N2 Cold exchangers
Gerald Pearson (gpearson@umaxc.weeg.uiowa.edu)
Wed, 13 Oct 93 15:20:25 -0500
Here in Iowa City for several months during each summer, neither the "cold
dissipators" (made from an aluminum extrusion) from Bruker nor the sponge
rubber pad trick prevents condensed water from dripping onto the magnet can
and running down the side. Since our rooms are air-conditioned, I find it
hard to believe that the humidity _everywhere_ else in the world is _always_
low enough so that no one else has this problem.
We've now got our machine-shop and electronics-shop guys building an
_electrically_ _heated_ de-icing device which also has Al fins. The heater is
a wire-wound resistor which will supply just enough heat to prevent water from
_ever_ condensing on the Al block and fins. The fins ought to prevent the
block from _ever_ becoming more than slightly warm, even with no N2 flow.
Once we've tried out the first one and made any appropriate design changes, we
plan to have 11 more made up for all the N2 ports on our MSL-300, AC-300, WM-
360, and AMX-600. I got started on this idea when Mike Abell (former GE
engineer) told us about the electrically heated (no fins, but with thermal
cut-out for the heater) defroster blocks-with-holes which he designed and
which GE put on their N2 ports for many years.
If anyone is interested, I could supply details after we've worked them out.
N2 LIQUID-LEVEL SENSORS: A simple, cheap, and reliable sensor can be based
on an ordinary 1/8-watt _carbon_ _composition_ resistor. When driven to it's
rated power, the resistance decreases sharply by several percent as it's
raised above the surface of the liquid. (e.g. For use with a 7.5-volt power
supply, R=(7.5*7.5/0.125)=450 ohms, so use a 470 ohm resistor.) If the
resistor is connected to a bridge circuit, either a meter deflection or a lit
LED could indicate that the resistor is still "wet". (I used this trick
extensively in the 1960's to monitor boil-off and cryogen fills of both N2 and
He in cryostats, usually connecting only a battery and milliammeter to the
resistor. In 8 years, I never saw a resistor fail in this application.) A
slightly more elaborate circuit could verify that a "wet" resistor has a
"reasonable" resistance, and therefore the sensor is OK. Use insulated
constantan wire (sold for use as thermocouple wire) instead of Cu wire to
connect the resistor to the outside world, to avoid significant heat leak down
the wires. The electrical heating should have negligible effect on the N2
hold time. Even if you had 3 resistors running continuously to monitor "low",
"medium", and "full" levels, the 3/8-watt of sensor power would boil off only
about 170 grams/day of N2.
--------------------------------------------------------------------
Gerald A. Pearson INTERNET: gpearson@umaxc.weeg.uiowa.edu
Chem. Dept., Univ. of Iowa VOICE: 319-335-1336
Iowa City, IA 52242-1219, USA FAX: 319-335-1270
--------------------------------------------------------------------