Test Dome Research Articles

New technique for oil backstreaming contamination measurements

Due to the large size and the number of diffusion pumps, space simulation chambers cannot be easily calibrated by the usual test dome method for measuring backstreaming from oil diffusion pumps. In addition, location dependent contamination may be an important parameter of the test. We have measured the backstreaming contamination in the Space Power Facility (SPF) near Sandusky, Ohio, the largest space simulation vacuum test chamber in the U.S.A. We used small size clean silicon wafers as contamination sensors placed at all desired measurement sites. The facility used diffusion pumps with DC 705 oil. The thickness of the contamination oil film was measured using ellipsometry. Since the oil did not wet uniformly the silicon substrate, two analysis models were developed to measure the oil film: (1) continuous, homogeneous film and (2) islands of oil with the islands varying in coverage fraction and height. In both cases, the contamination film refractive index was assumed to be that of DC 705. The second model improved the ellipsometric analysis quality parameter by up to two orders of magnitude, especially for the low coverage cases. Comparison of the two models for our case shows that the continuous film model overestimates the oil volume by <50%. Absolute numbers for backstreaming are in good agreement with published results for diffusion pumps. Good agreement was also found between the ellipsometric results and measurements done by x-ray photoelectron spectroscopy (XPS) and by scanning electron microscopy (SEM) on samples exposed to the same vacuum runs.

Production of extreme high vacuum using a new bakeable‐type cryopump with Gifford–McMahon refrigerators

When we evacuate a vacuum system with a usual cryopump, temperature of vacuum baking is limited to the value lower than that efficient to reduce the ultimate pressure sufficiently. Especially, a cryopump which can be baked at high temperature (∼200 °C) during regeneration has a potential for the production of extreme high vacuum. We have developed a new type of cryopump with Gifford–McMahon refrigerators which can be baked at such high temperature during regeneration. In order to keep the temperature of refrigerator heads below the maximum sustainable temperature 70 °C, the cryopanels are separated from refrigerators and are cooled with helium gas circulating from refrigerators. Using the new cryopump, an aluminum alloy test dome coated with TiN was pumped down to the pressure less than 1×10−10 Pa in a few days after 150 °C vacuum bakeout. The low pressure of extreme high vacuum could be kept at least for 25 d. Major constituents of the final residual gas were m/e=2 (H2 ) and m/e=28 (N2 and CO).

7157. Physical vapor deposition equipment evaluation and characterization using statistical methods: VS Dharmadhikar et al, J Vac Sci Technol, A8, 1990, 1603–1607

The authors investigated the possibility of achieving an Extreme High Vacuum (XHV) environment by employing a commercial G-M refrigerator-cooled 10 K cryopump instead of a costly 4 K refrigerator-cooled cryopump. In order to minimize the outgassing rate which might disturb the evaluation of the cryosorption effectiveness in the XHV region, all vacuum walls of the cryopump vessel and the test dome were made of 5052 aluminum alloy the surfaces of which had been specially treated. After 70 days system conditioning, the ultimate pressure (between 3 and 5 × 10−11 Pa) was obtained repeatedly by 3 or 4 days evacuation. The partial pressure behavior of several gases was observed by changing the cryopanel temperature to evaluate the ultimate pressure limit of cryosorption pumping at the relatively higher temperature of 10 K. The experimental result shows that the cryosorption pumping at 10 K is still effective below 10−11 Pa.

Ｇ‐Ｍ冷凍機を用いたクライオポンプによるＸＨＶ実現の可能性

Abstract The authors investigated the possibility of achieving an Extreme High Vacuum (XHV) environment by employing a commercial G-M refrigerator-cooled 10 K cryopump instead of a costly 4 K refrigerator-cooled cryopump. In order to minimize the outgassing rate which might disturb the evaluation of the cryosorption effectiveness in the XHV region, all vacuum walls of the cryopump vessel and the test dome were made of 5052 aluminum alloy the surfaces of which had been specially treated. After 70 days system conditioning, the ultimate pressure (between 3 and 5 × 10 −11 Pa) was obtained repeatedly by 3 or 4 days evacuation. The partial pressure behavior of several gases was observed by changing the cryopanel temperature to evaluate the ultimate pressure limit of cryosorption pumping at the relatively higher temperature of 10 K. The experimental result shows that the cryosorption pumping at 10 K is still effective below 10 −11 Pa.

Capability of obtaining extreme high vacuum by commercial G-M refrigerator-cooled cryopump

The authors investigated the possibility of achieving an Extreme High Vacuum (XHV) environment by employing a commercial G-M refrigerator-cooled 10 K cryopump instead of a costly 4 K refrigerator-cooled cryopump. In order to minimize the outgassing rate which might disturb the evaluation of the cryosorption effectiveness in the XHV region, all vacuum walls of the cryopump vessel and the test dome were made of 5052 aluminum alloy the surfaces of which had been specially treated. After 70 days system conditioning, the ultimate pressure (between 3 and 5 × 10 −11 Pa) was obtained repeatedly by 3 or 4 days evacuation. The partial pressure behavior of several gases was observed by changing the cryopanel temperature to evaluate the ultimate pressure limit of cryosorption pumping at the relatively higher temperature of 10 K. The experimental result shows that the cryosorption pumping at 10 K is still effective below 10 −11 Pa.

Development of a bakeable cryopump for extreme high vacuum

For the production of extreme high vacuum with a cryopump we have developed a new cryopump/refrigerator system in which the refrigerator is separated from the cryopanel housing to keep its cold head below 70°C during 200°C bakeout. The cryopanels are cooled by circulating helium gas from a two-stage refrigerator. For a stainless steel test dome polished by electro-chemical buffering technique, the lowest total pressures attained were 5.3 × 10 −8 Pa before bakeout, and less than 6 × 10 −9 Pa after 30 h bakeout at 130°C.

Recommended procedures for measuring the performance of positive-displacement mechanical vacuum pumps

It is essential that uniform and realistic test procedures be used for measuring the performance of vacuum apparatus. This paper recommends procedures for measuring the pumping speeds and ultimate pressures of positive-displacement mechanical vacuum pumps. It replaces outdated material originally published in three separate sections between 1963 and 1967. Changes include emphasis on more modern pressure gauges such as the capacitance manometer and spinning rotor gauge, an attempt to harmonize procedures with existing international standards, a shift to Système International units of pressure, and minor rewording to include dry vacuum pumps. Minor changes have been made in test dome geometry to reduce sources of error. Material has been added concerning some of the causes of pumping speed variations during continuous operation of a given pump. Other new material covers the problems of measuring water vapor tolerance and water vapor pumping capacity. Practical details which minimize measuring errors are discussed.

Measurement of the effective pressure distribution in axial direction in a dynamic vacuum system

In this paper, a theoretical analysis of pressure distribution along the axial direction in a dynamic vacuum system with an oil diffusion pump is presented. Theoretical relationships between the measured pressure and different orientations of the gauge openings were derived by taking into account the gas beaming effect inside a test dome that is connected to an oil diffusion pump of the same diameter. A precise experimental study to verify the theoretical relationships between the measured pressures and the three orientations of the tubulated gauge was carried out using a 104‐mm‐diam test dome connected on top of an oil diffusion pump of the same diameter. Experimental results obtained on the pressure distribution in the axial direction at four tubulated gauge positions are presented and discussed.

コンダクタンス変調法による超高真空での真空ポンプの排気速度測定

The conductance modulation method has three advantages useful for the measurement of pumping speed in ultrahigh vacuum region. First, the flow rate of gas for the speed measurement is not required if the rate is constant. Secondly, the absolute value of pressure in the test dome is also not necessary if the sensitivity of pressure gauge is unchanged during the measurement. Thirdly, the effects of the positions of pressure gauge and gas inlet in the test dome are relatively small.The orifice system composed of three discs is used as the variable conductance which has two definit values of conductance determined by the Monte Carlo calculation. One of the two values can be selected from outside of the ultrahigh vacuum chamber. The pumping speed of a combination pump made of a sputter ion pump and a titanium getter pump can be successfuly measured down to 10-7 Pa for nitrogen and hydrogen.

Throughout determination in pumping speed measurements: comparison of two methods

The rate of pressure rise and the inverted pipette methods of throughout determination in pumping speed measurements are compared. The values obtained for the pumping speed of a NW63 oil diffusion pump using an ISO 63 test dome are the same for both methods showing that the rate of pressure rise method is valid for pumping speed measurements with small ISO test chambers.

High pressure pumping speed measurement on sputter ion pumps

Sputter ion pumps are nowadays available for operating at pressures up to 10−3 Torr. Problems arise when measuring pumping speed in the high pressure range (namely, 10−5–10−3 Torr) where the usual Fisher–Mommsen test dome is inapplicable. As a matter of fact assumptions on which it is based are not verified at such high pressures. In the present work a method is described for measuring pumping speeds in the high pressure range (10−6–10−3 Torr). Examples of application of this method to different pumps are given. Pumping speed plots are compared with the ones obtained with the Fisher–Mommsen test dome. In the 10−6–10−4 Torr range, where these two methods overlap, pumping speeds obtained by both methods are in fairly good agreement. By combining the two methods it is possible to characterize pumping speeds in the whole operating range.

Gas beaming effect: measurement of pumping speed and determination of effective conductance

This paper reports the experimental results of the effect of gas beaming on the pumping speed measurement of an oil diffusion pump at the four different positions of the test dome for a tubulated gauge connection and also for the three different orientations of the gauge inlet. The results obtained are compared with the theoretically calculated values. The mathematical relationships for determining the gauge positions Y o (to measure the intrinsic speed) for the remaining two orientations A and C of both, a tubulated as well as an orifice gauge inlet, are derived and the values of Y o/ R thus determined are also presented. The determination of correct value of effective conductance is made and the theoretical values of the dome constant for different sizes of the test dome for both the tubulated and an orifice type gauges are determined. In the light of experimental results obtained, the relationship between the intrinsic speed, measured speed and the effective conductance due to the gas beaming are discussed.

Cryopump speed measurement with the AVS/ASTM and ISO test domes

Correction factors are presented for speed measurement of cryopumps using nude ionization gauges, which take into account the effects produced by the reflection of gas molecules from the cold surface of the cryopump

Cryopump regeneration studies

A cryopump (APCI Model HV202-8C) connected to a pump speed test dome was used to evaluate the time and effectiveness of its regeneration using alternate methods of purging and heating the pump. Regeneration effectiveness was determined by measuring the pumping speed and helium adsorption capacity of the charcoal after cool down. The pump was loaded with 8 g of water and 10 std.L of argon prior to regeneration. For all tested regeneration methods the helium speeds were within 80% of the maximum speed. Warm up and evacuation without extra heat or a purge followed by cooling can be done in 4 h and 15 min. Regeneration was accomplished in 2 h and 45 min using a heater on the outside of the housing and a nitrogen purge to bring the cryopump to 310 K, followed by evacuation with a mechanical pump to 100 mTorr without heat or purge, then closing the roughing valve and starting the cryopump. An initial water vapor pressure of 100 mTorr does not increase the cool-down time compared to a lower starting pressure. Tests with other gases showed that argon and nitrogen have to be pumped below 20 mTorr and hydrogen and helium have to be pumped below 4 mTorr to avoid long delays in cool down.

Study of molecular flow inside a test dome connected to the same diameter diffusion pump

The paper presents a study of molecular flow inside a test dome connected to the same diameter diffusion pump. The theoretical analysis of the problem connected with the tubulated as well as the orifice type pressure gauges has been briefly reviewed by introducing the new factors J t and J orf for the tubulated and the orifice gauges respectively. The values of these factors for different gauge positions inside the test domes of different sizes (L = R to L = 8R) are determined. The mathematical relationships for determining the measured pressure for different orientations of the gauge inlet at the different positions are also derived. The improved values of the gauge position ‘Y o/R’ to measure the intrinsic speed) for a tubulated gauge are determined by using the correct value of wall flux distribution. To verify the theoretical analysis for tubulated gauge, an experimental study using a 104 mm diameter pump connected to the same diameter dome of effective length 6 R is made by measuring the pressures at the four different positions. The analysis of the experimental results obtained shows a good agreement with the theory. A new approach to intrinsic speed measurement, by measuring the pressure at the different positions of the test dome, has been presented in which the measurement of throughput is not required. By measuring the pressures in different directions of the inlet of the gauge, the number of molecules s −1 crossing the dome at any position is determined and is compared with the number of molecules s −1 actually introduced inside the test dome by the gas throughput.

Comparison of two pumping speed measuring methods of oil diffusion pumps

This paper presents a comparative study of two pumping speed measurement methods recommended for oil diffusion pumps. The methods used are the test dome method and the conductance tube method with an auxiliary pumping system which is used at very low pressures when the throughput is below 10 −5 Nm s −1 (∼ 0.1 μl s −1). It is found that, in order to obtain the comparative results in the overlapping pressure region on the speed measurement of oil diffusion pumps by these two methods, the conductance tube in the conductance tube method should be located at a height 1.2 D from the pump mouth when the position of the inlet gauge is at a distance D/2 from the pump flange, i.e. at the same position as recommended in the dome method to obtain the intrinsic speed of the pump.

The appropriate test domes for pumping speed measurement

The appropriate test domes for pumping speed measurement

The appropriate test domes for pumping speed measurement

The current test domes for pumping speed measurement with two gauges and a calibrated orifice have been reviewed. The key to investigation of test domes is to choose a suitable assumption characterizing the inlet aperture of the pumps. In order to set up an appropriate test dome instead of an infinite one, first we should determine the height of the lower chamber of the dome, then determine the gauge position for pressure measurement. A set of equations that show the relation between the height, the position and the radius of the calibrated orifice have been derived, and solved by the Monte Carlo method. The heights and positions of appropriate test domes have been obtained for three orifices of different radii. The check shows that in appropriate test domes the same pressure appears in three main directions, and therefore simulates infinite test domes more perfectly than ones as given in the tentative standard ISO/DIS 3556/1 and others.

A Method For Testing The Effects Of Fluids On Rat Skin Using Acrylic Chambers

A method is presented whereby test fluids may be kept in contact with the skin of the back of the rat for 5 to 6 days. Methyl methacrylate chambers with hemispherical test domes were made and attached to rubber washers of larger diameter, which in turn were attached to the skin of the back of the rat using an alkyl cyanoacrylate (Eastman 910 adhesive). Each chamber had a removable stainless-steel screw to allow it to be filled and replenished with fluid as required. To prevent the rat gnawing the chambers from its back, twice weekly 4 mm was removed from the lower incisors and 1 1/2 mm from the upper incisors, using a diamond disc in a dental handpiece. The young adult rats were fed a soft diet and after an initial temporary fall in weight, they moved about the cages in a normal manner.

Monte Carlo design of a 3−gauge test dome

The preliminary results of the Monte Carlo analysis of the 3−gauge (or conductance tube) test dome reported at the 6th International Vacuum Congress1 have been extended to the noncosine distribution for molecules reflected back into the test dome from a pump being tested. The gauge locations for the measurement of ’’intrinsic pumping speed’’ are presented showing the influence of reflection coefficient on optimum gauge location. The distribution from one actual pump configuration has been obtained from the Monte Carlo calculations and applied to the test dome design. The pumping speed obtained from this design for a 3−gauge test dome is compared with the results obtained with a Fischer−Mommsen test dome.