> Not so. I'm not an idiot, Rich. I always set the carrier control according
> to Kenwood recommendations -- meter deflection within the ALC zone.
I have to agree with Dick here.
My IC-706, no matter how it is adjusted, spikes. So did a 950SDX
I owned for a week.
The 775DSP I had drove my AL-1500 to 4 kW output on leading
edges when I opened the load control up, almost like a pulse
transmitter. When I didn't open the load up I got constant grid
overload detector faults.
The FT1000 doesn't have a bit of that problem if I adjust the drive
correctly, neither does the 751A under any condition.
It's tough on the PA to bang it with a millisecond long pulse at
three or four times rated drive, especially with MOX cathode tubes.
By the way, I measured the pulse power on a peak storage
wattmeter. It looks a lot smaller on a model 43P, because the
meter movement is so slow and the pulse so short.
> Actually, I normally set it as low as possible, just under the point where
> the output stops rising. But my 950 spikes with meter deflection anywhere
> in the ALC zone, even when set to just start indicating at the bottom of
> the zone, as you have recommended. The rig was checked by a
> factory-authorized technician, who found the ALC to be correctly adjusted
> and working to specs.
What specs are those? What we really need are some standard
specs. ;-)
> In my opinion, it's a design problem: if you look at the transmit timing
> curves in the 950 service manual, you can see that after a 15ms delay, CW
> output is initiated by a sharply-rising square wave signal called CKY. A
> signal to turn on the ALC is initiated at exactly the same time, but rises
> and decays more gradually than CKY. In my opinion, this sometimes allows
> the output to spike to the max before the ALC has turned on. The timing is
> probably very close, one way or the other. By increasing the CW waveform
> rise time, the ALC has more time to catch up. Just a theory.
That is what happens. The ALC is so slow, the envelope rise time
is faster than the ALC response time. One cure with some rigs is
to add an external regulated adjustable dc ALC source, that you
use as a power control.
> By the way, the spikes are easily observed on an 87A amp, which fault
> trips in response. All I have to do is send a steady stream of dits in QSK
> mode. Neither an FT-990 nor an FT-1000mp has ever driven that fault.
Think of all that voltage built up in the tank when the tube is banged
with all that drive!
73, Tom W8JI
w8ji@contesting.com
--
FAQ on WWW: http://www.contesting.com/ampfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|