>Jon
>KE9NA
>But speaking of chokes and choke impedances, here is a question that
>another member of the reflector posed to me a week or so ago. Maybe
>someone can answer it:
>
>The average output impedance of the tube is about 2 to 3 KOhm. From my
>RF knowledge, the tank circuit when tuned, looks like approximately the
>same impedance. However, a typical RF choke at 80 m has an impedance
>around 2 KOhms or so. This is right in the ballpark of the impedance of
>the tube, it's not a "high" impedance relative to the tube's output.
>So...why does the choke work???
Yes, that's close. One would want to account for the inductive reactance of
the choke, in the value of the first capacitor shunted with the tube
capacitance. It reduces the capacitive reactance, being out of phase with
it. The pi network or whatever, should account for this in the calculation
of the cap value. Otherwise it will be off a bit, esp at the lowest
operating freq.
In bigger tubes, the plate R is low, on order of <500 Ohms, so a 2500-5000
Ohm choke reactance is great.
That's what i just did, and using the network analyzer, I did a plot from
1-100 Mhz of the pi. Then removed the plate choke. Very little if any
effect was noticed on the through swept response. Since i have a narrow
band amp, that's easy. I have two operating frequencies to work on, and i
can live with 30 minutes to change freq. So built two different chokes,
both optimized to be operated on the inductive side of parallel resonance
(self resonance). That way I get high Z in circuit, even with strays of the
cabinet detuning it a bit.
As for the choke killing the RF power without self destructing, I am not
sure I can envision it. If there is DC current and voltage being metered,
and yet no RF output, then the laws of conservation of energy say that
there has to be power dissipated somewhere. either as RF or heat or
something. Perhaps there was a very high frequency oscillation going, that
was sending RF out to the load? But not showing on a wattmeter?
If the choke was a dead short circuit for RF, then yes, you may never get
the plate swing. Still, you shouldn't see plate current then. And the plate
bypass cap may be destroyed quickly, creating a DC short path. THis is like
the parasitic theories, a chicken and egg question. If you could stick a
bad choke in a functional working amplfier at full power, with a magic
switch, I would expect it to be blown up quickly.
But if the choke is already there, at series resonance, then the RF
wouldn't even tune close, and your amplifier would be operating in some
wierd mode, probably at a totally off frequency, could even be pumping
power into a parasitic circuit, made partially by the new short circuit (RF
shorted choke). This is all speculation, as nothing is that perfectly a
short or open, but only degrees.
-----------
NOT THE WAY TO TEST RF CHOKES UNDER POWER!......
I remember standing there with a technician once, next to an operating RF
dryer system, rated at 30 KW CW at 27 MHz. We had a long stick with a neon
bulb in the hot end. We probed an operating plate RFC, and found that it
did indeed have the desired 1/4 wave voltage response. The doors were off.
Those were the days (not too long ago I might add) where we could feel the
RF, and you didn't grab the handrail behind you while performing this
test.....
I do not condone that this be attempted with amateur amplifiers, as you may
either get killed by the HVDC or destroy your amplifier with an inadvertant
arc.
John
K5PRO
--
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
|