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.

"Ludwik commented that ionization chamber detected current is
proportional to n*q*v."

I'm a bit confused: Isn't nqv, just J the definition of current density (I / area)? So no surprise?

Yes, the anode is like a probe detecting the separation of charge in the chamber. when the charges alight on the electrodes they are neutralized and the EMF drops to zero, well, the equilibrium value. Thanks ME, I had never thought about it. This maybe? explains why using the PD on a low value resistor in the cathode circuit gives the same result as from a capacitor at the anode. (Inverted, of course.) Another reason for a high EMF is to turn it into a prop. counter, of which ME is certainly aware. I suspect they are less sensitive (and more expensive) as a smoke detector, tho.

bc, doesn't remember in the dozens of articles about counters (drift *, wire, prop., G-M, etc. ever reading about EM's / LK's point.

p.s. a paradox? conservation of charge requires the # of + = - ions? And if the - are more mobile, then the currents (or Js) aren't =? What am I missing?

* "When this avalanche reaches the positive electrode, it gives rise to a measurable current, the size of which is proportional to the original number of ions created. The ratio between the final number of electrons collected and the initial number deposited is called the gas gain, and for practical detectors is typically on the order of 10^4-10^6." A typically erroneous description?

Edmiston, Mike wrote:

A couple follow-up comments...

* * * CO Monitors * * *

Don Polvani wondered if a regular smoke detector would be just as good
at detecting carbon monoxide (from problems with a gas furnace or gas
water heater) as specific CO monitors one can purchase.

Certainly an ionization smoke detector could detect a malfunctioning
natural-gas furnace if a reasonable amount of products of combustion
would be getting into the household air near the furnace. However, the


* * * Ion velocity in ionization chambers * * *

Ludwik commented that ionization chamber detected current is
proportional to n*q*v.

Quite correct. I remember when this was a surprise to me, and I suspect
others might likewise be surprised. In a former life, I designed,
built, tested nuclear detectors. Many were ionization chambers.

When I first started working with ionization chambers I assumed the ions
travel to the electrodes, hit the electrodes, and that was when they
were detected. This is wrong. Once the ions hit the electrode, the
detection is finished (not started). The detection of ions in an
ionization chamber results from "seeing" the charges move through the
space between the electrodes. In order to get high signal-to-noise
ratio, you not only prefer large n and large q, you also want large
drift velocity. One way to help get large drift velocity is high
voltage across the chamber, and that is one big reason for operating
ionization detectors at high voltage.

Another velocity determining factor is the ionization medium. I did a
lot of work with liquid argon as the ionization medium, and I measured a
lot of drift velocities of electrons in liquid argon as a function of
temperature, pressure, purity, etc. A problem with using liquid argon
as an ionization-chamber medium is only half the ions are mobile. When
the ionizing radiation produces Ar+ and electrons, the electrons are
quite mobile, but the Ar+ is not. That means the signal from Ar+ is
practically nil. By contrast, a silicon or germanium radiation detector
is also an ionization chamber with solid Si or Ge as the medium, and
here the charge carriers are electrons and holes. Both are very mobile,
so the signal is at least twice the size from a Si or Ge chamber as from
a liquid-Ar chamber.

Michael D. Edmiston, Ph.D.
Professor of Chemistry and Physics
Bluffton University
Bluffton, OH 45817
Forum for Physics Educators