A IAA-8 2-0 077

A Cryogenic Valve for Spacecraft Applications

L.J. Salerno, NASA Ames Research Center,
Moffett Field, CA; and A.L. Spivak, Transbay
Downloaded by University of Michigan - Duderstadt Center on April 3, 2017 | http://arc.aiaa.org | DOI: 10.2514/6.1982-77

Electronics, Richmond, CA

,
AlAA 20th Aerospace
Sciences Meeting
ihn~e~~leb-i January 11-1 4, 1982/0rlando, Florida

For permission to copy or republish, contact the American Institute of Aeronautics and Astronautics
1290 Avenue of the Americas, New York, N Y 10104
 A CRYOGENIC VALVE FOR SPACECRAFT APPLICATIONS

L . J. Salerno"
Ames Research Center, NASA, Moffett Field, California

and

A . L. Spivakk
Transbay Electronics, I n c . , Richmond, California

Abstract The approach described here employs a remote1.y

controllable cryogenic valve which was fabricated
With the current interest in infrared astron- using commercia1l.y available components. The
omy and the number of proposed orbiting cryogeni- objective was to develop an inexpensivc ve1.v~whi.ch
cally cooled infrared telescopes, n requirement has could meet project specifications. The present
developed for cryogenic valves to operate between valve configuration lends itself to applications
room and liquid helium temperatures. in such projects a s Cosmic Background Explorer
(COBE), Shuttle Infrared Telescope Facility
(SIRTF) , German Infrared Laboratory (GIRL), and
Downloaded by University of Michigan - Duderstadt Center on April 3, 2017 | http://arc.aiaa.org | DOI: 10.2514/6.1982-77

A remotely controllable cryogenic v a l v e is

described which employs commercially available Gravity Probe B (GPB), all of which require
components. The bellows-type stainless-steel cryogenic systems.
valve is operated by n miniature dc motor and
integral gearset. Coupled to a power transmission, A standard cryogenic angle valve was used as
an overall gear ratio of nearly 3 5 0 0 : l is realized the major component. The valve body was machined
assuring reliability of operation at low to reduce its weight, and the valve seat material
temperatures. was changed to copper. The valve was equipped
with a motor and integral planetary gearset. In
Valve performance (leak rate) data are pre- addition, a powcr transmission for the valve
sented at LN2, LHe, and S f H e temperatures with assembly provided further g e a r reduction e n s u r i n g
torque as a parameter for valve closing torques of reliable seating and sufficient operating torque
18 N-m, 27 N-m, 31 N-m, and 35 N-m. With a clos- at cryogenic temperatures. The valve assembly
ing torque of 31 N-rn, a leak rate of w a s housed in an evacuated test Dewar which allowed
2.8 x s c c / s e c was achieved at 2 K, while with cooling to liquid helium temperatures.
n torque o f 18 N-m, the leak rate at 300 K was
less than 3 x 1 0 O s c c i s e c . A small computer was employed to automate
valve testing and assure consistency in data acqui-
sition. valve performance was evaluated on the
Introduction basis of reliability of operation and of leak rate
data at room temperature, liquid nitrogen tempera-
With the current interest in infrared astron- ture, liquid helium temperature, and superfluid
omy and the advent of superfluid helium cooling helium temperature. Since the currently proposed
systems for orbiting astronomical telescopes, projects have no set requirement for leak rate
requirements have developed f o r space-compatible performance, criteria set forth in th IRnS (Infra-
cryogenic valves to operate at both room and red Astronomical Satellite) project were chosen
liquid helium temperatures. to evaluate the valve. These criteria call f o r a
leak r3tc o f less t h a n 5 x 10-* s c c s at room tem-
In a spacecraft system it is advantageous to perature and 0.3 sccs at superfluid helium temper-
have the Dewar completely filled with superfluid ature. Mechanical considerations such as weight
helium below atmospheric pressure and valved off. and volumc were of secondary importance.
There are two reasons why this is desirable.
First, if the Dewar is filled at atmospheric p r e s -
sure and subsequently evacuated in orbit, there is Apparatus
some cryogen lost through boil-off. Additionally,
the density of liquid helium at superfluid tempera- Figure 1 shows the cryogenic valve assembly as
tures is Significantly greater than at 4 . 2 K. installed in the Dewar before testing. The motor,
Combined, these factors can produce a cryogen loss with integral gearset ( 4 8 5 : 1 ) , operates from a
of approximately 1/3. Since cryogen lifetime is nominal voltage of 28 Vdc. The power transmission
an important consideration in a space mission, provides a further reduction of 7.2:l resulting in
loss must be minimized. A cryogenic valve pro- an overall gear ratio of nearly 3500:l. This ratio
vides an effective solution to the problem of was selected on the basis of testing, to provide
containing superfluid helium in a flight Dewar. for uniform and consistent operation at cryogenic
At least one cryogenic valve exists in the liter- temperatures. An earlier prototype with lower
ature,' but to date, v a l v e s of the Size and con- torque had w o r ~ eleak rate performance due to
figuration described here have not been reported, insufficient torque. The present gear ratio pro-
although one manufacturer is working on such a vided an effective solution.
design (R. Hopkins, Ball Aerospace Systems Divi-
sion, Boulder, Colorado, private communication). The valve itself is a bellows-type d o u b l e
sealed angle valve. Seating is accomplished by
*Research Scientist. Member AIAA. means of an actuator employing Acme power trans-
.:Research Technician. mission threads. A copper valve seat was

This paper is declared a work of the US.Government and

therefore is in the public domain. +Nupro SSBUW, Nupro Company, Willoughby, Ohio.

1
 A Hewlett-Packard 9825 calculator was employed
to cycle the valve, take data from the temperature
sensors, and record the leak rate as well as per-
forming secondary functions necessary to system
operations.

Figure 2 shows the test setup for the cryogenic

valve system. The helium gas supplied for testing
helium was gaseous technical 99.0% pure, Navy
Grade A. A pressure regulator maintained a constant
pressure of 760 torr (14.7 psia). Electric solenoid
valves were installed in the entrance and exit lines
of the test Dewar with manual valves for backup
during adjustments. A 0-50 sccm flowmeter monitored
the leak rate with the ability of installing a
conventional mass spectrometer-type le& detector if
warranted. A vacuum vessel was connected to a two-
stage rotary-type mechanical pump to ensure that no
a i r could enter the system due to cryopumping of the
test Dewar and plug the lines.
Downloaded by University of Michigan - Duderstadt Center on April 3, 2017 | http://arc.aiaa.org | DOI: 10.2514/6.1982-77

Experimental Procedure

The valve was leak checked at 300 K and flushed

with gaseous helium. The Dewar was then evacuated
to 5 x lo-' torr and cooled to 7 1 K. The valve was
cycled several times to ensure uniformity of opera-
tion, and the leak rate measurement cycle was ini-
Fig. 1 Cryogenic valve assembly.
tiated. The cycle began by closing the cryogenic
valve and allowing it to seat. Seating torque was
set by limited presetting of the power supply cur-
installed for the testing reported here. The rent drawn by the motor. This stall current was
valve was modified for low-temperature operation
maintained at the end of the cycle for 1 sec thereby
by disassembly and lubrication with molybdenum
assuring valve seating. A variety of motor currents
disulfide, after removal of existing petroleum-
was selected as part of the test program. At each
based lubricants, which become glassy at low
temperature, the goal was to test the valve at motor
temperatures. The Acme threads were burnished
'J with Teflon. currents of 1.0, 1.5, 1.75, and 2.0 A (corresponding
to delivered torques of 1 8 , 2 7 , 31, and 35 N-m,
respectively) to determine the effect of closing
Silicon diode temperature SenSOKS were
torque on leak rate.
installed on the valve body, cold plate, and motor.
TO minimize heating of the valve assembly, the
After valve seating, the leak rate was moni-
motor and gearset were thermally anchored to the
tored. When the change i n successive (GO-sec
cold plate by a copper braid, and the frame was
intervals) leak rate measurements dropped to less
anchored by additional thermal straps. A s well,
than 0.05 sccm, the leak rate was assumed to have
the valve inlet and outlet lines were coiled and
stabilized, the test was concluded, and data were
thermally anchored to the cold plate a s was the
recorded. By closing the solenoid valves after
valve body itself. Indium was used between all
joints for maximum heat transfer. each run. a trapped volume of gas remained within

0-50 psi ABSOLUTE

PRESSURE GAUGE
DEWAR DEWAR
r;\ WALL WALL
PRESSURE
REGULATOR
I 77K,SHlELD
VACUUM
VESSEL
A

SOLENOID
VALVE
VALVE MECHANICAL
DC
PUMP
HEAT SINK TO HEAT SINK TO
COLD PLATE COLD PLATE
-

2
 the cryogenic valve assuring further cooling, an In examining Fig. 3 more closely, it is appar-
important considefation at superfluid temperatures. ent that a closing torque of 18 N-m is insufficient
Testing was performed at 77 K 4.2 K, and 2 K. Tem- to meet the established performance requirements
perature determination below 4 . 2 K was made not whereas raising the torque to 31 N-m exceeds the
only with the silicon diode temperature Sensors but requirement by an order of magnitude. Addition- b
also by monitoring the test Dewar helium pressure. ally, down to 4 . 2 K, the 31 N-m and 35 N-m data are,
After the runs were completed, the sensor mounted for a l l purposes, indistinguishable within the
on the valve body was checked and, if necessary, range of values obtained while the 2 1 N-m data is
recalibrated against vapor pressure data to assure only slightly less desirable. This is significant,
accuracy of the results obtained. since power requirements in particular space mis-
sions may dictate selection of specific motorlgear-
set combinations and different gearsets which would
Results and Discussion affect the time required for opening and closure.
In these cases, satisfactory operation could still
Figure 3 shows leak rate data obtained as a be provided while power consumption was maintained
function of temperature with closing torque a s a at an acceptable level, an important consideration
parameter. Lines connecting the data points are in long duration missions. As a final note, the
intended as an aid to the eye and should not be use of Navy Grade A helium for leak rate measure-
construed as representing actual or interpolated ments poses an interesting consideration. The dis-
Downloaded by University of Michigan - Duderstadt Center on April 3, 2017 | http://arc.aiaa.org | DOI: 10.2514/6.1982-77

data. Data for room-temperature operation are not tributor of the particular gas employed verified
presented since with 1 A of motor current (18 N-m that the sample contained impurities of 6 ppm N 2 ,
torque), the leak rate at 300 K was less than 2 ppm 0 2 , 1.5 ppm H20 vapor, and 0 . 5 ppm unspeci-
3 x IO-' s c c l s e c (below the limit of the leak fied impurities. A liquid nitrogen trap between
detector). At 2 K, data were taken with a motor the helium supply and the cryogenic valve inlet
current of 1.15 A (31 N-m torque) only since resid- served to remove the oxygen and water vapor, but
u a l friction in the motor assembly prevented smooth nitrogen and other possible impurities such as
operation at currents less than this value. At hydrogen may have entered the valve and could pos-
2.0 A , the life cycle of the motor was consider- sibly have augmented leak rate performance at 2 K.
ably shortened, a consequence of operating at Clearly, additional testing is warranted to deter-
nearly double the manufacturer's published stall mine what, if any, effect these impurities have on
current rating. At this high Current, brush leak rate values. It is suggested that boil-off
damage due to heat buildup and binder outgassing gas from a helium supply Dewar be used for testing
limited motor lifetime to 10-15 cycles. While this to assure high purity. A special grade high purity
may seem of concern at first glance, it should be bottled helium gas could be employed; however, over
noted that the motor employed was a standard, com- a large number of cycles, this could prove expen-
mercially available item, and in actual practice, sive since a significant volume of gas is passed
stringent low-temperature performance specifica- through the valve during each cycle to assure that
tions would require that motor selection be based
on established test results at 2 K. Different
seating occurs while there is a flow through the
valve.
u
brush materials or treatment, as well as special
lubrication techniques, would be needed to assure
motor reliability. Conclusion

LEGEND The data reported indicate that a motorized

0 18 N-m valve can be designed to operate in a cryogenic
System over a temperature range from 300 K to 2 K
V 27 N-m
with acceptable 1.eak rate characteristics.
0 31 N-m
A35 N-m Applications in future space projects may dic-
tate extended performance requirements as well as
0 different configurations. AS more sophisticated
5 optical systems are developed, specifically infra-

.
x
5 100 - red telescopes for deep space operation, remotely
controlled cryogenic valves will offer a method of
Y maximizing cryogen lifetime and mission effective-
n ness. I n addition to operation in the main cryogen
c
VI
r3 - systems, these valves can be utilized in support
equipment such a s helium dilution refrigerators,
c
U cryostats, and adiabatic demagnetization
a refrigeration.
Y
2
A
Reference
10 :
'Roach. P. R.. Ketterson. T. B.. and Kuchnir.
- M., "Hydraulically Actuated Valve for Very Low
. Temperatures," Review of Scientific Instruments,
41 d I Vol. 4 3 , No. 6, June 1972, pp. 898-899.
1 10 100
TEMPERATURE, K

Fig. 3 Temperature dependence of leak rate with

closing torque as B parameter. Error bars denote
the range of values achieved.