WO1997012216A1

WO1997012216A1 - Vacuum absolute pressure measurement with resonator to excess temperature

Info

Publication number: WO1997012216A1
Application number: PCT/EP1996/004134
Authority: WO
Inventors: Herbert Bartuch; Hans Weiss
Original assignee: Leybold Vakuum Gmbh
Priority date: 1995-09-26
Filing date: 1996-09-21
Publication date: 1997-04-03
Also published as: DE19535651A1

Abstract

Pressure in a range between less than 10?-1 to 105¿ Pa can be determined with a high degree of sensitivity and largely independently of the gas type or composition. A silicon beam resonator (12) of 50 micron in thickness, 1 mm in width and 10 mm in length is stimulated by thermal microresistances (18) and heated via a strain gauge (20) to 10 °C excess temperature and measured. From the frequency increase and amplitude increase, an absolute pressure is obtained in a mass number characteristics family (m1, m2, m3). In addition, in a range above 10 Pa, an indication of the mean mass number (m) of the gas or gas mixture under investigation can also be obtained.

Description

VACUUM ABSOLUTE PRESSURE MEASUREMENT WITH RESONATOR TO OVER TEMPERATURE

The invention relates to a method for measuring absolute pressure in a gas-diluted room according to the preamble of claim 1.

Vacuum pressure measurements in a range from <10 ~ ¹ Pa to 10 ⁵ Pa are usually carried out with commercially available, according to the Pirani principle, heat conduction vacuum gauges or with capacity gauges. The former have the disadvantage that the pressure cannot be specified regardless of the type of gas and that there is only a slight sensitivity, particularly in the range> 10 ³ Pa. The measurement inaccuracy is around 50% there, in the remaining range around 20%. The capacitance gauges with an error of 0.5% on average are considerably more precise and measure regardless of the type of gas, but these advantages are offset by high manufacturing costs and sensitivity to vibration.

Various attempts have been made to expand the measuring range for the thermal vacuum measuring principle and to increase sensitivity. For example, thin foils made of good heat-conducting materials (semiconductors, metals) were provided with resistance meanders, which serve both to heat up and to detect the foil temperature. Depending on the size of the heated surfaces, high sensitivities of up to approx. 5-10 ^J Pa were achieved, especially in the lower pressure ranges. For pressures> 10 ³ Pa, the sensitivity very quickly became unacceptably low due to the prevailing pressure-independent convection as the determining heat transport mechanism. A significant improvement in this pressure range can be achieved by placing an additional heat sink over the heated surfaces at a distance (approx. 30 μm) adapted to the mean free path of the gas molecules. In this area, the heat is then transported even at relatively high pressure by pressure-dependent heat conduction, so that the change in temperature can be measured, for example, using thermopiles.

Resonant systems, such as quartz oscillators or mechanical beam or tongue resonators, show a characteristic dependence of their resonance frequency on the ambient pressure, which Tedoch is hardly suitable for pressure measurement in the range <10 ³ Pa. If the oscillation amplitude is additionally measured for beam resonators and integrated into the measurement signal, the result is only a shift in the lower detection limit of the pressure by approximately two powers of ten to 10 ¹ Pa. When using a micromechanically manufactured tongue or paddle resonator, there are usable amplitude changes down to 10 ^-1 Pa.

A detection of the type of gas in the measuring room or of changes in the gas composition is not possible with all known vacuum measuring methods. The invention provides a method of the type mentioned is to be so improved that it can reduce the pressure in the range of <1 ₀ ^-1 Pa to 10 ⁵ Pa with high sensitivity weitge¬ basis regardless of the type of gas or gas composition sen erfas¬.

This object is achieved by the features of claim 1.

The basic idea of the invention is to determine the resonance frequency and, at the same time, the mechanical vibration amplitude of a mechanical resonator, for example a bar resonator, tongue resonator or the like, which is at a certain excess temperature with respect to the environment, depending on the absolute pressure and the mean mass to measure the number of gas present in the measuring space, in order to first determine a characteristic field for absolute pressure p _a , average mass number m (type of gas), vibration amplitude A and resonance frequency f, which can be described mathematically or graphically. A gas-independent, unambiguous assignment to the prevailing absolute pressure is thus obtained, so that subsequently measured values of the vibration amplitude A and the resonance frequency f can be easily assigned to the prevailing absolute pressure with the aid of the characteristic field determined once. The simultaneous detection of the oscillation amplitude and resonance frequency has the additional advantage that in the pressure range from> 10 ² Pa to normal pressure the measurement effect at the resonance frequency, which decreases due to the decreasing temperature dependence of the thermal conductivity of the gas surrounding the resonator, is compensated by an increasing sensitivity in the amplitude measurement becomes. A micromechanical beam oscillator is used as the sensor element. The resonance frequency of such a beam oscillator initially changes only in connection with increasing damping of the resonator (the relationship between resonance frequency and damping in the case of a forced oscillation is known analytically). Up to this point, the resonance frequency and the damping and thus the amplitude are also dependent quantities in this case too. This means that it is not possible to determine two variables by measuring only one independent variable. It is only through the trick relating to the invention that the beam oscillator is additionally heated up that the frequency and the amplitude become independent variables, and thus make it possible to determine two variables (pressure and mean molar mass) from two measured variables (frequency and amplitude of the oscillator). . The heating of the bar results in a frequency dependence on the gas pressure, which is due to the thermal conductivity of the gas. The resonance frequency of the beam oscillator depends very much on the voltage state of the beam. If the bar is heated with a certain heating power, a certain pretension is set due to the corresponding thermal expansion. The pressure-dependent heat conduction of the gas, especially in the pressure range from 10 ^~ 1 to 10 ^ PA, results in a change in the overtemperature and thus a change in the pretension. The effect directly affects a change in the resonance frequency of the bar. The _U nteransprüche relate to advantageous embodiments of the invention. In particular, claim 4 is directed to the use of a known Si resonator with strain gauge full bridge circuit, which has proven to be particularly expedient for carrying out the method according to the invention.

_A Nhand the figures, the implementation of the method with the aid erfindungsge¬ MAESSEN is of such a per se known resonator explained in greater detail. It shows:

FIG. 1 shows a top view of a resonator known per se,

FIG. 2 shows a section along the line II-II in FIG. 1,

FIG. 3 shows a section along the line III-III in FIG. 1,

FIG. 4 shows a first embodiment of a measuring cell with a base for installation in the measuring cell,

FIG. 5 shows a sensor array constructed using the resonator according to FIG. 1 for installation in the measuring cell according to FIG. 4,

FIG. 6 shows another embodiment of a measuring cell with a built-in sensor array and base,

FIG. 7 shows a top view of the base shown in FIG. 6,

FIG. 8 shows a graphical representation of the change in resonance frequency measured with the measuring cell according to FIGS. 1 to 7 as a function of the absolute pressure, Figure 9 is a graphical representation of the amplitude change as a function of absolute pressure

FIG. 10 shows the characteristic field determined by plotting the change in resonance frequency against the change in amplitude with absolute pressure p and m as parameters.

The known mechanical resonator shown in FIGS. 1 to 3 is in H. Bartuch et al., "Resonant silicon sensors with electrothermal excitation and DM _S in metal thin film technology", 6th international trade fair with congress for sensor technology & system technology, Nuremberg, conference proceedings 2 (1993), pp. 17-24. This resonator, generally designated 10 in FIGS. 1 to 3, has, for example, a 50 μm thick and 1 mm wide silicon beam 12 as a bending oscillator, which is surrounded by a, for example, approximately 380 μm thick and thus comparatively rigid silicon frame 14. So that the beam 12 can swing freely, V-shaped longitudinal slots 16 are provided on both sides between the beam 12 and the frame 14. At the ends of the bar 12, the micro-heating resistors 18 necessary for thermal excitation are placed, while in the middle of the bar and in the vicinity of the bar ends, the strain-gauge resistors 20 are arranged in a suitable full-bridge circuit in such a way that the greatest possible bridge detuning is achieved when the excitation occurs. Connection pads are attached to the frame for power supply. Since this Si resonator 10 and its mode of operation are known to the person skilled in the art, they need not be explained in more detail here.

Using the micro-mechanically produced Si resonator with strain gauge full bridge shown in FIGS. 1 to 3 for converting the mechanical vibrations into an electrical signal voltage, a suitable measuring element according to the Figures 4 to 7 obtained by the fact that on the one hand the gate Resona¬ 1 ₀ 24 (Fig.4) or 26 is installed (Fig. 6) and operated in a tubular measurement cell, that at the same time by appropriate electronic control of the strain gauge resistors 20 be used as heating elements for generating a sufficiently high excess temperature (eg 10 K) of the resonator beam 12, and that on the other hand the V-shaped slots 1 ₆ , which separate the beam 12 from the frame 14, have a width of the order of magnitude have an average free path length of the gas molecules, ie a few μm to a few 10 μm. Since the DM _S resistors 20 are connected to form a full bridge and there is a sufficiently large bridge supply voltage, an alternating electrical voltage which is directly proportional to the mechanical oscillation can be tapped via the bridge diagonal, and the frequency and amplitude of an alternating electrical voltage which is not shown can be picked up by the owner However, the successor electronics available to the person skilled in the art can be evaluated according to the invention. A number of resonators 10 is combined according to FIG. 5 or FIG. 6 into a so-called sensor array 28, which can be installed in the measuring cell 24 together with a base 32 holding the electrical feedthroughs 30.

Figures 4 and 5 on the one hand and Figures 6 and 7 on the other hand show two possible embodiments of the measuring cell 24 and 26 with internals.

The one-off determination of the Afp -m characteristic field typical for a specific design of resonator and measuring cell is now carried out by flange-mounting the measuring cell 24 or 26 to a recipient which can be filled with different types of gas and which must be equipped with a measuring facility for absolute measurement. For different types of gas, the change in the resonance frequency (Fig. 8) or amplitude change (Fig. 9) as a function of pressure. Applying the determined frequency and amplitude values with p _a and m as parameters results in the characteristic field to be determined, which is shown in FIG. 10.

10, the absolute pressure to be measured can be determined as an intersection with an isobar at any time in an unknown gas or gas mixture by measuring the current f, A value pair with a measuring element 24 or 26 corresponding to the characteristic field. In addition, measurements in the range> 10 ¹ Pa give an indication of the average mass number m of the respective gas or gas mixture.

The measuring accuracy of the absolute pressure p _a and the average mass number m depends exclusively on the graphical procedure decisively on the number of gas types or gas mixtures included in the initial measurements. When using suitable interpolation methods (eg Tschbyschew approximation), the parameterization distance for p _a and m can be kept practically arbitrarily small, so that the accuracy of the values to be determined only depends on the error of the approximation coefficients. This determination can be carried out in a manner familiar to a person skilled in the art with the aid of suitable follow-up electronics.

Claims

PATENT CLAIMS

1. A method for measuring absolute pressure in a gas-diluted room, in which the resonance frequency and the oscillation amplitude at the resonance frequency of a mechanical resonator are measured and evaluated to determine the absolute pressure, characterized in that the resonator is related to the temperature of the gas ver is brought to excess temperature in a thin room, so that, depending on a variable absolute pressure (p _a ) as the first parameter and the variable mean mass number (m) of gases in the room one after the other as a second parameter, a characteristic field of the measured resonance frequencies (f ) and vibration amplitudes (A) is determined and that a gas-independent assignment of the currently measured values of the resonance frequency (f) and the vibration amplitude (A) to the absolute pressure (p _a ) prevailing in the gas-diluted space is determined becomes.

2. The method according to claim 1, characterized in that the characteristic field (Afp _a ~ m) mathematically described and / or is graphically represented.

3. The method according to claim 1 or 2, characterized gekennzeich¬ net that the characteristic field (Afp _a ~ m) for the determination of the respective current absolute pressure (p _a ) is interpolated electronically evaluated. 