Since this topic has some legs I'll expand a bit more on my previous post.
I acquired several LM1949 driver ICs from Newark (they seem to be pretty good to deal with and small orders are ok). Spec sheets and circuits are on the National Semiconductor web site.
Two essential papers are:
"Pulsed gas injection for on-line calibration of residual gas analyzers," B.R.F. Kendall, J. Vac. Sci. Technol A, Jan/Feb 1987.
"Controllable leaks using electrically pulsed valves," J.L Dobson and B.R.F. Kendall, J. Vac. Sci. Technol A, May/June 1990.
Dobson and Kendall used the Moto MC3484 which I've never found a source for. In their experiments, 2-6 millisecond pulse widths would dependably meter air at atmospheric inlet pressures at 40 to 140 microliters/pulse.
Helium at an inlet pressure of 10 Torr would yield 40 microliters/pulse for a 6 ms pulse. In the range of 1 to 10 Torr (flow substantially in transition region) volume (STP) is basically independent of the pressure.
Above about 200 Torr inlet pressure the flow is choked and approximately linear with pressure for a constant pulse width.
Bosch injectors, given the metal seals, have a closed conductance of about 10-5 to 10-6 liters/sec and the authors state that newer valves (1990 vintage) were better than that.
Duty cycles were adjustable to 75% (several hundred cycles per second) although reliability tests were confined to 10-20/sec.
Steve
Created on Friday, February 23, 2001 7:40 PM EDT by Steve Hansen