[Amps] 3.5 kV 2A REGULATED Power Supply: Schematic ?

[Ian White, G3SEK]

R. Measures wrote:
>1. Do not skimp on anode and screen V unless you are a QRP fan.   If I 
>was going to build an amplifier with a 4cx3500A, I would start with a 
>least 6000V under load on the anode and 1200V on the screen . 
>Otherwise, you might not be able to utilize the full emissive capability.
>2. Neutralize  (see http://www.somis.org/D.a.05.GIF).
>3.  Under no circumstances tune the grid with a capacitor.
>4.  Except for >100MHz, consider using the 8171/4cx10,000D.  It does 
>not require a noisy high-pressure blower; it costs less, it can make 
>more P; a low IMD version is available; plus it also has handles.
>5.  For any tetrode, to prevent a potentially fatal, runaway condition 
>of reverse screen current, bleed a constant 25 - 30 mA from the screen 
>supply.

Several people, including Rich, have pointed out that screen and control 
grid voltage stability is much more important than anode voltage 
stability for tetrodes.

However, anode voltage stability cannot be completely ignored, because 
maximum RF drive and anode current will coincide with minimum anode 
voltage. So right at the moment when you want maximum undistorted RF 
output, the tube has the least capability to deliver it.

If the B+ sags too much, the result can be a large reduction in 
available RF output at low distortion... or a large increase in IMD if 
you insist on being greedy.

This applies especially to tubes that need high current but at low 
voltage. The GU84b is a typical example - say 2.0kV at 1.2A - and of 
course two of any tube is two times harder on the power supply.

As a rule of thumb, 10% voltage regulation is a good figure to aim for - 
in other words, with full drive the B+ sags to 90% of its zero-drive 
value. You can live with 15%, but worry if the voltage sags as much as 
20%. Don't worry about the voltage rising (within reason) when the amp 
is on standby.

To achieve such good voltage regulation, you need a transformer with 
very low winding resistances. Voltage doubling is not a good idea for 
high-current supplies, because it *always* has worse regulation than a 
full-wave bridge unless the winding resistances are extremely - no, make 
that extraordinarily - low.

A very common amateur solution is to increase the anode voltage and run 
the tube in more of a high voltage / low current mode. Most tubes will 
handle this quite comfortably. For example, the GU84b seems to run 
happily at 2.8-3.0kV instead of the rated 2.0kV, and that is much better 
for the B+ supply.

Triodes are even more sensitive to anode voltage (because they don't 
have a constant-voltage screen grid) so the same problems and the same 
solutions apply - and even more strongly.

However, I don't see much role for precision regulated B+ supplies in 
amateur amps, except maybe to allow you to use an existing transformer 
that would deliver too high a voltage for the tube(s) you want to use; 
or perhaps for portable use where the mains regulation is very poor.


Just one final point about tetrodes...
Rich wrote:
>5.  For any tetrode, to prevent a potentially fatal, runaway condition 
>of reverse screen current, bleed a constant 25 - 30 mA from the screen 
>supply.

25 - 30mA will often be OK, but it won't prevent runaway in all possible 
cases. Some tubes - or pairs of tubes - will generate larger negative 
screen currents than that. The important thing is to bleed *more* 
current than *whatever* maximum screen current the tubes can generate, 
at the worst possible combination of tuning/loading/drive.

The worst case will be at intermediate levels of RF drive, so runaway is 
most likely to happen while you're tuning up or testing a new amp. If 
you can get safely through the critical period when you and the new amp 
are breaking each other in, then you'll be home free. So do bleed *lots* 
of standing current through the screen regulator of a new amp (you may 
be able to reduce it later, when you've found out how the amp behaves) 
and also use current trips, VDRs and everything else you can think of to 
protect the tube(s).




-- 
73 from Ian G3SEK         'In Practice' columnist for RadCom (RSGB)
                            Editor, 'The VHF/UHF DX Book'
http://www.ifwtech.co.uk/g3sek