
operator tuning one tight as a dicks hat band with no modulation or 
"swing". I have seen this done time and time again. These carriers may 
be two or three times what it should be when tuned correctly. Then 
again, it's also according to how good of a warranty a manufacturer will 
give too! Anyhow, that's the way I look at the subject and may be wrong.

Will Matney


Dear Sirs,
       In reading about how the SB-220 will be used on AM  I wonder
if any of the '220 users out there have connected a known-good 50 ohm
dummy load AFT of the amplifier and a known-good Reflected RF Power
meter between the exciter and amplifier and measured what the 
amount of reflected power is with and without the amplifier on line.

       Is it not true that the SB-220 pass-through circuitry and the
input tuning circuitry sometimes displayed greater than ten percent
reflected power with the amplifier in STBY and an equivalent amount
of reflected power while keyed?

       In developing an A-M signal and worrying about the ability of
the amplifier to withstand prolonged keying, is it not wise to determine
if an inherent defect of design is adding to the problem?

       An additional cooling option is available at Max-Gain on the web.

They sell a Tarzan fan that can be bolted to the exterior of the '220,
doing away with the internal fan altogether. The Tarzan fan is much more
robust and coupled with the dual-level airflow in the '220/'221 series,
should provide more than adequate cooling for long key-down times.

        The only other issues to tackle are the anode and filament
transformer capabilities.

Respectfully,

Hal Mandel
KA1XO


_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps