by Hardik Dave, ESC (QNA)/KSC/NASA
In the 21st century, NASA and the world have started looking into deep space astrophysics. Acquiring deep space images requires a space telescope with a large optical mirror, such as the James Webb Space Telescope (JWST), the Planck telescope, the Herschel telescope, etc. To carry out various space missions at the beginning of the century, NASA needed to be able to do large diameter space telescope testing at ground level in a large diameter thermal vacuum chamber (THVC) that required 20K helium shroud cooling and a vibration-free clean high vacuum of 1 x 10-6 torr suited to space simulation.
Because it would be costly to build a new chamber, the decision was made to consider using existing large diameter THVCs with diffusion pumps. However, because the old chambers designed and fabricated by NASA in 1960 were based on diffusion pumps, which are unacceptable because of oil contamination that can migrate onto the optical element and mirror, they would have to be replaced. From the family of clean high vacuum pumps available, two types of pumps—a titanium sublimation pump and an ion pump—could be considered, but they have their own limitations for a large chamber and are not the ultimate answer for a high performance vacuum system. Therefore, a cryopump system can be considered the best available and most economical clean high vacuum pump.The Space Power Facility at Plum Brook Station, NASA GRC, is one of the world’s largest chambers, with a 100-foot diameter and 120-foot height, requiring a contamination-free clean high vacuum range of 1 x 10-6 torr for the space payload of NASA’s next mission or to test any large space telescope. The large Chamber A at NASA Johnson Space Center, in which the JWST test is planned to be done, has similar requirements. The vacuum chamber has a large volume of around 20 million liters or more, and both chambers require a cryopump to replace the diffusion pump to achieve the clean high vacuum required. The JWST qualification test, the final thermal vacuum test to simulate deep space environment at ground level before its launch, has the same requirement. I was involved as a design engineer for cryopump design of both the above-mentioned chambers and am currently providing support as a cryogenic/mechanical engineer at NASA Kennedy Space Center Exploration and Space Communications (ESC). All details provided in this article are either based on published data or the author’s knowledge and judgments formed from his experience.
It is comparatively easy to design a new cryopump vacuum system, but difficult to modify an existing vacuum one. No single source of information or software is available that provides a complete design solution for a specific complex cryogenic system. The challenge for the design was to provide cryopumping capability and to determine how many cryopumps and what pumping speed was required. The cryo-pump is a major part and needs to be correctly matched to the process. This type of large volume chamber includes an LN2 and 20K helium shroud and the space telescope supporting structure, so the designer has to calculate the total gas load (outgassing + permutation + virtual leak) of the chamber and all components.
Vacuum: At the starting point, the chamber is filled with an atmosphere of air (nitrogen, oxygen, water vapor, etc.). A roughing system can be used to achieve 10-3 torr vacuum, the point where the water vapor becomes the predominant partial pressure—98 percent of the total pressure. To achieve high vacuum, the cryopump is required to take care of the large water vapor load, including the rest of the gas molecules as condensable gases and non-condensable gas molecules.
How the cryopump works: The cryopump is a gas-capture-type high vacuum pump using the Gifford-McMahon refrigeration technique to achieve cryogenic temperatures of helium as low as 10 to 20K. This type of pump, connected to an external helium compressor, produces a two-stage cooling effect. The cryo-compressor, vacuum vessel, two-stage cold head cylinder and drive unit displacer together produce a GM closed-cycle helium refrigeration system enabling the cryopump to create a clean high vacuum to simulate the space environment. During operation, high pressure helium from the compressor enters the cold head, then flows through the displacer-regenerator assembly, crankcase, and back to the compressor. The expansion of helium gas takes place in the specially designed displacer-regenerator assembly and provides cooling to the first-stage and second-stage cold station.The cryopump includes a first-stage 80K array which condenses water vapor and hydrocarbon vapor. The first stage 80K also consists of a thin, specifically designed and optically dense radiation shield to reduce heat load. The manufacture of a specially designed adsorbing array of 10 to 20K creates the second-stage cold station inside the vacuum vessel. The 10 to 20K array traps nitrogen, oxygen and argon, while the specially processed charcoal adsorbing array traps helium, hydrogen and neon. It also requires a regeneration rough vacuum pump to exhaust all absorbed gases into the atmosphere when the cryopump gets saturated and can no longer perform its work. This process requires an electro-pneumatically operated clean high vacuum isolation valve, leak tested for a leak rate of 1 x 10-9 torr-liters/sec with helium mass spectrometer leak detection for chamber isolation, connecting the cryopump to the chamber. The connecting pipe must be a large diameter with short length to reduce conductance losses at the molecular gas flow region and achieve maximum pumping capacity.
We need to consider crossover pressure and throughput of the cryopump. Basic engineering formulas indicate that we need large numbers of cryopumps that require detailed design evaluation to provide the most economical, safe and reliable cryopump vacuum system.
The pumping speed of the cryopump can be defined using the standard equation:
S = Q/P, where S = pumping speed (liters/sec), Q = total gas load (torr-liters/sec), P = vacuum pressure (torr) and the conductance losses as applicable. Different formulae are available for each specific calculation approach. Let us assume we need to design a system to take care of a total gas load of 18 t-L/sec and to achieve a design vacuum pressure of 1×10-5 torr. It would require a cryopump effective pumping speed as follows: pumping speed S = Q / P = (18 tL/s)/(1×10-5 torr) = 1,800,000 L/sec of air by cryopump.
Cryopump sizing: To achieve a very high pumping speed, we require a specially designed large cryopump with an approximately 50-inch diameter and an additional liquid nitrogen circulation cooled shroud. Normally this LN2 shroud works at 100K to 140K temperatures and each square centimeter area may provide about 10 to 15 L/sec water vapor pumping speed, depending on the manufacturer’s design.
Manufacturers provide cryopump capability data for different gases like N2, H2, Ar, He, water vapor, etc., but the data is confusing because it depends on many other factors, such as pump geometry, high vacuum valve layout and conductance losses. Therefore, the designer must complete a detailed technical evaluation for net effective pumping speed of the cryopump at high vacuum range in the molecular flow region for each gas molecule. Now we are monitoring parts per millions of each gas molecule particle. The pumping speed of each gas varies due to its mass number, such as nitrogen (mass=28) and argon (mass=40), and heavier gases that have lower pumping speeds. Effective pumping speed depends on the geometry of the cryopump installed on the chamber.Cryopump selection: Heat load capacity determines the gas pumping speed required in order to achieve vacuum. The conversion of cryopump capability from heat units of watts to vacuum units—liters/sec of pumping speed—is another interesting topic, but it is beyond the scope of this article. Commercially available cryopumps use large cryo-refrigerators, each having an approximate estimated heat load capacity, depending on the manufacturer, or as outlined in Figure 1.
For nitrogen gas, cryopump capacity at a temperature of 20K will require 12 to 15 watts of heat dissipation. This can take a maximum gas load of approximately 7 to 20 torr-liters/sec without any type of cryopump failure and up to approximately 5.1 x 10-3 or -4 torr vacuum. The commercially available high pumping speed of cryopumps varies: for N2 35,000 to 65,000 L/s; for hydrogen 15,000 to 35,000 L/s; for H2O 125,000 to 180,000 L/s and also various pumping speeds for argon, helium and other gases. Again, the capacity of the cryopump depends upon the manufacturer’s design.
The same cryogenic technology is needed for the most economical design to handle heat load capacity in watts for superconducting magnets used in medical equipment such as magnetic resonance imaging, cryobiology, or for cryogenically cooled materials. Quality control and safety are most important for achieving reliable cryopump performance. Selection of the cryopump based on heat load capability and crossover capacity requires technical evaluation of cryopump suppliers like PHPK, SHI, Dynavac, Oxford, CTI, Austin Cryogenics, Janis Research, Leybold and other leading suppliers.
Cryopump requirements to eliminate vibration in space telescopes: Space telescope operation requires a vibration- and noise-free environment and works at a certain frequency only during its certification and qualification of optical and IR measurement testing to achieve clear images during actual performance in space. This qualification measurement requires a long time—around five to eight hours or more during THVC testing. This makes it necessary to stop all cryopumps, because the cryopump cold head motor may work on certain frequencies and create noise/vibration that interferes with the optical measurement reading needed by the engineer to qualify the space telescope design. During this time, all thermal and vacuum systems will be stopped to provide a vibration-free orbital environment. We need to provide a clean high vacuum, better than 1 x 10-5 torr, at the required cryogenic temperature. The cryopump is the best and most economical approach to satisfy this specific space telescope requirement inside the large THVC to carry out a Space Telescope Qualification Test at ground level.
Cryopump requirement to achieve contamination-free orbital environment: The space telescope requires a no-contamination environment. The gas load, which includes inert gases such as nitrogen, oxygen, argon, etc., as well as the non-condensable gases like helium, neon and hydrogen, needs to be removed during the high vacuum cycle. The inert and non-condensable gas load plus outgassing, virtual leaks, permeation and real leak will act as contamination for the space telescope and may condense at cryogenic temperature of optical payload of the telescope. This is not acceptable and requires the cryopump and its regeneration to remove all contamination from the chamber to simulate a contamination-free space orbital environment.
Molecular Contamination Control
During thermal vacuum testing, contamination control is most important for space telescope performance. Efforts must be made to select low outgassing materials (CVCM level of 0.1% and TML of 1.0% by weight) for all applications. For materials where data does not exist, it may be necessary to test outgassing characteristics in accordance with the ASTM E595: “Methods of Test, Total Mass Loss and Collected Volatile Condensable Materials from Outgassing in a Vacuum Environment.”
Clean Vacuum Chamber Certification is required before the space telescope is installed inside THVC for final testing, as per specific project requirements for space telescope or satellite optical payload to be tested. This includes the following:
The Thermoelectric Quartz Crystal Microbalance (TQCM) and Cryogenic Quartz Crystal Microbalance (CQCM) can be used with a control system that measures and records condensable mass deposited on a piezoelectric crystal. Extreme accuracy is required when comparing the exposed measurement crystal to the reference crystal located in the same TQCM head. A computer-controlled thermoelectric device provides the high degree of crystal temperature control required for accurate frequency measurement. Two TQCM sensing units may be used, 10 MHz or 15 MHz. The data may be sent to the control unit and the computer control station, then to a data acquisition facility. For TQCM the crystal temperature was operated between -50 and +100° C within 1 to 0.1° C accuracy. The chamber must be pumped down to a test pressure of 10-5 torr or less, at which point the TQCM is turned on and set for the appropriate operating temperature as per orbital and payload requirements. As the test article outgassed and materials condensed on the TQCM sensing crystal, the crystal frequency increased in direct proportion to the amount of deposition from the test article outgassing. The function and control of a CQCM is similar to that of a TQCM. The CQCM with mass sensitive piezoelectric crystals were operated between -268 and 127°C as or if required.
The Residual Gas Analyzer (RGA): The RGA was used to measure the partial pressures of ionized molecules over a mass range of 1 to 300 atomic mass units (AMU) where the best sensitivities were below 100 AMU. Using a combined RF and electrostatic field formed by two metal rods, the RGA scanned the mass range and detect the partial pressures of each element or compound fraction. The RGA probe was located in the thermal vacuum facility and was oriented to maximize the detection of the outgassing species and also provide PPM of each gas.
The success of space telescope testing depends on the design and operation criteria discussed here for a clean high vacuum system for optical IR measurements. It requires three different engineering communities—cryogenic, vacuum technologist and optical IR engineer—to work out different approaches, considering the requirements of payload to be tested and reduction of total test cost. The proposed cryopump is an economical solution to the need for very high vacuum and an ultra-clean working environment.