> From: Rich Measures <measures@vc.net>
> Three times the anode supply potential of 2900v would be 8700v. It makes
> little sense to me that the anode voltage could possibly soar to such
> heights when a tank is essentially out of resonance.
If you read what I wrote carefully, you'll see I specified the tank
IS resonant, the load is simply removed.
How this happens is simple.
A tube appears like a diode in series with a time varying
resistance loading the choke/blocking cap junction.
The tube pulls the junction towards the chassis. That "tug" causes
the end of the blocking cap towards the tank to go more negative than
the chassis reference.
If the tank is underloaded, the tank swings too far negative. If it
does that the tube simply cuts off, a tube with negative anode
voltage does not conduct. No amount of instantaneous grid voltage
will cause the tube to conduct so the tank "free-wheels". The tube
does NOT pull the tank back up positive, it simply can't.
By the time the tank starts to swing up near the peak positive
voltage positive (remember it changes slowly because of the Q, so the
exciter is already cutting the tube off. Because of this the tank is
free to overshoot positive. The RF choke's high impedance allows the
voltage at the anode to exceed the supply voltage.
The limit to this excursion is ONLY the Q of the tank, the mismatch
presented, and voltage breakdown someplace in the system. Most
ignition systems use this very same effect to fire spark plugs, and
many switching supplies operate this way. So will your PA, if you
overdrive it for a given amount of coupling to the load.
Motorola describes this effect in solid state PA's, since it is well
known to cause the destruction of high power solid state devices
during operation into high VSWR loads. You can find it mentioned on
page 254 of Radio, RF, and Video Applications (DL413D rev 1) as
they discuss reliability of solid state FM broadcast modules in
high power transmitters. They say ""the amplifier must be designed
with the capability of supporting a VSWR of 3:1 as a minimum. This
point can be very determinant when considering that on a high
efficiency circuit (non-class A) the collector voltage swing can be
close to three times the collector supply voltage."
With single ended vacuum tubes and larger mismatches it can be even
more. I've measured PA's that produce TEN times the dc supply voltage
on the tube when the load is removed!
It's pretty well known among RF engineers that the peak anode (or
collector) voltage can easily exceed the supply voltage by more than
two times, and as many as several times, depending on tank loading.
Of course the truth doesn't sell kit$, so guys after your money
ignore it!
> In an SB-220, saturation begins to occur at roughly 150W drive.
> Increasing the drive beyond 180w does essentially nothing to the output
> level or to the peak voltage across the Tune C.
You call yourself a liar.
50 watts can overdrive the SB-220, if the loading control is
closed too far. Anyone who has ever watched a grid current meter, or
listened to a splattering signal from an underloaded amp, will know
even a few watts can overdrive and saturate an amp (causing it to
switch hard) when it is underloaded.
Rich expects everyone to believe the tube suddenly "saturates"
at 150 watts in this case, but when it's a parasitic he suddenly
whistles a different tune. In one case the tube can't saturate...
in another it can draw so much current it bends tungsten wires and
explodes high fusing amperage chokes.
Just like the tank arcing... Rich says a parasitic causes and arc
because it can produce many times the rated voltage and arc a part
or switch over...but at the same time he claims fundamental
RF drive can NEVER do the same thing. He argues with himself.
Which way is it Rich? Does the tube saturate at 150 watts, and that's
it. Or does the current behave like you argued a week ago with
Peter Chadwick?
Can the tank generate voltages much greater than the supply system
and arc, or is it always limited to two times the dc?
Make up your mind.
73, Tom W8JI
--
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
|