I have printed out Jim's control list. I obtained a really sweet little item from the Mrlin P. Jones Company (big electronics catalog house) 800 652-6733, part #8418-KT, price $55.00.
http://www.mpja.com
The whole deal was in kit form and the visual appearance is that of a parallel port security dongle feed through or parallel port gender bender. What it is is a nice little, al in one A to D and digital i/o do-flotchy. it is an 8 channel, 12 bit A/D converter (8 single ended or four differential) with 4 digital outs and 4 digital ins. All run off of the parallel port. It comes complete with instructions and software including a Q-basic library and a formal running/testing environment in visual basic. Really nice!!!
I plan on making my own custom Q-basic program built around their driver library of subroutines. Actually the item is from Australia based on an article that appeared over there. (article included on disk with software).
Jim's list is fairly complete, but I have found that I am capable of cutting on and off most major long term components such as fore and diff pumps with a simple switch. So those will not be controlled by cpu.
I am planning on having pressures of both the fore line and chamber, temperature of the chamber body, voltage applied and current to the device formally displayed on screen. (all A/D conversions).
The key control to gain ascendency over will be setable and forgetable constant current built around carefully crafted closed loop servoing of the power. A simple electronic constant current control is no good, as the key to success will be the "designed in", multiple level hysteresis routines which can only be intelligently written from hands on past knowledge. (much like Steve noted)
An electronic control would be a huge noise maker in an area where noise is not needed (due to the need for a very electrically quite neutron counting arena.) Electronic controls respond in microseconds (lotsa' hash). We need slop..........lotsa slop. There will be auto shutdown routines at hot grid detect and anomolus pulse ignore routines. Sure it could be made electronic, but what a nightmare when program control is easy to modify over hard wiring on a pitching fusion ship deck in a storm.
Again, baby steps.......
Richard Hull
Created on Monday, March 12, 2001 2:50 PM EDT by Richard Hull