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: Re: Grid loss not losses (cooling)
Date: Jul 14, 1:28 pm
Poster: Jim Lux
On Jul 14, 1:28 pm, Jim Lux wrote:
>All of the above were suggested here in Richmond last year by one of our groups of amateur scientists. (I have a meeting every month of science types who putter around with high energy physics.) She suggested hypodermic tubing also with flowing LN2!!! Boy that is really out there for the amateur budget!!
Hypodermic tubing isn't all that expensive.. And, what is the tradeoff on higher cooling power vs increased area blocked?
LN2 isn't all that impractical.. If you are only looking at 20-30 watts being dissipated... heat of vaporization for LN2 is probably a few hundred J/g (water=2200J/g, R12= 130 J/g), so you'd boil less than a gram per second of LN2...A liter would last 10-20 minutes...
But what is wrong with good old DI water? It is an insulator, and fairly good heat transfer medium.
>alcohol fluid cooled by dry ice or such to
Do you need to be that cold? If now, you are running hot (glowing), just running at room temp might be a signficant difference.
>>A suitable coolant might be high pressure hydrogen gas.. low viscosity, high conductivity,
>The hydrogen gas is a bit scary especially at high pressures. That pressure won't go thru the insulating tygon and a leak would be disasterous.
As if generating millions of neutrons and having HV around isn't scary enough???
>The actual power in the inner grid is minimal compared to the disappation of the outer wall. In theory, it would seem that they would both deal with the same power. They do of course carry the same energy, being a simple series circuit, but the electrons accelerate to large kinetic energies and slam into the outer wall casing heating it to very high levels. The heating of the inner grid is by ion bombardment and at the point of current runaway by field emission.
IR losses would be quite low (even with tungsten),, the currents are pretty low.
Probably mostly ion bombardment...
The KE of the electrons is the same as the KE of the ions, though (same accelerating potential)..
If you make an assumption of low temps (reasonable for the outer wall), radiative transfer should be quite low (since it is T^4).
Then, you are just looking at simple conduction from the inner grid to the outside world (or to a coolant)...
A useful comparison might be X-ray tubes, which get hot anodes for the same reason, eh? A typical Xray tube is cooled by conduction through a copper bar, but is dissipating some 10 kW when it is on (100 kV @ 100 mA).
What about making the grid out of fins (with the long axis radial from the center of the chamber) rather than wires? The (impractical) idea being that the ions (at least the ones we are interested in) are heading in and out and the cross section in that direction is small...
For precision, and based on grid construction, the thing would be a night mare of resisitive wye-delta matrices. Still, that's better than the reactive wye-deltas which drove me nuts in college years ago (during the age of the huge log-log EE slide rules)
Yes.. but this is what computers programs (like Excel) that solve linear equations are for... Put it in, let it chug..