X-Message-Number: 2194
Date: Sun, 2 May 93 12:55:10 CDT
From: Brian Wowk <>
Subject: CRYONICS Temperature Stability
Will Dye:
> Insulation can change with time,
> especially with earthquakes, maintenance personnel clunking around,
> etc. Also, the interior of the room will change somewhat as new
> patients are added. In other words, you may have to re-tune your
> insulation periodically. That could be a real mess, since the
> only way I can think of involves 1) switching off power & letting
> LN2 take over, 2) taking LOTS of measurements all over the room
> to see if 3) any part of the insulation is getting things too cold
> or too hot. Even after you solve the measurement problems, letting
> things get too cold/too hot is bad for the patients.
I think there is a misunderstanding here concerning what
insulation needs to adjusted. The insulation to fine tune is not the
room insulation, but the *LN2 tank* insulation.
The LN2 backup system consists of a simple 50 gallon tank in
one of the heat exchange cells surrounded by a few inches of foam
insulation. The fan system circulates room air past the outside of
the tank. During commissioning of the room, you adjust the LN2 tank
insulation so that a steady state temperature of -130'C is achieved
while running in LN2 backup mode. Steady state operation exists when
the heat flow from outside the room (about 500 watts) equals the heat
flow into the LN2 tank. This "steady state" is inherently stable: If
the room temperature rises above -130'C, heat flow from the outside
decreases, and heat flow into the LN2 increases, thus lowering the
room temperature. If the room temperature goes below -130'C, the
reverse occurs.
Trymer foam insulation loses up to 30% of its insulating power
over a number of years that depends on how well you seal it from room
air. (This occurs because of air diffusing into the foam cells, and
diluting the CFC blowing agent.) Periodic (perhaps annual)
recalibration of the LN2 backup mode will therefore be necessary.
There are many possible ways of easily altering the heat transfer
charateristics ("insulation") of the LN2 tank during recalibration,
and room air flow temperature would vary by only a few degrees during
the process. Because of the way patients and ballast are insulated
from the room air flow (see discussion below), this temperature drift
during recalibration would have essentially no effect on patients.
> Maybe I'm being too paranoid. I just have these fears of turbulent
> airflow & heatflow creating semi-stable, cyclic structures (like
> the Great Red Spot of Jupiter) that introduce temperature cycling
> on the patients. Sure, the temperature differences would be small,
> but how small is small enough?
Let's review those aspects of my Cold Room design that make
patient temperature inherently uniform and stable. Heat flowing into
the top and bottom of the room is intercepted by -130'C over and under
air flows as shown in the diagram below. Heat flowing into the sides
of the room is intercepted by -130'C sheet aluminum panels that
conduct heat to the over/under airflow.
Side View
|--------------------------------------------------|
| Insulation |
| |
| A--------------------------------A | A
| A -130'C Overhead Air Flow A | A = 3mm sheet
| A--------------------------------A | A aluminum
| A Insulation A |
| A |------------------------| A |
| A | | A |
| A | Patients & Ballast | A |
| A | | A |
| A |------------------------| A |
| A Insulation A |
| A--------------------------------A |
| A -130'C Underfloor Air Flow A |
| A--------------------------------A |
| |
| Insulation |
|--------------------------------------------------|
Patients and ballast are thus surrounded by -130'C on all sides
(-129'C at the midpoint of the aluminum walls). It is therefore
physically impossible for their temperature to stabilize at anything
other than -130'C.
Note the insulation (4" foam) around the patient/ballast
compartment. This *strongly* protects patients from any spatial or
temporal variations in air temperature due to refrigerator cycling,
maintenance, or LN2 backup recalibration. In fact, this protection is
so great that airflow temperature could be raised all the way up to
0'C for several minutes at a time for defrosting operations. In fact,
you could defrost at 0'C for as long as one hour at 0'C with patient
temperatures rising by only a fraction of a degree. Clearly
temperature variations during LN2 recalibration would be insignificant
by comparison.
--- Brian Wowk
Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=2194