Ionization gauge and method of using and calibrating same

Abstract

Controller circuitry and method for controlling the operation of an ionization gauge having a source of electrons, an anode, and an ion collector electrode, including storing a plurality of calibration data sets for at least collector electrode currents and gauge sensitivities obtained with at least one reference gauge at representative values of known pressures; and calculating the gauge sensitivity corresponding to an unknown pressure in response to at least one of the data sets together with a measured value of the ion collector current. The data sets may also include representative values of the heating powers of the electron source so that the calculated sensitivity may also be compensated for variations in the electron source heating power. The ionization gauge may be of the Bayard-Alpert type and include a shield which defines a shielded volume within which an electron source, anode, and collector electrode are disposed where the shield is at least partially open to permit transfer of gas molecules into the shielded volume and where the shield completely encloses the electron source, the anode and the collector electrode so that potentials external to the shield means do not disturb the electric charge distribution within the shielded volume to thus stabilize the sensitivity of the gauge.

Description

BACKGROUND OF INVENTION

1. Field of the Invention

The present invention relates to vacuum gauges such as Bayard-Alpert (BA) ionization gauges and to systems and methods for operating and calibrating such gauges.

2. Discussion of the Prior Art

The BA gauge is the simplest known non-magnetic means of measuring very low pressures and has been widely used worldwide essentially unchanged since being disclosed in U.S. Pat. No. 2,605,431 in 1952.

A BA gauge system consists of a BA gauge and controller circuitry. A prior art BA gauge consists of a heated cathode for emitting electrons, a cylindrical grid or anode for accelerating the emitted electrons to ionizing energy, a very small cross section collector electrode on axis for collecting the ions formed within the anode volume by collision of energetic electrons with gas molecules and a vacuum envelope surrounding the gauge electrodes and attaching to a vacuum system wherein an unknown gas pressure is to be measured. Controller circuitry consists of a means to apply suitable potentials to the anode, to the cathode and to the collector electrode, means for heating the cathode to provide a controlled electron emission current and means for measuring the collector current and means for calculating and displaying the indicated pressure.

As a first order approximation, it can be stated that the ion current i.sub.+, to the collector electrode is proportional to the electron emission current i.sub.--, and to the gas density in the gauge or at constant temperature to the gas pressure, P.sub.G, in the gauge. Thus,

i.sub.+ =Si.sub.-- P.sub.G Eq. 1 where S is a constant of proportionality commonly called the gauge sensitivity.

S can be calculated by measuring i.sub.+ and i.sub.-- when the gas pressure P.sub.GK in the gauge can be determined by other calibration means. Thus,

S=i.sub.+ /(i.sub.-- P.sub.GK) Eq. 2 where P.sub.GK is known pressure in the gauge. The value of S thus determined can be utilized to calculate the value of an unknown gauge pressure P.sub.x which produces an ion current, i.sub.+x at the same value of i.sub.-- used to determine S in Eq. 2 provided conditions in the gauge do not change. Thus, P.sub.x =i.sub.+x /(i.sub.-- S) Eq. 3

Very complex and costly non BA gauges have been described which are claimed to provide excellent accuracy of measurement over a limited pressure range. However, these laboratory devices are entirely unsuitable for everyday use in research and industry and have yielded few clues if any as to how to improve the accuracy of the simple prior art BA gauge.

Numerous studies over the years have demonstrated conclusively that S is not constant in prior art BA gauges. For example, Poulter and Sutton at the National Physical Laboratory in the United Kingdom reported that a typical prior art BA gauge calibration "drifted at a rate of -1.4% per 100 operating hours when it was kept under vacuum but similar gauges exposed to atmospheric conditions showed sharp changes of up to 25% in their sensitivity", K. F. Poulter & C. M. Sutton, Vacuum, 31, 147-150 (1981). In another representative study, Tilford at the U.S. National Bureau of Standards reported that typical prior art nude BA gauges (BA gauges without vacuum envelopes) had sensitivities which varied from 70% to 110% of that specified by the manufacturer, C.R. Tilford, J. Vac. Sci. Technol., A1(2), 152-162 (1983). In another study, repeated calibrations of seven prior art "broad range" BA gauges showed sensitivities varying from 52% to 67% of their specified values, C. R. Tilford, K. E. McCullogh, H. S. Woong, J. Vac. Sci. Technol., 20(4), 1140-1143 (1982).

Ionization gauge systems must be calibrated against a primary or secondary pressure standard to be useful. Users of prior art BA gauge systems have been provided with three alternatives to secure required accuracy of vacuum measurement.

1. Calibrate the ionization gauge system in situ utilizing, for example, an expensive spinning rotor gauge. Such a solution is being actively promoted now by several spinning rotor gauge suppliers. The prior art gauge may change calibration at anytime, so frequent in situ calibration is required. This alternative provides the best accuracy bu is extremely time consuming, inconvenient and costly. 2. Send the ionization gauge system to a separate calibration facility and have the system calibrated--an expensive, time consuming alternative at best because the prior art gauge may change calibration anytime with use. This alternative enables the user to comply with certain governmental requirements but is unlikely to result in improved accuracy over the long term. 3. Use the generic calibration data supplied by the manufacturer assuming it is accurate for the prior art BA gauge in use. This generic calibration data is typically obtained by measurements on one or more prototypes and consists of a fixed value of the sensitivity S for a given gas type and a list of the nominal applied electrode potentials.

The huge majority of users opt for the last alternative because of the large expense and inconvenience of either of the other alternatives even though using generic calibration data provides vacuum measurements grossly in error.

What is needed is better means for providing generic calibration data that will yield more accurate vacuum measurements.

Many researchers have pointed out for years that S is not a constant but depends on gas species, electron energies, emission current, electric field distributions, gas density, kinetic energy of the gas molecules in the gauge, plus several other parameters. But from all this work there have been no solid clues on how to improve the accuracy of the prior art BA gauge and the inaccuracy remains the same as when the device was invented. From the vantage point achieved through computer simulation of electron and ion trajectories in a BA gauge, it is now possible to discern deficiencies in the prior art which help cause the observed large inaccuracies in prior art BA gauge systems.

It is well known that the performance of BA ionization gauges can be seriously affected by a phenomenon known as surface ion desorption. An effective way of minimizing this effect is by utilizing an anode which has minimum surface area. Thus, to minimize surface ionization inaccuracies all low pressure BA gauge designs utilize transparent grids with minimal surface area. When such open grids are used, energetic electrons are not confined to the anode volume but travel a significant fraction of their total path length outside the grid. These energetic electrons traveling outside the anode volume can impinge on exposed insulating surfaces and uncontrollably change the surface potential of the exposed insulator. Their trajectories may also be changed by uncontrolled potentials outside the anode volume. Utilizing computer simulation, one can determine if energetic charged particles will impact an exposed surface or if trajectories are influenced by potentials outside the anode volume for any given configuration of surfaces, potentials, energy and initial trajectory of the charged particles.

The total electric charge distribution in a BA gauge is the sum of the charge distribution on the surfaces exposed to impact by charged particles and the charge distribution due to free charges within the gauge volume. In prior art BA gauges, means have not been provided for adequately fixing the electric charge distribution in that region of the gauge accessible to energetic charged particles. Thus, the electric charge distribution can vary from measurement-to-measurement in the same gauge or from gauge-to-gauge at any given pressure. Thus, because the electric charge distribution which exists during calibration cannot be duplicated during use, inaccurate pressure indications result.

Applicants have found by computer simulation that seemingly trivial changes in the electrode geometry or in surface potentials in the gauge volume accessible to energetic charged particles can cause large changes in electron trajectories. It is well known that changes in electron trajectories cause changes in the radial position, r.sub.o, of ion formation resulting in changes in the angular momentum, m r.sub.o v.sub.T, of the ion, where m is the mass of the ion and v.sub.T is the tangential component of its velocity. Changes in angular momentum cause the probability of ion collection to change as is well known, thus causing the gauge sensitivity to change. Changes in gauge sensitivity cause measurement inaccuracies.

Prior art BA gauges can be categorized as to the manner in which and the degree to which the electric charge distribution on surfaces exposed to charged particle impact in the gauge are uncontrolled.

1. Glass envelope BA gauges

The huge majority of BA gauges in use have glass vacuum envelopes where the entire inner surface of the gauge is exposed to energetic charged particles. Energetic electrons and ions can impact on these insulator surfaces and charge them to uncontrolled potentials. Thus, in glass gauges the electric charge distribution on the insulator surfaces is totally uncontrolled. Abrupt changes in indicated pressure of 50% or more are regularly observed at 10.sup.-5 Torr or higher particularly at lower emission currents.

2. Glass envelope BA gauges with conductive coatings

Conductive coatings such as stannous oxide and platinum have been deposited on a portion of the inner glass wall of BA gauges and held at a fixed potential, as disclosed, for example, in U.S. Pat. No. 3,839,655. However, in all such prior art devices the glass insulators and substantial areas of the inner glass surface have remained uncoated and exposed to energetic charged particles. Thus, the exposed glass insulator surfaces can charge up uncontrollably and change the electric charge distribution in such prior art devices. Applicants have found that changes in port diameter or even in the length of the port tubulation affect the charge distribution in prior art devices.

3. Nottingham gauge

Nottingham at M.I.T. positioned a screen grid between the glass wall and the BA gauge electrodes but did not enclose the ends of the screen grid, W. B. Nottingham, AVS Vac. Symp. Trans. 8 494 (1961). The anode ends were enclosed with grids but the screen grid ends were left open. Thus, the exposed glass surfaces on both ends of the gauge can charge up uncontrollably and change the electric charge distribution in the gauge.

4. Nude gauge with exposed feedthru insulators

So-called nude gauges have the electrode assembly mounted on a metal flange which is penetrated by multiple feedthru insulators. In this regard, see, for example, Varian Associates Series UHV 24 Specification Sheet. The collector electrode feedthru is invariably covered with a metal shield to prevent soft x-rays from impinging on the collector electrode support thus decreasing the indicated base pressure of the gauge. The remaining feedthru insulators are exposed to energetic charged particles and can charge up uncontrollably and change the electric charge distribution.

5. Nude gauge used with undefined enclosure geometry

Nude gauge electrode assemblies are intended to be inserted into a metal vacuum system. The interior surfaces of the vacuum system surrounding the electrode structure help to define the electric field configuration in regions traversed by electrons. A nude gauge is typically calibrated in one vacuum system geometry and used in a different geometry or even exposed to different potentials. The electric charge distribution in such cases is different during use from that present during calibration. Thus, the indicated pressure cannot be accurate.

The American Vacuum Society Standard 6.4, "Procedure For Calibrating Hot Filament Ionization Gauges Against A Reference Manometer In The Range 10.sup.-2 -10.sup.-5 Torr" contains no directives, cautions, nor any hints that the electric charge distribution must be controlled to achieve accurate pressure indication.

In AVS Standard 6.4 there is a note: "The position of the tube relative to the chamber wall and the temperature of the chamber wall exposed to the tube are factors in the gauge calibration and, hence, should duplicate actual usage as nearly as possible". This note is made in the context of ensuring that all gauges are symmetrically placed so that gas temperatures and conductances are the same for the test gauges and reference manometers.

U.S. Pat. No. 3,742,343 discloses a Groszkowski gauge wherein a cylindrical electrical conducting screen is described for use in a nude ion gauge to ensure repeatability of pressure measurement in various configurations of the vacuum system. However, the screen described does not completely enclose the end of the screened region. Thus, variations in the electric charge distribution on unshielded exposed surfaces can affect the sensitivity.

In 1986 Hseuh and Lanni at Brookhaven National Laboratory tested 400 custom nude BA gauges in different specified diameter metal envelopes and observed considerable variation in sensitivity S, see H. C. Hseuh & C. Lanni, J. Vac. Sci. Technol. A5 (3) Sep/Oct 1987, p.3244. Neither screens on the port end nor insulator shields were present during these tests. Thus, the electrical charge distribution on exposed surfaces could vary during testing, thus affecting the sensitivity.

In a recent research paper, Peacock and Peacock reported on tests on nude gauges with and without closed ends on the anodes, see R. N. Peacock and N. T. Peacock, j. Vac. Sci. Technol., A Vol. 8, No. 4, Jul/Aug 1990. The authors state, "To obtain reproducible conditions all nude ionization gauges were operated in nipples 97 mm long by 33 mm inside diameter. These were provided with a screen at the chamber end of the nipple to prevent ion coupling between gauges. The tubulated glass gauge had a similar screen on its flange." However, no insulator shield is provided on the tested gauges.

In addition to the foregoing considerations, to the best of applicants' knowledge, no prior art nude gauge manufacturer nor distributor has provided the pertinent dimensions of the controlled environment used to provide a generic calibration of the gauge. Thus, the user has no means to duplicate the generic calibration geometry during use and the pressure indications are therefore likely inaccurate.

6. BA gauges with large tolerance electrode geometry

A cursory examination of any ten commercially available prior art BA gauges with the same part number will reveal significant dimensional differences in electrode positions. Variations of as much as 0.25" are evident in some products. Because the electric charge distribution is not at all the same in such gauges as in the prototype gauge which was actually calibrated by the manufacturer, large inaccuracies in indicated pressure can be observed when the generic calibration data is used.

7. BA gauges with electrode geometries that change with time.

Designers of prior art BA gauges have failed to take the necessary precautions to maintain all electrode positions constant with use. Most glass BA gauges utilize untensioned and unsupported hair pin shaped cathodes which typically sag and move by tenths of inches in long term use. Bifilar wire anodes sag and warp in long term use. Ion collectors that are not adequately stress relieved can change shape over time especially when subjected to high temperatures during electron bombardment degassing. In some nude gauges the collector electrode position changes with externally applied force on the feedthru pin. Bimetallic joints cause electrode motion as the power input into the gauge varies. In some prior art BA gauges the collector electrode is so long and flimsy that in use it vibrates with a relatively large amplitude that changes with external conditions. Because the electric charge distribution changes uncontrollably with time in such gauges, the gauge becomes inaccurate.

8. BA gauges with electrodes fixed relative to one another but with large tolerance between electrodes and enclosure.

In many prior art gauges it is common to observe relatively large variations in the location of the entire electrode structure relative to the vacuum envelope. Such variation, of course, affects the electric charge distribution in the volume accessible to energetic electrons. Changes in the electric charge distribution cause inaccuracies in the indicated pressure gauge to gauge.

Thus, the situation with prior art BA gauge systems is as follows: The electric charge distribution on surfaces in prior art gauges is so ill-controlled that there is little point in providing generic calibration data for a given gas type other than (1) a constant value for the sensitivity for use at all pressures, (2) nominal values of potentials applied to the cathode and anode relative to ground, and (3) a nominal value of the emission current used during calibration.

Thus, even though it has long been known that the BA gauge calibration depends on a number of variables, it has not been possible to correct the indicated pressure in a prior art BA gauge system for changes in any or all of the variables.

What is needed is a better approach to providing a more accurate BA gauge system.

SUMMARY OF THE INVENTION

A primary object of the present invention is to provide significantly greater accuracy of vacuum measurement than has heretofore been possible with prior art ionization gauges such as BA gauges.

Applicants have discovered that the ion collection probability in a BA gauge and therefore the BA gauge sensitivity is far more influenced by minor variations in the electric charge distribution than has been previously assumed. Seemingly trivial changes in electrode geometry or in surface potentials can cause large changes in electron trajectories. As is well-known in the art, changes in electron trajectories cause changes in the radial position, r.sub.o, of ion formation resulting in changes in the angular momentum, m r.sub.o v.sub.T, of the ion, where m is the mass of the ion and v.sub.T is the tangential component of its velocity. Changes in angular momentum cause the probability of ion collection to change thus causing the gauge sensitivity to change. Changes in gauge sensitivity cause measurement inaccuracies whenever the sensitivity has been assumed to be constant.

It is thus a further primary object of the present invention to provide means to maintain electric charge distributions on surfaces exposed to charged particle impact, constant gauge-to-gauge and over time so that the gauge sensitivity, although not a constant over the pressure range, is more repeatable measurement-to-measurement in the same gauge and more reproducible gauge-to-gauge when preferably used with a well-defined set of calibration measurement parameters. Thus, greater throughput may be obtained in various industrial processes since it becomes unnecessary to operate at needlessly low pressures where time may be wasted until the system can be pumped to achieve low indicated pressures.

The above calibration parameters are those which affect the gauge sensitivity, namely gauge type, electrode voltages, gas type, emission current, cathode heating power and pressure. This more stable sensitivity can then be corrected utilizing algorithms and stored generic values of the collector current, cathode heating power and gauge sensitivities obtained during a prior calibration run at representative values of each parameter. Up to now there has been no hope of correcting the sensitivity of prior art gauges because the electric charge distribution on gauge surfaces has been so unstable.

It is a further object of the present invention to provide hardware and software for storing generic values for gauge type, electrode voltages, gas type, emission current, collector current, cathode heating power and sensitivity. Hardware and software are provided for measuring the collector electrode current at an unknown pressure, P.sub.x, selecting the correct sets of stored generic values, calculating the correct values of the sensitivity and indicated pressure and displaying the correct value of the indicated pressure. It is a further object to measure, in addition to the collector electrode current, the cathode heating power and to also utilize the measured cathode heating power when calculating the correct values of sensitivity and indicated pressure.

It is a further object to provide a method for utilizing the above hardware and software to produce a more accurate measurement than has previously been possible.

It is a further object of the present invention to provide a calibration method which compensates for changes in work function of the gauge cathode from gauge to gauge and for a particular gauge over a period of time.

These and other objects of the invention will become apparent from a reading of the following specification taken with the drawings and claims.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic cross-sectional view through the axis of a first illustrative embodiment of a gauge in accordance with the invention.

FIG. 2 is a cross-sectional view thru the axis of another illustrative embodiment of a gauge in accordance with the invention.

FIG. 3 is a schematic end-on view of the gauges of FIGS. 1 and 2.

FIG. 4 is a schematic block diagram of an illustrative gauge and controller in accordance with the invention.

FIG. 5 is an illustrative example of selected data shown in Table 1.

FIGS. 6 and 7 are illustrative examples of calculated data.

DESCRIPTION OF PREFERRED EMBODIMENTS

Reference should now be made to the drawing where like reference numerals refer to like parts.

Referring to FIGS. 1 and 3, a nude gauge assembly 10 of the BA type in accordance with the present invention comprises a gas permeable, metallic shield 12 (which may be cylindrical in configuration), an open, metallic, insulator shield 14, a cathode or electron source 16, an anode 18 (which may also be cylindrical in configuration), a collector electrode 20, insulated vacuum feedthroughs 22, a vacuum enclosure 24, and a flange 28 to attach the gauge to the vacuum system.

The shield 12 and the insulator shield 14 are preferably connected to each other and are preferably grounded and serve to define a shielded volume 26 wherein the electric charge distribution on the gauge surfaces is not disturbed by potentials external to shielded volume 26.

In general, shield 12 and insulator shield 14 may be considered a shield means or outer electrode which is at least partially open (mesh-like, for example) and which completely surrounds the cathode, anode and collector electrode disposed within shielded volume 26. The shield means is at least partially open to permit transfer of gas molecules in and out of the shielded volume. Moreover, the shield means (and insulator shield 14, in particular) is at least partially open at 14' to permit electrical connections to the cathode, anode, and collector electrode so that the latter electrodes are electrically isolated from the shield means.

Due to the grounding of the shield means, any electrons or ions generated within the shielded volume which contact the shield means will be passed to ground while any potentials (which may be quite large depending on the process) external to the shielded volume will have no effect on the electric charge distribution on gauge surfaces within the shielded volume. Thus, stabilization of the gauge sensitivity is achieved in this first important respect. Means and procedures for obtaining further stabilization are described below.

FIGS. 2 and 3 illustrate another embodiment of a non-nude (modest) gauge 32 wherein a metallic vacuum enclosure 24 serves the same function as longitudinally extending portion 12" of the shield 12 in FIG. 1. A metallic, gas permeable plasma shield 30 serves the same function as the end 12' of the shield 12 in FIG. 1. Thus, in FIG. 2, the shield means includes vacuum envelope 24, plasma shield 30, and insulator shield 14 where, as in the FIG. 1 embodiment, the shield means is preferably grounded.

If the vacuum enclosure 24 is made of glass, an electrically conductive coating may be provided on the inner surface thereof to provide the function of portion 12" of FIG. 1. Alternatively, a screen grid may be provided between a glass vacuum enclosure where the screen would provide the function of portion 12" of FIG. 1. Other means for completely enclosing shielded volume will occur to those of ordinary skill in this art.

In the embodiments of FIGS. 1 and 2, the dimensions of the shield means are preferably maintained substantially the same in both embodiments. Moreover, the locations of the cathode, anode, and collector electrode with respect to the shield means are substantially the same over time and from gauge to gauge so that there is no change in the electric charge distribution on surfaces within the shielded volume which would otherwise be caused by variations in the location of these elements with respect to the shield means.

Moreover, in FIGS. 1 and 2, all dimensions of hot cathode 16, anode 18 and collector electrode 20 are maintained fixed over time and gauge to gauge so there is no change in the electric charge distribution on gauge surfaces due to changes in electrode dimensions.

Furthermore, in FIGS. 1 and 2 the positions of cathode 16, anode 18 and collector electrode 20 are maintained fixed relative to one another over time and gauge to gauge so there is no change in the electric charge distribution due to electrode location.

The location of the gauge electrode surfaces can be fixed over time in the same gauge and from gauge-to-gauge by utilizing good mechanical design and construction techniques which are well-known in the art of electron tube design and construction. The cathode, as is known, can be prevented from sagging in use by utilizing a small spring to tension the cathode. Motion due to bimetallic joints can be avoided by proper joint design. Close tolerance electrode structures can readily be assembled using proper jigs and fixtures as is well-known in the art. Proper design of the feedthrough insulator as is well-known in the art insures that reasonable stresses applied to the feedthrough outside the gauge will not move the surfaces inside the gauge.

Referring to FIG. 4, illustrative controller circuitry in accordance with the present invention consists of a cathode heating power control 40 which maintains a constant emission current, i.sub.--, from the hot cathode, 16, in accordance with a pre-selected value stored in the parameter setting block 42. Also stored in block 42 is the gauge tube type, electrode voltages and gas type information which the user has preselected. A multiplexer 44 and an A/D converter, 46, digitizes the measured analog values of the ion collector current, i.sub.+, the cathode heating current i.sub.K, and the cathode heating voltage, V.sub.K.

As described below, the collector current i.sub.+ may be uniquely utilized in the calibration technique of the present invention to compensate for the variation in gauge sensitivity S with pressure. Digitized values of i.sub.+ are supplied to the computation block 48 where this data is combined with generic pressure calibration information stored in memory 50. The correct value of S is calculated. Then the correct value of unknown pressure P.sub.x, in the gauge is calculated in block 48 and fed to the pressure display 52, for display.

Moreover, as will also be described below, the cathode heating voltage and current (power) may be uniquely utilized in the calibration technique of the present invention to compensate for varying characteristics of the cathode such as the work function thereof. Digitized values of the cathode heating voltage and current and the collector current i.sub.+ are supplied to the computation block 48 where this data is combined with generic pressure calibration information stored in memory 50. The correct value of S is calculated. Then the correct value of the unknown pressure P.sub.x, in the gauge is calculated in block 48 and fed to the pressure display, 52, for display.

The above calibration methods and apparatus are preferably utilized with the gauge configurations of FIGS. 1-3 and equivalents thereof but may also be employed with other ionization gauges including BA gauges provided they are sufficiently stable. In this regard, it is noted that the foregoing calibration methods and apparatus can be utilized because of the high degree of stability of the sensitivity of the gauges of FIGS. 1-3 and equivalents thereof. That is, heretofore, due to the instability of the sensitivity of prior art gauges, it has not been practical to utilize the calibration methods and apparatus of the present invention.

A description will first be given of how the cathode heating power may be utilized to compensate for varying characteristics of the cathode such as the work function thereof. It will then be seen that the utilization of the collector current i.sub.+ to compensate for sensitivity variations may be treated as special case when it is not desired or necessary to correct for changes in the cathode heating power.

Generic pressure calibration data as shown in Table 1 is stored in memory. This data is preferably obtained by averaging measurements utilizing two or more (twenty, for example) BA ionization gauge systems in accordance with the present invention using known pressure calibration apparatus. In particular, data set A of Table 1 corresponds to low cathode heating powers needed to produce an electron emission current of 100 .mu.A, data set B to medium cathode heating powers required to produce this emission current, and data set C to high cathode heating powers. Thus, with respect to data set A, i.sub.1, W.sub.1A, and S.sub.1A are the average collector current i, cathode heating power W, and sensitivity S of twenty (for example) gauges of the present invention for a first known calibration pressure; i.sub.2, W.sub.2A, and S.sub.2A are the average collector current, cathode heating power, and sensitivity for a second known calibration pressure greater than the first calibration pressure; etc. The number of calibration pressures will extend over the desired pressure measurement range.

TABLE 1__________________________________________________________________________EXAMPLE OF CALIBRATION DATA SETS FOR A GIVENGAUGE TYPE AND FOR GIVEN ELECTRODE POTENTIALSData GasSet, Type, i.sub.-- i.sub.1, W.sub.1, S.sub.1 i.sub.2, W.sub.2, S.sub.2 i.sub.3, W.sub.3, S.sub.3__________________________________________________________________________A N.sub.2 100 .mu.A i.sub.1, W.sub.1A S.sub.1A i.sub.2, W.sub.2A, S.sub.2A i.sub.3, W.sub.3A, S.sub.3AB N.sub.2 100 .mu.A i.sub.1, W.sub.1B S.sub.1B i.sub.2, W.sub.2B, S.sub.2B i.sub.3, W.sub.3B, S.sub.3BC N.sub.2 100 .mu.A i.sub.1, W.sub.1C S.sub.1C i.sub.2, W.sub.2C, S.sub.2C i.sub.3, W.sub.3C, S.sub.3CD N.sub.2 1 mA i.sub.1, W.sub.1D S.sub.1D i.sub.2, W.sub.2D, S.sub.2D i.sub.3, W.sub.3D, S.sub.3D__________________________________________________________________________

In accordance with the invention, i, W, and S are obtained for the above series of calibration pressures for a plurality of different cathode heating powers required to obtain a predetermined emission current with slightly different degrees of contamination of the emitting surface of the cathode. Thus, data set B will be selected to have cathode heating powers greater than those of data set A while data set C will be selected to have cathode heating powers greater than those of data set B.

The cathode heating powers of data sets A, B, and C preferably extend over the range of expected cathode heating powers of gauges in actual use. Of course, the number of data sets may be changed to insure appropriate representation over the expected range of cathode heating powers. In use, a gauge may be temporarily exposed to contamination such as oxygen from the system being monitored whereby the work function of the cathode will be increased to a certain extent depending upon the degree of exposure. With further usage of the gauge, the degree of change of work function may increase or decrease. Thus, the amount of cathode heating power required to obtain a predetermined amount of emission current will vary from gauge to gauge and over time with respect to a particular gauge. In accordance with the present invention, this variation in the cathode heating power can be compensated by utilizing calibration data typified by data sets A, B, and C.

As discussed above, data sets A, B, and C respectively correspond to cathode heating powers required to obtain an emission current of 100 .mu.A. 100 .mu.A is a typical emission current utilized in measurements of high pressure ranges. In lower ranges, a larger emission current such as 1 ma is used and this is exemplified by data set D. Although not shown, there would also be in Table 1, data sets E and F which would respectively correspond to data sets B and C. Of course, whatever number of data sets are required for expected pressures, gas types, emission currents and cathode heating powers can be obtained and employed in Table 1.

For purposes of illustrating the method of using the calibration method of the present invention, assume an unknown gas pressure, P.sub.x, is to be measured in N.sub.2. Gas type N.sub.2 is entered into the parameter settings block 42 by the user using a selector switch, for example. Similarly the desired emission current setting, say, i.sub.-- =100 .mu.A, is entered by the user into block 42. Alternatively, the controller 38 may be arranged to select the optimum value of emission current depending on the current value of the pressure in the gauge as is well-known in the art. This automatically selected value of i.sub.-- is then fed to the parameter setting block 42. Parameter setting block 42 is also set for gauge type and electrode potentials applied.

Measurements of the collector current, i.sub.+x, and the cathode heating current, i.sub.kx, and voltage, V.sub.kx, are then obtained at an unknown gas pressure, P.sub.x, in the gauge as is well known in the art.

Assume the cathode heating power W.sub.x =i.sub.kx V.sub.kx to produce the required emission current is calculated in the computation block 48 as is well known in the art. Further assume W.sub.1A .ltoreq.W.sub.x .ltoreq.W.sub.2B. It remains to calculate the correct value of the sensitivity so that P.sub.x can be calculated and displayed.

The controller circuitry has been previously programmed as is well known in the art to select the two data sets stored in memory 50 which most closely approximate the actual conditions existing during measurement of the unknown pressure P.sub.x. Thus, for the assumed conditions for this example of:

______________________________________1. Gauge type = same as used for cali- bration data stored in memory2. Electrode potentials = same as used for cali- bration data stored in memory3. Gas type = N.sub.24. i.sub.-- = 100 .mu.A5. W.sub.1A .ltoreq. W.sub.x .ltoreq. W.sub.2B'______________________________________ data sets A and B are selected out of all the data sets stored in memory because these data sets most closely approximate the conditions existing while the unknown pressure, P.sub.x, is being measured. The unique feature of this aspect of the present invention is the selection of the data sets on the basis of the calculated cathode heating powers to thereby compensate variations in the work function, for example, of the cathode from gauge to gauge or with respect to time for a particular gauge. This utilization of the calibration data is preferably employed with the gauge configurations of FIGS. 1-3 and equivalents thereof but may also be employed with other ionization gauges including BA gauges provided they are sufficiently stable.

Note data sets A and B are selected since W.sub.x occurs between (a) W.sub.1A where W.sub.1A is the cathode heating power associated with the first calibration pressure and the low cathode heating power utilized during the calibration procedure and (b) W.sub.2B where W.sub.2B is the cathode heating power associated with the second calibration pressure and the medium cathode heating power.

Referring to FIG. 5, the gauge sensitivity, S.sub.XA, corresponding to any measured value of collector current i.sub.+x, can be calculated in the computation block 48 according to the equation

(S.sub.XA -S.sub.1A)/(S.sub.2A -S.sub.1A)=(i.sub.+x -i.sub.1)/(i.sub.2 -i.sub.1) or by rearrangement S.sub.XA =(S.sub.2A -S.sub.1A)(i.sub.+x -i.sub.1)/(i.sub.2 -i.sub.1)+S.sub.1A where all the terms on the right are known either from direct measurement or from the data stored in memory 50. S.sub.XB can be calculated similarly. S.sub.XA and S.sub.XB are indicated on FIG. 5.

Referring to FIG. 6 the cathode heating power W.sub.XB corresponding to any i.sub.+x can be calculated in computation block 48 according to the equation.

(W.sub.XB -W.sub.1B)/(W.sub.2B -W.sub.1B)=(i.sub.+x -i.sub.1)/(i.sub.2 -i.sub.1) or by rearrangement W.sub.XB =(W.sub.2B -W.sub.1B)(i.sub.+X -i.sub.1)/(i.sub.2 -i.sub.1)+W.sub.1B where all the terms on the right are known either from direct measurement or from data stored in memory 50. W.sub.XA can be calculated similarly. W.sub.XA and W.sub.XB are indicated on FIG. 6.

Referring to FIG. 7, the corrected value of the gauge sensitivity, S.sub.x, corresponding to any value of i.sub.+x can be calculated in computation block 48 according to the equation

(S.sub.x -S.sub.XB)/(S.sub.XA -S.sub.XB)=(W.sub.XB -W.sub.x)/(W.sub.XB -W.sub.XA) or by rearrangement S.sub.x =(S.sub.XA -S.sub.XB)(W.sub.XB -W.sub.x)/(W.sub.XB -W.sub.XA)+S.sub.XB where all the terms on the right are known from prior calculations.

S.sub.x indicated on FIG. 7 is the correct value of the gauge sensitivity corresponding to the actual conditions which existed when i.sub.+x was measured at the unknown pressure P.sub.x.

This value of S.sub.x is then used in Eq. 3 to calculate the correct value of the unknown pressure P.sub.x which can then be displayed as is well known in the art.

When it is not desired to correct for changes in cathode heating power, data sets A, B and C become identical and FIG. 5 reduces to a single curve, say curve A. In this example the correct value of S, namely S.sub.XA, can be calculated by interpolation along curve A. This correct value, S.sub.XA, is then used to calculate and display the correct value of the unknown pressure P.sub.X corresponding to i.sub.+x in FIG. 5 where the i.sub.1 and i.sub.2 values of FIG. 5 are selected since these values most closely approximate the measured value of i.sub.+x corresponding to the unknown pressure. Thus, it can be seen that the utilization of the collector current i.sub.+ to compensate for sensitivity variations may be treated as a special case when it is not desired or necessary to correct for changes in the cathode heating power.

As can be appreciated from the foregoing, interpolation techniques are employed to obtain Sx. Alternatively, if the number of data sets of Table 1 is sufficiently large, non-interpolative techniques may be employed. It should also be noted that the curves of FIGS. 5-7 are linear. Although collector current variation and cathode power variation may not be linear over the pressure range of interest, non-linear portions may be approximated by a series of linear portions as is generally known in curve approximation.

Claims

1. In controller circuitry for controlling the operation of a predetermined ionization gauge having a source of electrons, an anode, and an ion collector electrode, said controller circuitry including means for biasing said electron source, anode, and collector electrode at predetermined potentials; means for substantially maintaining the electron emission current from said electron source at a predetermined value; and means for measuring the ion current to the collector electrode to obtain a measured value of the ion collector current at an unknown pressure; the improvement comprising:

means for storing a plurality of calibration data sets for at least collector electrode currents and gauge sensitivities obtained with at least one reference gauge at representative values of at least known pressures where said reference gauge has substantially the same gauge sensitivity at any given pressure as does said predetermined gauge;

means responsive to at least one of said data sets together with said measured value of the ion collector current for calculating the gauge sensitivity corresponding to the unknown pressure; and

means responsive to the calculated sensitivity, the measured ion collector current, and the electron emission current for calculating the unknown pressure.

2. Controller circuitry as in claim 1 including means for selecting those two collector electrode current values which most closely approximate the measured ion current to the collector electrode.

3. Controller circuitry as in claim 1 where each of said plurality of calibration data sets is also obtained for representative values of one or more of the electron emission current, the type of gas whose pressure is being measured, and the potentials applied to the electron source, the anode, and the collector.

4. Controller circuitry as in claim 1 where said ionization gauge includes shield means defining a shielded volume within which said electron source, said anode, and said collector electrode are disposed where the shield means is at least partially open to permit transfer of gas molecules into the shielded volume and to permit external electrical connections to the electron source, the anode, and the collector electrode and where the shield means completely encloses the electron source, the anode and the collector electrode so that potentials external to the shield means do not disturb the electric charge distribution on the electron source, the anode, and the collector electrode to thus stabilize the sensitivity of the gauge.

5. Controller circuitry as in claim 4 where the gauge is of the nude type and has insulated electrical feedthroughs for said electron source, anode, and collector electrode and where the shield means includes (a) a longitudinally extending portion, (b) a first end portion which shields a first end of the shielded volume, and (c) a second end portion which shields the other end of the shielded volume from at least said insulated electrical feedthroughs.

6. Controller circuitry as in claim 4 where the gauge is of the non-nude type and has insulated electrical feedthroughs for said electron source, anode, and collector electrode and where the shield means includes a vacuum enclosure, a plasma shield for shielding one end of the shielded volume, and an insulator shield which shields the other end of the shielded volume from at least said insulated electrical feedthroughs.

7. Controller circuitry as in claim 4 where the gauge is of the Bayard-Alpert type.

8. In controller circuitry for controlling the operation of a predetermined ionization gauge having a source of electrons, an anode, and an ion collector electrode, said controller circuitry including means for biasing said electron source, anode, and collector electrode at predetermined potentials; means for substantially maintaining the electron emission current from said electron source at a predetermined value; means for measuring the heating power of the electron source to obtain a measured value of the heating power at an unknown pressure; means for measuring the ion current to the collector electrode to obtain a measured value of the ion current at an unknown pressure; the improvement comprising:

means for storing a plurality of calibration data sets for collector electrode currents, heating powers of the electron source, and gauge sensitivities obtained with at least one reference gauge at representative values of known pressures and cathode heating powers; where said reference gauge has substantially the same sensitivity at any given pressure and heating power as does said predetermined gauge;

means responsive to at least one of said data sets together with said measured values of the ion collector current and electron source heating power for calculating the gauge sensitivity corresponding to the unknown pressure so that the calculated sensitivity is compensated for variations in the electron source heating power; and

means responsive to the calculated sensitivity, the measured ion collector current, and the electron emission current for calculating the unknown pressure.

9. Controller circuitry as in claim 8 including means for selecting those two electron source heating power values which most closely approximate the measured electron source heating power.

10. Controller circuitry as in claim 8 where each of said plurality of calibration data sets is also obtained for representative values of one or more of the electron emission current, the type of gas whose pressure is being measured, and the potentials applied to the electron source, the anode, and the collector.

11. Controller circuitry as in claim 8 where each said data set is a function of the amount of electron source heating power required to provide a predetermined amount of electron emission current from the reference gauge where the work function of the electron source is varied over a predetermined range approximately corresponding to an expected range of work function variation of the electron source.

12. Controller circuitry as in claim 8 where said ionization gauge includes shield means defining a shielded volume within which said electron source, said anode, and said collector electrode are disposed where the shield means is at least partially open to permit transfer of gas molecules into the shielded volume and to permit external electrical connections to the electron source, the anode, and the collector electrode and where the shield means completely encloses the electron source, the anode and the collector electrode so that potentials external to the shield means do not disturb the electric charge distribution on the electron source, the anode, and the collector electrode to thus stabilize the sensitivity of the gauge.

13. Controller circuitry as in claim 12 where the gauge is of the nude type and has insulated electrical feedthroughs for said electron source, anode, and collector electrode and where the shield means includes (a) a longitudinally extending portion, (b) a first end portion which shields a first end of the shielded volume, and (c) a second end portion which shields the other end of the shielded volume from at least said insulated electrical feedthroughs.

14. Controller circuitry as in claim 12 where the gauge is of the non-nude type and has insulated electrical feedthroughs for said electron source, anode, and collector electrode and where the shield means includes a vacuum enclosure, a plasma shield for shielding one end of the shielded volume, and an insulator shield which shields the other end of the shielded volume from at least said insulated electrical feedthroughs.

15. Controller circuitry as in claim 12 where the gauge is of the Bayard-Alpert type.

16. An ionization gauge comprising:

a source of electrons;

an open anode defining an anode volume;

a collector electrode for collecting ions formed by impact between said electrons and gas molecules within said anode volume so that the pressure of the gas can be measured; and

shield means defining a shielded volume within which said electron source, said anode, and said collector electrode are disposed where the entire shield means is electrically conductive and at least partially open to permit transfer of gas molecules into the shielded volume and to permit external electrical connections to the electron source, the anode, and the collector electrode and where the shield means completely encloses the electron source, the anode and the collector electrode so that potentials external to the shield means do not disturb the electric charge distribution on the electron source, the anode, and the collector electrode to thus stabilize the sensitivity of the gauge.

17. An ionization gauge as in claim 16 where the gauge is of the nude type and has insulated electrical feedthroughs for said electron source, anode, and collector electrode and where the shield means includes (a) a longitudinally extending portion, (b) a first end portion which shields a first end of the shielded volume, and (c) a second end portion which shields the other end of the shielded volume from at least said insulated electrical feedthroughs.

18. An ionization gauge as in claim 16 where said longitudinally extending portion is cylindrical in configuration.

19. An ionization gauge as in claim 16 where the gauge is of the non-nude type and has insulated electrical feedthroughs for said electron source, anode, and collector electrode and where the shield means includes a vacuum enclosure, a plasma shield for shielding one end of the shielded volume, and an insulator shield which shields the other end of the shielded volume from at least said insulated electrical feedthroughs.

20. An ionization gauge as in claim 19 where said vacuum enclosure is metallic.

21. An ionization gauge as in claim 16 wherein the shield means is substantially maintained at a reference potential.

22. An ionization gauge as in claim 21 wherein the reference potential is ground potential.

23. An ionization gauge as in claim 16 where the gauge is of the Bayard-Alpert type.

24. A plurality of ionization gauges each providing coinciding measurement outputs under identical conditions, each said gauge comprising:

a source of electrons;

an open anode defining an anode volume;

a collector electrode for collecting ions formed by impact between said electrons and gas molecules within said anode volume so that the pressure of the gas can be measured to provide said measurement output of the gauge; and

shield means defining a shielded volume within which said electron source, said anode, and said collector electrode are disposed where the entire shield means is electrically conductive and at least partially open to permit transfer of gas molecules into the shielded volume and to permit external electrical connections to the electron source, the anode, and the collector electrode and where the shield means completely encloses the electron source, the anode and the collector electrode so that potentials external to the shield means do not disturb the electric charge distribution on the electron source, the anode, and the collector electrode to thus stabilize the sensitivity of the gauge;

wherein the locations of the electron source, anode, and collector electrode with respect to each other are substantially the same for each said gauge, whereby said gauges produce said coinciding measurement outputs under identical conditions.

25. A plurality of gauges as in claim 24 where the locations of the electron source, anode, and collector electrode with respect to each other are substantially the same for each gauge.

26. A plurality of gauges as in claim 24 where the dimensions of the shield means, the electron source, the anode, and collector electrode are substantially the same for each gauge.

27. In a method for controlling the operation of a predetermined ionization gauge having a source of electrons, an anode, and an ion collector electrode, said method including the steps of biasing said electron source, anode, and collector electron at predetermined potentials; substantially maintaining the electron emission current from said electron source at a predetermined value; and measuring the ion current to the collector electrode to obtain a measured value of the ion collector current at an unknown pressure; the improvement comprising the further steps of:

storing a plurality of calibration data sets for at least collector electrode currents and gauge sensitivities obtained with at least one reference gauge at representative values of at least known pressures where said reference gauge has substantially the same gauge sensitivity at any given pressure as does said predetermined gauge; and

calculating the gauge sensitivity corresponding to the unknown pressure utilizing at least one of said data sets together with said measured value of the ion collector current; and

calculating the unknown pressure utilizing the calculated sensitivity, the measured ion collector current, and the electron emission current.

28. A method as in claim 27 including selecting those two collector electrode current values which most closely approximate the measured ion current to the collector electrode.

29. A method as in claim 27 where each of said plurality of calibration data sets is also obtained for representative values of one or more of the electron emission current, the type of gas whose pressure is being measured, and the potentials applied to the electron source, the anode, and the collector.

30. A method as in claim 27 including providing shield means to define a shielded volume within which said electron source, said anode, and said collector electrode are disposed where the shield means is at least partially open to permit transfer of gas molecules into the shielded volume and to permit external electrical connections to the electron source, the anode, and the collector electrode and where the shield means completely encloses the electron source, the anode and the collector electrode so that potentials external to the shield means do not disturb the electric charge distribution on the electron source, the anode, and the collector electrode to thus stabilize the sensitivity of the gauge.

31. A method as in claim 30 where the gauge is of the nude type and has insulated electrical feedthroughs for said electron source, anode, and collector electrode and where the shield means includes (a) a longitudinally extending portion, (b) a first end portion which shields a first end of the shielded volume, and (c) a second end portion which shields the other end of the shielded volume from at least said insulated electrical feedthroughs.

32. A method as in claim 30 where the gauge is of the non-nude type and has insulated electrical feedthroughs for said electron source, anode, and collector electrode and where the shield means includes a vacuum enclosure, a plasma shield for shielding one end of the shielded volume, and an insulator shield which shields the other end of the shielded volume from at least said insulated electrical feedthroughs.

33. A method as in claim 30 where the gauge is of the Bayard-Alpert type.

34. In a method for controlling the operation of a predetermined ionization gauge having a source of electrons, an anode, and an ion collector electrode, said method including the steps of biasing said electron source, anode, and collector electrode at predetermined potentials; substantially maintaining the electron emission current from said electron source at a predetermined value; measuring the heating power to the collector electrode at an unknown pressure to obtain a measured value of the heating power; and measuring the ion current to the collector electrode to obtain a measured value of the ion current at said unknown pressure; the improvement comprising the steps of:

storing a plurality of calibration data sets for collector electrode currents, heating power of the electron source and gauge sensitivities obtained with at least one reference gauge at representative values of known pressures and cathode heating powers where said reference gauge has substantially the same sensitivity at any given pressure and heating power as does said predetermined gauge;

calculating the gauge sensitivity corresponding to the unknown pressure utilizing at least one of said data sets together with said measured values of the ion collector current and electron source heating power so that the calculated sensitivity is compensated for variations in the electron source heating power; and

calculating the unknown pressure utilizing the calculated sensitivity, the measured ion collector current, and the electron emission current.

35. A method as in claim 34 including selecting those two electron source heating power values which most closely approximate the measured electron source heating power.

36. A method as in claim 34 where each of said plurality of calibration data sets is also obtained for representative values of one or more of the electron emission current, the type of gas whose pressure is being measured, and the potentials applied to the electron source, the anode, and the collector.

37. A method as in claim 34 where each said data set is a function of the amount of electron source heating power required to provide a predetermined amount of electron emission current from the reference gauge where the work function of the last-mentioned electron source is varied over a predetermined range approximately corresponding to an expected range of work function variation of the electron source.

38. A method as in claim 34 including providing shield means to define a shielded volume within which said electron source, said anode, and said collector electrode are disposed where the shield means is at least partially open to permit transfer of gas molecules into the shielded volume and to permit external electrical connections to the electron source, the anode, and the collector electrode and where the shield means completely encloses the electron source, the anode and the collector electrode so that potentials external to the shield means do not disturb the electric charge distribution on the electron source, the anode, and the collector electrode to thus stabilize the sensitivity of the gauge.

39. A method as in claim 38 where the gauge is of the nude type and has insulated electrical feedthroughs for said electron source, anode, and collector electrode and where the shield means includes (a) a longitudinally extending portion, (b) a first end portion which shields a first end of the shielded volume, and (c) a second end portion which shields the other end of the shielded volume from at least said insulated electrical feedthroughs.

40. A method as in claim 38 where the gauge is of the non-nude type and has insulated electrical feedthroughs for said electron source, anode, and collector electrode and where the shield means includes a vacuum enclosure, a plasma shield for shielding one end of the shielded volume, and an insulator shield which shields the other end of the shielded volume from at least said insulated electrical feedthroughs.

41. A method as in claim 38 where the gauge is of the Bayard-Alpert type.

42. A method for measuring pressure with an ionization gauge having a source of electrons; an open anode defining an anode volume, a collector electrode for collecting ions formed by impact between said electrons and gas molecules within said anode volume so that the pressure of the gas can be measured; said method comprising the steps of:

providing a shield means to define a shielded volume within which said electron source, said anode, and said collector electrode are disposed where the entire shield means is electrically conductive and at least partially open to permit transfer of gas molecules into the shielded volume and to permit external electrical connections to the electron source, the anode, and the collector electrode and where the shield means completely encloses the electron source, the anode and the collector electrode so that potentials external to the shield means do not disturb the electric charge distribution on the electron source, the anode, and the collector electrode to thus stabilize the sensitivity of the gauge.

43. A method as in claim 42 where the gauge is of the nude type and has insulated electrical feedthroughs for said electron source, anode, and collector electrode and where the shield means includes (a) a longitudinally extending portion, (b) a first end portion which shields a first end of the shielded volume, and (c) a second end portion which shields the other end of the shielded volume from at least said insulated electrical feedthroughs.

44. A method as in claim 42 where the gauge is of the non-nude type and has insulated electrical feedthroughs for said electron source, anode, and collector electrode and where the shield means includes a vacuum enclosure, a plasma shield for shielding one end of the shielded volume, and an insulator shield which shields the other end of the shielded volume from at least said insulated electrical feedthroughs.

45. A method as in claim 42 where the gauge is of the Bayard-Alpert type.