Hi Hans & Peter
>>>Let us start a discussion on placing the plate choke instead at the
antenna
>>> end of the pi/L filter to eventually get rid of these series
resonance
>>>problems? It works fine according to what I have seen in, I believe,
>PAT HAWKERs RADIOCOMM column?
>This appeared in Pappenfus' book back in 1964, and I started that
discussion
>in a letter to Pat many years ago now. The problem is that the tuning
caps
>now have to be rated for at least twice the plate voltage - both of
them.
>The 'Princess' transmitter that appeared in the 1970 RSGB handbook tried
to
>minimise this by putting HV fixed capacitors in series with the tuning
>caps, but without bleed resistors, this doesn't help too much. The other
>problem is that the blocking capacitor now needs to be fairly big, and
>capable of handing at least 5 amps - 12 for safety.
There is an article by Wayne Cooper in 73 Magazine, August 1965, pp.
18-21 which shows that connection in a small KW amp. The author
apparently didn't take full advantage of the connection as he used a 1mH
choke -- much larger than required in this instance.
>By putting the choke to the middle of the pi - L, you can use a smaller
>blocking C and a bigger second inductor, but the current rating still
needs
>to be about 12 amps for a 1500 watt amp (safety factor of 2).
An article in June '78 Ham Radio, p.98, shows that connection.
>But you still end up swapping one problem for another. If you use vacuum
>variables, that gets around the problem. But Pappenfus says:
At the cost of doubling the voltage rating of the vacuum variables!
>' This transmitter uses a unique plate feed circuit to overcome the
>plate-feed-choke problem of sufficient inductance without self
resonances in
>or near the 2-30Mc range. High voltage vacuum capacitors are used which
can
>withstand the d-c as well as the r-f plate voltages. The tubes ( 4 off,
>4CX5000A - RZP) are series fed from the 50 ohm end of the output network
>where the r-f voltage is much lower than on the tube plates. The plate
feed
>choke, the blocking capacitor and the antenna static-drain choke are
chosen
>to form a high pass filter of 50 ohms characteristic impedance. In this
>manner, relatively low inductance chokes and a low capacitance blocking
>capacitor can be used. An arc in vacuum capacitors, due to a lightning
>transient, (my italics!) for example, would be damaging because it would
>provide a dc path. For this reason, a "crowbar circuit" is employed to
short
>circuit the high voltage power supply in the event an arc starts.'
>
That sounds like the AM-1640(XW-1)/GRT amplifier.
I have a copy of section 4 (Theory of Operation) of the manual from
that critter.
It details the RF/DC diplexer (18uh chokes / 2000pf caps) on the
output as well as many other interesting features: such as running the
4CX5000 's in AB2 (with predistortion to restore linearity), a shunt
connected parasitic suppressor, RF feedback, autotuning, and of course
the crowbar (thyratron) protected power supply.
The manual states that the crowbar protection is for the benefit of
saving the tubes & vacuum caps should they arc for any reason...
>I guess the bottom line is that there ain't no free lunch!
There may be a slightly less costly lunch...
I posted most of this back in '98 but, there are likely many on the
group who haven't seen this information.
**********************************************************************
There is a technique which I observed in a 2-36 MHz Hughes Aircraft
airborne kilowatt amplifier that was designed circa 1958.
The fixed portion of the pi coil is fabricated from what appears to be
a section RG-402. The center conductor is connected directly to the
plate of the tube and the plate blocking capacitor is connected in
"conventional" fashion between the plate and the outer conductor of the
semi-rigid cable plus the plate tuning capacitor. The other end of the
center conductor that emerges where this coil is connected to the
bandswitch and the lower frequency coils is connected to the HV supply
through a rather ordinary looking 1mHy choke of sufficient current
capability.
When this arrangement is operating at it's highest frequency, the
effective impedance presented at the top of the choke is only 50 ohms!!!!
Of course, as more coils are switched in the impedance level rises but,
since the operating frequency is simultaneously decreasing, the applied
stress on the choke probably remains somewhat constant. An amplifier
employing a pi-L output network would raise the impedance level somewhat
though the stress on the choke would still be greatly lessened as
compared to the conventional connection.
RG-402 is rated for 5KV or thereabouts DC, so this technique would be
applicable to a legal limit amplifier. I suppose the method could be
extended to higher power levels by employing RG-401 (7.5KV) or if that
isn't available, just running some high voltage / high temperature
(silicone) wire through a piece of tubing before fabricating the coil.
I've never seen this exact circuit employed in any amateur equipment
nor even described in the literature.
***********************************************************
If you are winding your own coils, you could expand upon the above
technique by extending the HV lead all the way through the Pi coil, and,
if applicable, even through the L coil! This would provide a 50 ohm
connection point for the plate choke without requiring the tank
components to accommodate the DC plate voltage.
Not free... but, certainly no where near as costly as requiring that
the plate and load capacitors be rated for the DC plate voltage plus the
RF.
73,
Marv WC6W
*
________________________________________________________________
YOU'RE PAYING TOO MUCH FOR THE INTERNET!
Juno now offers FREE Internet Access!
Try it today - there's no risk! For your FREE software, visit:
http://dl.www.juno.com/get/tagj.
--
FAQ on WWW: http://www.contesting.com/ampsfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|