X-Message-Number: 2010
Date: 23 Mar 93 06:12:00 EST
From: "Steven B. Harris" <>
Subject: Nifty Idea Plus...
To Brian Wowk:
I'm glad you reconsidered the idea of boiling LN2 with an
electric filament to get cold vapor: it's a 3 to 1 waste of
power, of course, for the cooling you get (so it really would
have to be just a fraction of a percent of the total boiloff, as
you say). Pumping power with a thermopile is a better idea, but
I wonder if such things work well at such small temp gradients.
Your "nifty idea" that liquid nitrogen should be kept in an
internal reservoir insulated JUST well enough so that boiloff
absorbs about the amount of heat that passes through the walls of
the main storage room at -135 C, is great. And yes, it really
looks as though this internal liquid nitrogen reservoir had
better be in the ceiling, for passive convection-stirring
reasons. Maybe the entire ceiling should be a thick piece of
insulation, above which is a (thin) liquid nitrogen layer. The
upper liquid nitrogen reservoir will have to have even thicker
walls (side and top) than the -135 room, to optimize costs.
How much foam should be between upper LN2 chamber and
lower -135 C chamber, with this large surface area? Let me see:
you want as much heat to flow into the LN2 through a ceiling
which is about 1/5th as large as the total remaining wall area
(through which heat is flowing into the main box from the
environment), but you want this to happen with a temp gradient
only 1/4th as large as between main room and ambient environment.
So I would guess as first approximation that the foam between LN2
and main chamber should be about as thick as insulates the main
chamber itself (MUCH more than 5 inches). With TWO reservoirs at
different temps and several constraints this would be a really
interesting minimax problem. You'd want the LN2 layer to be as
large as possible for good convection, but as thin as possible to
minimize the amount of needed side insulation, subject to the
constraint that you want a certain reservoir size big enough that
you don't have to fill more than once a week.
But wait: no need to work this out yet. Why have the actual
liquid nitrogen spread out in a thin layer of ceiling area, when
all you need is the *temperature* of it, over that area? Here's
a small improvement idea: all you really need is a conductive
metal plate that size (the whole ceiling area) at liquid nitrogen
temperature, insulated from the main room by a meter of foam or
so, and insulated from outside ambient temps by even more. The
embedded conductive metal ceiling plate can simply form part of
the metal bottom of a tall, thin cylindrical foam LN2 reservoir
(no moving parts or pipes) sitting on top of the main room, and
of a volume to give the proper capacity to give a weekly fill
time, with dimensions picked to maximize insulation for the
money. Not a steel drum, but a foam tube sitting on a much
larger heat-sink disk or metal square. Help us on the numbers
for this. Since the LN2 reservoir itself will be relatively
small, one can (should) spend more lavishly on foam thickness
here than the 40% greater temp gradient alone would suggest.
Again, you suggest we should also pick numbers here so that
natural LN2 boiloff is not *quite* enough to give you the -135
main room temp you seek, leaving a little extra cooling to be
done by your fine control system. Again, as you note, the
advantage of main cooling systems such as these over those based
around a Dewar holding the LN2, is that this all works without
any valve or control.
LN2 fill --> -----
| |
----------------| |----------------
| |~~~~~| |
| Foam | LN2 | Foam |
| ______________|_____|______________ |
| ^Conductor Plate^ |
|_______________________________________|
| |
| | | | |
| v v v |
| convection |
| |
| |
| |
| Insulated Main Storage |
| -135 C |
| |
| |
| |
+_______________________________________+
To Bobby Hesselbo:
Your economies of scale remind strongly of the giant freezers
envisioned in Ettinger's original book. It's probably eventually
going to come to that, somewhere (now where were those abandoned
missile silos for sale...?)
Comment To Rick Shroeppel:
You missed the thread of a while ago in which we discussed
ethyl chloride thermal ballasts, which operate nicely at -135 C
using a cheap and relatively nontoxic (though still flammable)
liquid. Clearly, there will have to be a lot of ethyl chloride
around in a good stable -135 C cryo-room, although patients will
probably not be floating in it. Maybe someone can make us some
small sealed plastic C2H5Cl pouches of the dinosaur-egg variety,
however, and they can be used as miscellaneous packing in patient
pods, etc.
Finally, while we're on the subject, I will admit that
although in the past I have criticized temp control schemes which
use no electronics as being silly in sort of Rousseauvian way in
this modern world, still it does occur to me that there might be
a way to use ethyl chloride to passively control that bit of
extra LN2 ingress that one would need for fine control in a -135
C room. Mike Darwin points out that there are pressure driven
impeller pumps which could be put in-line on an LN2 system (a low
pressure LS-160 line, say) to generate power directly for a fan
via an external connecting shaft. I'm thinking that such an
external fan could also spin in a container of liquid ethyl
chloride. When the ethyl chloride slushed/froze the fan
would be stopped, and the impeller in turn would freeze, stopping
LN2 flow from the small storage Dewar into the main room until
the ethyl chloride liquified again. One moving part in this, but
on a small scale, duplicated, and with a main passive convection
backup as described above doing most of the cooling work without
it, it might be acceptable.
Steve Harris
Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=2010