813 tube coil

From:  Alan Sharp [SMTP:AlanSharp-at-compuserve-dot-com]
Sent:  Tuesday, June 02, 1998 11:08 AM
To:  INTERNET:tesla-at-pupman-dot-com
Subject:  813 tube coil

Thanks to those who responded to my questions
about the 813 valve.

I've got some data and ideas now. I think I've got
the filament sorted out. The filament doubles as the cathode,
so I need a 5-0-5V supply - to keep everything symetrical.
I also need to adjust this to 5A per tube, I'll also use an inrush current
limiter -
to avoid a switch on surge shortening the life of the filaments.

Naturally no one supplies 5-0-5V transformers at 10A - so
I'll probably use a 100VA transformer kit and wind my own secondary -
(then I can put on / take of turns to get to the exact current/voltage).

I've got Brent Turner's book (1992 edition) - I'll use the design 
given there - monkey see, monkey do - I'm not going to try for
an original design with my first valve circuit. The design is for a
triode but I've got 2 extra grids on a pentode. 

Looking at the pin out:

Pin 1,7 Filament, Cathode - OK got these figured.

Pin  3. Screen Grid - what do I do with this? Tie to control grid? Bias to
of anode voltage?
Pin 4 Control Grid -  I think this is the same as triode grid.
Pin 5 Beam Plate - is this the same as the suppressor grid - connect to

Top cap: Plate - Anode?

Peter suggested: 
>the 1st grid went to ground via the grid leak
>resistor and cap. The second grid was tied to the HT via a +/-20k
>resistor. The third grid was tied to the filaments.  I used AC direct
>from a microwave xformer.

I think I've got the main PSU figured,
1) Rig it so that it cannot be switched on unless the heaters are on.
2) 2200V AC microwave transformer fed off my 4A variac,
no need to rectify - the tubes will do that but I can experiment with DC
1A meter to monitor HV current.

Next get hold of a tuning cap - 500pf at 5KV - or build one! I've got
Some 1000pF 28KV mica caps which will be more than adequate
for the circuit.

Does this look right?

Have fun,

Alan Sharp, UK