I see. A sort of push-pull oscillator with the link between the plate of
two tubes acting as an inductance.
Therefore you try to keep this inductance low by using sheet metal
instead of wire. Resonance should then be at a frequency high enough for
the tube gain to be low enough so no oscillation occurs.
I had a closer look at that unusual angled anode connection, and I saw
that the piece of fiberglass-epoxy PCB is double-sided and is used as
the plate capacitor! From its size, I calculated that the capacitance of
this piece of PCB should be about 300 pF. This was probably found by the
builder of the amp to be too little, so he added a 500 pF 10 Kv doorknob
capacitor in parallel.
Dielectric loss in fiberglass-epoxy at shortwave frequencies should be
acceptable apparently, because no discoloration of the copper due to
heat was apparent.
So, here is yet another use for FR4 fiberglass epoxy PCB: plate cap in a
shortwave PA!
73 de Angel Vilaseca HB9SLV
Alex Eban a écrit :
> That was done in order to avoid VHF ring oscillation, a common problem with
> paralleled tubes.
> They ha d a tendency to oscillate at a frequency set by the tray inductances
> of the tank circuits and the output capacitance of the tubes. This was the
> builder's way to reduce the stray capacitances to a minimum. Bear in mind:
> this is not the same as parasitic suppressing! This is intended to avoid
> problems stemming from another mechanism.
> We did something similar with an 8 tube amplifier producing 150kW for a
> short wave radio station.
> Alex 4Z5KS
>
> -----Original Message-----
> From: Angel Vilaseca [mailto:avilaseca@bluewin.ch]
> Sent: Tuesday, June 30, 2009 10:10 AM
> To: Alex Eban
> Cc: dezrat1242@yahoo.com; felipe@dxwatch.com; amps@contesting.com
> Subject: Re: [Amps] Sweep tube amp by DL9AH, made by HB9AWI
>
> Hi Alex,
>
> An interesting idea, but since there is plenty of room in the tank
> circuit compartment of the amp, I think I will first try replacing the
> Fuchskreis with a pi-circuit, if the anode impedance of the 5 pentodes
> in parallel allows it.
>
> There is another enigma with this amp: Instead of having the 5 anode
> nipples connected together with pieces of wire, there is an angled plate
> about 2 x 3 inches above the tubes and 3x3 inches on the side made of
> sheet metal and PCB that connects the anode connectors together and to
> the plate cap and HV coil. Never saw such a thing!
> I wonder if this is for matching purposes, to add some capacitance from
> anodes to ground? Quite awkward I think, especially considering that the
> whole plate is at anode voltage!
>
> Thanks for your comments
>
> Vy 73 de Angel Vilaseca HB9SLV
>
> Alex Eban a écrit :
>
>> ...if anyone wants to get away from the tested and true PI or PI-l
>>
> circuits
>
>> a better choice would be a link coupled tank circuit or a multiband
>>
> circuit
>
>> similar with the Z-Match innards. This is exactly what it was designed
>>
> for!
>
>> At the impedance levels encountered in multitube setups- a few hundred
>>
> ohms-
>
>> you get very reasonable "Q"'s and you have to switch only the output side
>> coupling coils (relatively low voltages).
>> Alex 4Z5KS
>>
>> -----Original Message-----
>> From: amps-bounces@contesting.com [mailto:amps-bounces@contesting.com] On
>> Behalf Of Bill, W6WRT
>> Sent: Monday, June 29, 2009 8:41 PM
>> To: Angel Vilaseca
>> Cc: felipe@dxwatch.com; amps@contesting.com
>> Subject: Re: [Amps] Sweep tube amp by DL9AH, made by HB9AWI
>>
>> ORIGINAL MESSAGE:
>>
>> On Mon, 29 Jun 2009 11:26:01 +0200, Angel Vilaseca
>> <avilaseca@bluewin.ch> wrote:
>>
>>
>>
>>> Advantages of this tank circuit are: simplicity, no bandswitch contacts
>>> to burn nor associated wiring to cause stray resonances.
>>> Potential inconvenients I can think of : Safety issue when
>>> bandswitching, Stability?
>>>
>>> Only trial and experimenting will tell!
>>>
>>>
>> REPLY:
>>
>> This is an ancient design and was pretty much abandoned years ago for
>> good reason. Take a look at handbooks from the '30s and '40s and you
>> will find it there. There are two major drawbacks to it:
>>
>> 1. Harmonic suppression is poor. The antenna is connected across a
>> coil (or part of one). At higher frequencies (think TVI), the coil
>> acts as a simple high-pass filter, just what you don't want. Compare
>> this to a pi-network where the antenna is connected across a capacitor
>> and just the opposite happens.
>>
>> 2That alone should be a show-stopper, but there is one more.
>>
>> 2. Band switching requires switching two taps for each band. Again,
>> compare to a pi-network which only switches one tap.
>>
>> As soon as the pi-network became well understood it almost totally
>> replaced the tapped-coil output. TVI reduction was a major factor in
>> adopting the pi-net, as you will note in discussions of the circuit in
>> handbooks of the early TV era. Another mechanical factor is that a
>> variable capacitor is much simpler and more reliable than a variable
>> (or tapped) coil.
>>
>> 73, Bill W6WRT
>> _______________________________________________
>> Amps mailing list
>> Amps@contesting.com
>> http://lists.contesting.com/mailman/listinfo/amps
>>
>>
>>
>>
>
>
>
>
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|