Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: [Phys-l] Ionization type smoke detectors.



I said,
"If the integration time of the charge-sensitive
amplifier is sufficiently long for the ions
to clear out of the chamber, then the track orientation
in the chamber becomes very important.
Tracks that make ions in one end of the chamber
can create a much larger integrated signal than tracks
created in the opposite end of the chamber."

Bernard said,
"My understanding of this must be deficient,
as I think you've got this backwards."

I think my explanation was more deficient than Bernard's understanding.

I should have said,
(1) Assume the + and - ions have different mobility.
(2) Assume the integration time is sufficient for the faster ions
to clear out of the chamber, but not the slower ones.
(3) Now we can say that track orientation is important.

Integrated signal is n*q*v*t = n*q*d.
Ideally this is true for both + and -.
Total integrated signal is n*q*d(+) + n*q*d(-) = n*q*[d(+) + d(-)]
d(+) + d(-) = chamber length L
Total integrated signal is n*q*L.
This assumes integrated time allows both charges to clear.

If only one sign of charge clears in the integration time (because the other charge is too slow) then the signal when the faster charge has the longer way to go will be larger than the signal when the slower charge has the longer way to go. When I wrote the previous message I had my mind fixed on my liquid argon detector in which the Ar+ was almost immobile, and the electrons were very mobile. But I neglected to say that. My integration times were about 2 us and this allowed the full n*q*d(-) to be detected regardless of whether d(-) was large or small, but n*q*d(+) was not seen. Therefore if the track occurred near the negative electrode, d(-) was large and I got a big signal. If the track occurred near the positive electrode then d(-) was small and I got a small signal. The only practical way to fix this was to add a grid, make sure all tracks were between the grid and the negative electrode, and then the detected path of the electrons was always the same length (from the grid to the positive electrode).

This all happened 28 years ago, so I have to remember it bit by bit. Sorry.

Michael D. Edmiston, Ph.D.
Professor of Physics and Chemistry
Bluffton University
Bluffton, OH 45817
(419)-358-3270
edmiston@bluffton.edu