Tom Rauch wrote:
>
>> >It was not Rich's idea. Eimac recommends a shunt screen regulator
>> >for smaller tetrodes. I used the circuit in two amplifiers and
>> >dynamic screen V regulation was better than 1%. A caveat: Operate
>> >zeners at no more than 33% of rated dissipation.
>
>> If I remember correctly, the measurements made by GW4FRX (and/or
>> G3SEK?)
John did the measurements and faxed me the screen dumps. I prepared the
bar charts and the web page on my site.
>showed a significant improvement in IMD if screen regulation
>> was improved from around 1% to around 0.1%. That's not the same as
>> saying that 1% regulation produces unacceptable results in the real
>> world, but shouldn't we aim for the better target? The effort of
>> building a better screen supply is usually minor compared with the
>> rest of PA building.
>
>Regulation will be much better than 1% if you wrap around the
>zeners with a gain stage. The overall cost will be the same, since
>you can use cheap HV NPN or PNP transistors and smaller zeners.
>
>In that case you can use a very high value resistor from the HV to
>the zeners, and the ESR of the zeners is less important.
A few additional points:
1. The "need" for good screen regulation will depend on the type of
tube, and each individual's self-respect about the quality of their
transmitted signal.
For the 4CX250/350 family the case is very convincing, but other types
of tubes may have lesser requirements for constant screen voltage... or
perhaps even greater requirements. We don't really know, because nobody
has made the same detailed tests for other tubes that GW4FRX did for the
4CX250/350s - or if they have, they haven't published them.
2. It may be just possible to get a low supply impedance with "amplified
zeners", but there remains the problem of additional series impedances:
metering circuits, "grid stoppers" (why?), overcurrent protection etc.
Because these are all connected in series with the output of the voltage
regulator, the supply impedance as seen by the screen tends to creep up
again.
A feedback regulator with more loop gain than a simple zener/transistor
combination will improve the regulation dramatically, and has three
significant advantages: (a) it allows most of the additional supply
impedances to be placed inside the feedback loop so that they don't
affect the regulation; (b) any series impedances that have to be placed
outside the loop will not increase the total impedance too much; (c) a
positive/negative overcurrent trip can be built into the system, without
worries about its series impedance. This feature is a real tube-saver!
Finally... most AMPS readers know that I do sell supply and control
boards for tetrodes and triodes - because I believe they are worth
having. For anyone who'd rather build than buy, schematics and other
details of the screen regulator and overcurrent trip are downloadable
from my web site.
73 from Ian G3SEK Editor, 'The VHF/UHF DX Book'
'In Practice' columnist for RadCom (RSGB)
http://www.ifwtech.com/g3sek
--
FAQ on WWW: http://www.contesting.com/FAQ/amps
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
|