Carl,
Righ you are. Yet here it's hard to know in the case of these tubes.
They are pulse rated. So what is the "proper" high voltage? The GI7B
range could be argued to run between 1800-9KV. Pretty big spread.
The ham application of these tubes, especially in RTTY type use (abuse?)
conditions is just about as far as you can get away from pulsed radar
duty. I could not find a reliable source of data on the best operating
sweet spots - and with 10 suggestions from the community, I had 10
different answers - there is not much data-based rules of thumb on the
operating domain and implications.
In my case, I ended up using a variac arrangement that would let me move
the plate from 1500-3400v. And it took a long time to work out the
"sweet" spot on the GI7b. For the GS35, I think the similar thing would
be needed there.
These are more esoteric comments though. Clarly the tube works and
works well, almost without regard to the B+. Amazing well.
These old tubes would be a good PhD project for an EE student who knows
how to use the Tagucci DOE methods technique.
73/jeff/ac0c
Carl wrote:
> To add to that IMD also gets worse as Ep is increased especially when
> it is operated above its published ratings for any class of operation.
>
> Carl
> KM1H
>
>
>
> ----- Original Message ----- From: "Jeff Blaine"
> <keepwalking188@yahoo.com>
> To: "Ulf Tjerneld" <star@frizon.org>
> Cc: <amps@contesting.com>
> Sent: Saturday, May 16, 2009 8:59 PM
> Subject: Re: [Amps] first time questions
>
>
>> Ulf,
>>
>> To make the pi calculations, you should consider using the LOADED plate
>> voltage. Idle plate voltage is not a factor. The difference, however,
>> is not very much. Maybe 10-20%...
>>
>> And on #3, you are right. The voltage and current are unimportant when
>> considered SEPERATELY. Unfortunately, there is not very much emperical
>> data published on the Russian tubes so picking the optimal B+ is hard to
>> do. Generally speaking, higher B+ is better as the efficiency and gain
>> is better. But in my experience with the GI7b, at a certain voltage
>> point, the efficiency of the amp starts to drop. At low B+, the PO is
>> fine - the trouble is the Ig gets high due to the low gain of the tube
>> at that lower B+.
>>
>> 73/jeff.ac0c
>>
>>
>> Ulf Tjerneld wrote:
>>> Hi all,
>>>
>>> I'm collecting parts for an amp-project. And I have som newbie
>>> questions:
>>>
>>> 1, Maybe this question is silly: When calculating the Pi network for
>>> an amplifier with GI7-B's I'm aiming for an unloaded plate voltage of
>>> 2500 volts. With one tube this means that the plate current would be
>>> 350 mA for an input power of about 875 watts and not dissipating more
>>> than 350 watts. If I'm instead going to use a pair of tubes, this
>>> means that I would input 700 mA in the Pi network formula. Right?
>>>
>>> 2, I have a perfect capacitor for the Pi-tank in the above example.
>>> But is rated exactly 2500 volts (Johnson something...). Would that be
>>> to little for an unloaded plate voltage of 2500 volts?
>>>
>>> 3, Most people that builds amplifiers based on the GS35B use a plate
>>> voltage of around 3500 volts. This would mean a plate current of about
>>> 700 mA for 2500 watts input. But let's say I only have 2500 volts, but
>>> enough iron in the transformer. Would I achieve equal results if I let
>>> the the tube draw more current? I.e. let the tube run with 2500 volts
>>> and a plate current of 1 A? And to extrapolate further, even go to a
>>> plate voltage of 2000 volts resulting in a current of 1.3 A? To
>>> summarize: does it matter what combination of voltage and current I
>>> use as long as it is within the specs of the tube?
>>>
>>> All the best
>>>
>>> Ulf/SM0NOR
>>>
>>>
>>> _______________________________________________
>>> 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
>
>
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|