[Amps] 2 x 4cx1500b Tube matching?

[Ian White, G3SEK]

>
>I want a build a dual 4cx1500b amplifier. I managed to obtain a batch 
>of unmatched 4cx1500b's all different date codes.
>
>I was wondering which method would be best to achieve matched and 
>balanced tubes? Since the cathode is tied to one side of the filament i 
>cant use a cathode resistors to equalise currents. Is there a drive 
>configuration that would allow me to drive these 2 tubes without 
>worrying about matched tubes?
>
>Assuming there is no drive configuration that will achieve this, would 
>it best to use seperate bias supplies to try and equalise each tube 
>individually? Someone suggested that i use the Super Cathode 
>Configuration, is there any merit to this configuration.
>
>I have looked at G3SEK regulator board kits, it seems these boards 
>require matched tubes.

When I was designing the boards, I looked very carefully at the 
possibility of dual independent screen and/or grid supplies to handle 
un-matched tubes... and decided it would be pointless.

It is a myth that un-matched tubes can be somehow 'fixed' by changing 
the screen or grid bias to equalize the zero-signal anode currents. The 
ZSACs are different because the tubes have slightly different internal 
geometry - particularly grid-cathode separation - and that also means 
they will have slightly different transconductance.

If you equalize the ZSACs of two non-matched tubes by applying different 
levels of bias, their operating conditions will still move apart when 
you increase the drive. There are various other possibilities for 
equalizing anode or screen currents, but if the two tubes are not 
identical you can still only equalize the operating conditions at *one* 
point on the operating line.


But don't start worrying yet. It may not be a problem that you have 
several tubes from the same manufacturer with different date codes. That 
doesn't automatically mean that they aren't well matched - it all 
depends on how accurately each one was made in the first place (and to a 
lesser extent on the number of hours operation since then). Test each 
tube, on its own and under the same DC operating conditions, and see 
which two out of the batch are most similar.

If you can't find any pair that are reasonably well matched, still don't 
spend your money on more elaborate power supplies. Either buy a pair of 
tubes that have been matched by the seller (in other words, tested and 
sorted from a large batch) or else buy some more tubes and sort them 
yourself.



>I would also like to DC ground the screen  and these boards dont appear 
>to support grid driven DC grounded screen.

Sure they do! They are designed to support all realistic combinations of 
grounding and driving. Only the DC grounding matters to the power supply 
- whether you drive cathode or grid is an RF design decision.



>I am not familiar with the procedures for matching tubes, can someone 
>explain what the exact procedures are for matching transmitting tubes 
>DC and RF wise. I  assume that resting equal grid current would be the 
>indication of a close match.

Equal resting *anode* current is a good indicator. Just build the 
two-tube power supply you first thought of, but plug in only one tube at 
a time to measure the ZSAC. Don't touch the bias controls while you're 
going this.

Then pick out the pair of tubes that are most similar. Plug 'em both in, 
and finally adjust the bias to 2x the manufacturer's recommended ZSAC 
for a single tube.


-- 
73 from Ian G3SEK          Editor, 'The VHF/UHF DX Book'
                           'In Practice' columnist for RadCom (RSGB)