In this space, visitors are invited to post any comments, questions, or skeptical observations about Philo T. Farnsworth's contributions to the field of Nuclear Fusion research.
Subject: Computer control
Date: Apr 12, 2:53 pm
Poster: Jim Lux
On Apr 12, 2:53 pm, Jim Lux wrote:
Richard commented on difficult transducers in a previous post in reference to computerized control. Let's think about this...
What sort of dynamic requirements...
What do we need:
1) Gas pressure control (Simple, a stepper on a needle valve would probably do.)
2) Pump controls (relays...if you don't just run it continuous)
3) Electrical controls
The real problem here is controlling the current/voltage of the excitation supply, mostly because it IS a glow discharge and has a nonlinear characteristic (that is probably time varying to boot, as the grids heat up).
Probably what you want is a constant current supply (more than a constant voltage supply). Perhaps a switching supply would be the way to do this, with a SMALL smoothing inductor (so the dynamic response would be good). At work, I am looking at switching supplies for TWTA's right now (8-10 kV) and I will see if I can scavenge a decent easy to build design. (For this application, it doesn't need to be space qualified, faster, or better, but cheaper would be nice.) Switchers are nice because the stored energy is low, so a short or flashover isn't as exciting.
The other approach is a big bulk supply (transformer, bridge and LC filter) with a series or shunt regulator made from some appropriate HV vacuum tube. A hairball shunt regulator I've heard of is to use a whole string of CdS photocells in series (for the voltage). And, with a lot of stored energy, a flashover is VERY exciting.
Metering voltage and current is trivial. Voltage is a suitable divider. Current is one of those Honeywell or F.W.Bell or... hall effect current transducers.
If you go to a multi grid system, and higher vacuums, I suspect the control problem becomes MUCH easier, because you are out of the glow discharge regime. You basically have a vacuum tube diode then, and you can change the current by changing the filament current or accelerating grid potentials.
In summary, it seems that the tough part is the power supply. Anybody got good ideas on an inexpensive source for the transformers for a HV switching supply in the kilowatt range (say, 50 kV @ 20 mA). Somehow, I don't think a TV flyback's going to cut it... (But maybe a bunch of HEI coils in parallel/series?... they are in the 100-200 watt range)
Switchers, as a class, live or die by their magnetic components.
- Re: Computer control - Apr 12, 5:27 pm
- Re: Computer control (switching power supplies) - Jim Lux Apr 13, 11:30 am