RU2507490C1 - Sensor of absolute pressure of high accuracy based on semiconducting sensitive element with rigid centre - Google Patents

Abstract

FIELD: measurement equipment.

SUBSTANCE: sensor of absolute pressure comprises a body with a nozzle, a sealing contact block, a metal membrane, a non-compressible liquid, a semiconducting sensitive element comprising a glass base and a square profiled semiconducting crystal, in the centre of the thin part of which there is a stiff centre of square shape, on the working part of the semiconducting crystal there is a bridge measurement circuit formed of four strain resistance gauges. The size of the stiff centre is determined based on the following ratio: $$l_{s.c.}>{}^{h_{s.c.}}/{}_{\mathrm{1,432}}.$$

Centres of some strain resistance gauges included into opposite arms of the bridge measurement circuit and perceiving relative positive deformations are located at the distance from the centre of the crystal determined on the basis of the following ratio: $$R_1=\frac{L_{tp}}{2}\cdot r_1\left(L_{tp},l_{sc}\right)=\frac{L_{tp}}{2}\cdot \sum _{i=0}^2{k}_i{\left(\frac{l_{sc}}{L_{tp}}\right)}^i.$$

Centres of other strain resistance gauges included into opposite arms of the bridge measurement circuit and perceiving relative negative deformations are located at the distance from the centre of the crystal determined on the basis of the following ratio: $$R_2=\frac{L_{tp}}{2}\cdot r_2\left(L_{tp},l_{sc}\right)=\frac{L_{tp}}{2}\cdot \sum _{i=0}^2{l}_i{\left(\frac{l_{sc}}{L_{tp}}\right)}^i.$$

EFFECT: higher accuracy of measurement.

7 dwg

Description

The present invention relates to measuring technique and can be used to measure the pressure of liquid and gaseous aggressive media.

A known design of a pressure sensor with a sensitive element based on the structure "silicon on sapphire" [1], which is rigidly connected by high-temperature glass solder to a ceramic cup (with the formation of a cavity) and mounted by surface mounting on an elastic corrugated membrane that is sealed between the case and the lid, contains live conductors tracks. When the strain gauges formed on the sensitive element are deformed and are included in the bridge measuring circuit, an electrical signal arises at its output, which is directly proportional to the applied measured pressure.

A known design of the sensor element of the membrane type pressure sensor [2], which is a n-type single crystal silicon crystal, the planar side of which is oriented along the crystallographic plane (100) with a recess on the back of the crystal, forming a square membrane in plan. Four p-type conductivity strain gauges are formed on the planar side of the membrane so that their longitudinal axes are parallel to one of the main axes of the membrane, which coincides with the crystallographic direction [110].

A known construction of a semiconductor absolute pressure sensor, the sensitive element of which is made in the form of a single-crystal silicon wafer, diffusion resistors are formed on one working side and a protective cover with a vacuum cavity is sealed, a recess is formed on the second side, forming a membrane under the strain gauges, a rigid center is formed on the membrane [3].

The closest in technical essence to the proposed solution is the design of a semiconductor absolute pressure sensor selected as a prototype [4]. Such a sensor contains a housing with a fitting, a sealing contact block, a metal membrane, an incompressible liquid, a semiconductor sensitive element consisting of a glass base and a shaped semiconductor crystal (square), in the center of the thin part of which a rigid center (square shape) is formed. The profiled semiconductor crystal is electrostatically connected to a glass base in a vacuum, a vacuum cavity is formed between them, which provides absolute pressure measurements. On the working part of the sensitive element, strain gages are formed, combined in a bridge measuring circuit.

The essential features of the prototype, common with the claimed device, are as follows: a body with a fitting, a sealing contact block, a metal membrane, an incompressible liquid.

A common drawback of the designs of the sensitive elements of the pressure sensors described in [1–4] is the relatively large nonlinearity of the measuring circuit, which is due to the fact that strain gauges located on the crystal that sense relative positive deformations and strain gauges that sense relative negative deformations are not uniformly deformed. As a result of this, an unequal change in the resistances of the strain gauges of adjacent arms of the bridge measuring circuit occurs. An error appears from the nonlinearity of the bridge measuring circuit, which reduces the accuracy of the measurement. The non-linearity of the measuring circuit of the sensor depends on the symmetry coefficient k and the relative changes in the resistance of the arms of the measuring circuit ε ₁ , ε ₂ , ε ₃ , ε ₄ [5]. For strain gauge sensors, in which the relative change in the resistance of one arm usually does not exceed 0.01, for k = 1 the magnitude of the nonlinearity is ~ 0.3–0.6% if two arms are working. In addition, the known technical solutions do not provide high sensitivity (due to the non-optimal location of the strain gages on the crystal), manufacturability (due to the lack of the possibility of optimal location of the strain gages with different ratios of the size of the rigid center of the square shape to the size of the thin part of the square semiconductor sensitive element) and reliability (due to the possibility of destruction of the thin part of the semiconductor crystal when exposed to overload pressures).

The aim of the invention is to improve accuracy by reducing the non-linearity of the measuring circuit of the sensor.

This goal is achieved by the fact that in the known absolute pressure sensor containing a housing with a fitting, a sealing contact block, a metal membrane, an incompressible liquid, a semiconductor sensing element consisting of a glass base and a square shaped semiconductor crystal, in the center of the thin part of which a rigid center of a square is formed forms, on the working part of the semiconductor crystal a bridge measuring circuit is formed, consisting of four strain gages, respectively According to the proposed solution, the size of the rigid center is determined from the ratio:

_zh.ts. where h - the thickness of the rigid center, which is selected from the ratio:

where h _incl - the thickness of the thin part of the semiconductor crystal, while the centers of the strain gauges included in the opposite arms of the bridge measuring circuit and perceiving relative positive deformations are located at a distance from the center of the crystal, determined from the relation:

where L _incl. - the size of the thin part of the semiconductor crystal;

r ₁ (L _including , l _f.c. ) - the relative distance to the location of the centers of the strain gauges, perceiving relative positive deformation;

l _women - the size of the rigid center located in the center of the thin part of the semiconductor crystal;

k is a polynomial coefficient taking values in accordance with table 1;

subscript i is the index defining the polynomial coefficient in accordance with table 1;

в соответствии с таблицей 1;superscript i is the degree to which the variable is raised $$\raisebox{1ex}{$l_{\mathrm{well}.c.}$}\!\left/ \!\raisebox{-1ex}{$L_{t.h.}$}\right.$$

in accordance with table 1;

Table 1 index i coefficient k _i 0 0.904 one 0,048 2 0,061

and the centers of other strain gauges included in the opposite shoulders of the bridge measuring circuit and perceiving relative negative deformations are located at a distance from the center of the crystal, determined from the relation:

where r ₂ (L _including , l _f.c. ) - the relative distance to the location of the centers of the strain gauges, perceiving relative negative deformation;

l is a polynomial coefficient taking values in accordance with table 2;

subscript i is the index defining the polynomial coefficient in accordance with table 2;

в соответствии с таблицей 2;superscript i is the degree to which the variable is raised $$\raisebox{1ex}{$l_{\mathrm{well}.c.}$}\!\left/ \!\raisebox{-1ex}{$L_{t.h.}$}\right.$$

in accordance with table 2;

table 2 index i coefficient l _i 0 0,079 one 0.936 2 -0.031

Figure 1 shows the design of the proposed absolute pressure sensor of high accuracy. The sensor comprises a housing 1 with a fitting 2, a sealing contact block 3, a metal membrane 4, an incompressible liquid 5, a semiconductor sensing element 6. An incompressible liquid is poured through a tube 7 located in the contact block 3.

Figure 2 shows a 3D model of a semiconductor sensing element of the absolute pressure sensor, consisting of a profiled semiconductor crystal 8, which is connected electrostatically to a glass base 9 in vacuum. A rigid center 11 is formed on the thin portion 10 of the semiconductor crystal 8. Between the semiconductor crystal 8 and the glass base 9 there is a vacuum cavity 12, which provides absolute pressure measurement.

On the working part of the semiconductor crystal 8, a bridge measuring circuit 13 is formed (Fig. 3, a), consisting of four identical strain gages 14-17 (Fig. 3, b). The centers of the strain gages 14, 16, perceiving relative positive deformations, are located at a distance R ₁ from the center of the crystal, determined from relation (3), and the centers of the strain gages 15, 17, perceiving relative negative deformations, are located at a distance of R ₂ from the center of the crystal, determined from relations (4).

Consider an example.

Take the size of the thin part of the semiconductor crystal L _incl. = 2.5 mm, the thickness of the thin part of the semiconductor crystal h _incl. = 30 μm.

In accordance with expression (2), we determine the minimum thickness of the rigid center:

h _f.c. ≥4 · h _incl. = 4 · 30 μm = 120 μm.

Assume h _zh.ts. = 200 μm. In accordance with expression (1), we determine the minimum size of the hard center:

L accept _zh.ts. = 1 mm.

Substituting the values in the formula (3), we obtain:

In this case, the relative distance $$r_{\mathrm{one}}=\frac{2R_{\mathrm{one}}}{L_{t.h.}}=\frac{2\cdot \mathrm{1,166}mm}{2.5mm}=0.933$$

Similarly, calculate the distance R ₂ in accordance with formula (4) and table 2.

In this case, the relative distance $$r_{}$$$${}_2=\frac{2R_2}{L_{t.h.}}=\frac{2\cdot 0.561mm}{2.5mm}=0.449$$

The ratio for the size of the hard center l _zh.ts. in accordance with (1) defined based on the fact that the rigid center minimum size can not be less than the ratio of the thickness h of the rigid center _zh.ts. to the tangent 6 of the angle α = 54.4 ° - the angle of etching of silicon [6], multiplied by 2 (Fig.3, a), i.e.

The ratio for the minimum thickness of the hard center h _z.ts. in accordance with (2), it was obtained as a result of modeling deformations by the finite element method. At certain values of the size of the thin part of the semiconductor crystal L _incl. , The size of the rigid center positioned in the center of the thin portion of the semiconductor chip, l _zh.ts. , the thickness of the thin part of the semiconductor crystal h _incl. the thickness of the hard center h _z.c. and recorded data on the maximum relative positive deformations and maximum relative negative deformations. The data obtained have been constructed according to the absolute values of the maximum relative deformation of 18 positive and negative relative maximum deformation ratio of the thickness 19 of the rigid center h _zh.ts. to the thickness of the thin part of the semiconductor crystal h _incl. (figure 4). From these dependences it was found that when relation (2) is fulfilled, the maximum relative positive deformations and the maximum relative negative deformations take constant values.

The ratios for the relative distance of the location of the strain gages r ₁ and r ₂ included in formulas (3) and (4) were obtained as a result of the simulation of deformations by the finite element method. At certain values of the thickness of the thin part of the semiconductor crystal h _incl. , The thickness h of the rigid center _zh.ts. satisfying condition (2), the size of the thin part of the semiconductor crystal L _incl. changed dimension value of the rigid center l _zh.ts. in the range from a value satisfying condition (1) to a value tending to the size of the thin part of the semiconductor crystal L _incl. , and the values of relative strains at the boundary of the thin part were determined, i.e. at a distance of L _incl. / 2, the values of the relative deformations at the boundary of the rigid center, i.e. at a distance l _zh.ts. / 2, as well as the value of the distance from the center of the crystal, where the relative negative strains take the maximum value. 5 shows the dependence of the relative deformation ε on the working part of the semiconductor chip 8 when the ratio size hard center l _zh.ts. to the size of the thin part of the membrane l _incl. , where curve 20 is the relative strain at the boundary of the thin part of the semiconductor crystal, i.e. at a distance of L _incl. / 2 from the center of the crystal; curve 21 are the values of the modulus of relative negative strains at the boundary of the rigid center, i.e. at a distance l _zh.ts. / 2 from the center of the crystal; curve 22 shows the absolute values of the maximum relative negative strains at a certain distance from the center of the crystal. It was found that the relative strains 20 and 21 at the boundaries of the thin part and the hard center of the semiconductor crystal have different values, therefore, the placement of strain gauges in these places leads to an increase in the nonlinearity of the bridge measuring circuit of the sensor. The values of the maximum relative negative strains 22 are located at some distance from the rigid center. Then, the values of the distances from the center of the crystal were determined, where the relative negative strains take the maximum value, and the relative positive strains take a value equal in absolute value to the maximum relative negative strains. The data obtained were approximated by a polynomial in the range from a value satisfying condition (1) to a value tending to the size of a thin part of a semiconductor crystal L _incl. , as a result, expressions (3) and (4) were obtained.

Figure 6 shows the dependence of the relative distance r ₂ , where the relative negative strains take the maximum value, on the ratio of the size of the hard center to the size of the thin part of the semiconductor crystal obtained as a result of modeling strains by the finite element method (23), and by the formula (4) ( curve 24).

Figure 7 shows the dependence of the relative distance r ₁ , where the relative positive deformations take a value equal in absolute value to the maximum relative negative deformations, on the ratio of the size of the hard center to the size of the thin part of the semiconductor crystal obtained as a result of modeling deformations by the finite element method (25), and by the formula (3) (curve 26).

The pressure sensor operates as follows. The measured pressure acts on the metal membrane 4, which transmits the pressure through the incompressible liquid 5 to the semiconductor sensing element 6 (Fig. 1), consisting of a profiled semiconductor crystal 8, connected electrostatically to the glass base 9 in vacuum (Fig. 2). As a result of the fact that the thin part 10 of the semiconductor chip 8 is formed a hard center 11, the size of l _zh.ts. which is defined by the condition (1) and the thickness h of the rigid center _zh.ts. determined from condition (2), the reliability of the sensor increases, because when exposed to overload pressures, the rigid center 11 rests on the glass base 9 and, thereby, the destruction of the thin part 10 of the semiconductor crystal 8 is prevented. As a result of the pressure on the working part of the semiconductor crystal 8, deformations occur which are perceived by the strain gauges 14-17 (Fig. 3, b) included in the bridge measuring circuit 13 (Fig.3, a). The change in the resistance of the strain gages is converted by a bridge measuring circuit into the output voltage. In connection with the location of the centers of the strain gauges 15, 17, perceiving relative negative deformations, at a distance R ₁ from the center of the crystal, determined from relation (4), they turn out to be located in the zone of maximum relative negative deformations. Since the centers of the strain gauges 14, 16 perceiving relative positive deformations are located at a distance R ₁ from the center of the crystal, determined from relation (3), they find themselves in the zone of relative positive deformations in absolute value equal to the maximum relative negative deformations. Due to this arrangement of strain gauges, the nonlinearity of the bridge measuring circuit of the sensor is reduced, due to this, the accuracy of the sensor is increased compared to the prototype.

In addition, the accuracy of the sensor is increased by increasing sensitivity, since the strain gauges 15 and 17 are located in the zone of maximum relative negative deformations, and the strain gauges 14 and 16 are in the zone of equal relative positive deformations, while the relative changes in the resistances of the strain gauges 15, 17 and 14, 16 are stacked in a bridge measuring circuit. The proposed technical solution provides high sensitivity due to the optimal location of the strain gauges on the chip. The proposed high-precision pressure sensor based on a semiconductor sensitive element with a rigid center has improved adaptability, since it seems possible to determine in advance the optimal location of the strain gages for various ratios of the size of the rigid center of the square shape to the size of the thin part of the square semiconductor sensitive element. Also, the proposed pressure sensor has increased reliability, since it eliminates the possibility of destruction of a thin part of the semiconductor crystal when exposed to overload pressures.

Thus, due to the distinguishing features of the invention, the accuracy of the sensor is improved by improving the linearity of the output characteristic, high sensitivity and manufacturability are ensured by arranging the strain gages in an optimal manner for various ratios of the size of the rigid center of the square shape to the size of the thin part of the square semiconductor sensitive element.

Information sources

1. Stefanovich V.A., Lebedev G.B., Nelina S.N. Pressure sensor // Pat. 2392592 Russian Federation, IPC G01L 9/04. / Application 2009116703/28 from 30.04.2009; publ. 06/20/2010.

2. Belikov L.V., Razumikhin V.M. The sensitive element of the membrane type // Pat. 93027803 Russian Federation, IPC G01L 9/04. / Application 93027803/10 of 05/18/1993; publ. 12/27/1995.

3. Danilova N.L., Pankov V.V., Sukhanov V.S. Microelectronic Absolute Pressure Sensor and Absolute Pressure Sensor // Pat. 2362133 Russian Federation, IPC G01L 9/04, H01L 29/84. / Application 2007148423/28 of 12.27.2007; publ. 07/20/2009.

4. Barinov I.N. Semiconductor strain gauge pressure sensors based on the KND structure / Components and technologies. 2009, No5. - S.12.-15.

5. Vasiliev V.A., Tikhonov A.I. Analysis and synthesis of measuring circuits of information converters based on solid-state structures // Metrology. - M., 2003. - No. 1. - S.3-20.

6. Raspopov V.Ya. Micromechanical devices / Tula State University - Tula, 2002 - 392 p.

Claims (1)

An increased accuracy absolute pressure sensor comprising a housing with a fitting, a sealing contact block, a metal membrane, an incompressible liquid, a semiconductor sensitive element consisting of a glass base and a square shaped semiconductor crystal, in the center of the thin part of which a rigid square-shaped center is formed on the working part of the semiconductor a bridge measuring circuit consisting of four strain gages is formed in the crystal, characterized in that the size of the a certain center is determined from the relation:

where h _zh.ts. - the thickness of the rigid center, which is selected from the ratio:

where h _incl - the thickness of the thin part of the semiconductor crystal, while the centers of the strain gauges included in the opposite arms of the bridge measuring circuit and perceiving relative positive deformations are located at a distance from the center of the crystal, determined from the relation:

where L _incl. - the size of the thin part of the semiconductor crystal;
r ₁ (L _including , l _f.c. ) - the relative distance to the location of the centers of the strain gauges, perceiving relative positive deformation;
l _women - the size of the rigid center located in the center of the thin part of the semiconductor crystal;
k is a polynomial coefficient taking values in accordance with table 1;
subscript i is the index defining the polynomial coefficient in accordance with table 1;
superscript i is the degree to which the variable is raised

in accordance with table 1;
Table 1 index i coefficient k _i 0 0.904 one 0,048 2 0,061
and the centers of other strain gauges included in the opposite shoulders of the bridge measuring circuit and perceiving relative negative deformations are located at a distance from the center of the crystal, determined from the relation:

where r ₂ (L _including , l _f.c. ) - the relative distance to the location of the centers of the strain gauges, perceiving relative negative deformation;
l is a polynomial coefficient taking values in accordance with table 2;
subscript i is the index defining the polynomial coefficient in accordance with table 2;
superscript i is the degree to which the variable is raised

according to table 2,
table 2 index i coefficient l _i 0 0,079 one 0.936 2 -0.031

Images