Session 11: Pressure Measurement

ITI104 - Instrumentacin e Informtica Industrial Curso 2013/2014

Grado en Ingeniera en Tecnologas Industriales

1. Introduction
This session examines the measurement of pressure. Some definitions are given first and then the various methods of pressure measurement are discussed.

2. Definitions
It is important to discuss the definitions related to pressure. There are three pressure definitions as depicted in Figure 1:

Figure 1: Relationship between Pressure Terms.

Absolute pressure: This is the difference between fluid pressure and absolute zero pressure. Gauge pressure: This is the difference between the pressure of a fluid and the atmospheric pressure (depends on atmospheric pressure). Absolute pressure = Gauge pressure + Atmospheric pressure: Differential pressure: This is used to describe the difference between two absolute pressures at two different points in the system. An example of such a situation is the difference in pressure that results from the introduction of an obstruction of the flow of a fluid in a pipe (Figure 1)

ITI104 - Session 11

2/21

Pressure Measurement

Figure 2: Difference in pressure caused by obstruction.

In terms of units, the SI unit of pressure is the Pa which is equivalent to Nm-2. This is such a small unit that it is impractical to use, and it is more likely to use the MPa as a practical unit (Nmm-2). Pounds force per square inch (psi) is used within the imperial system (1 psi is equal to 6 894.75729 Pa).
Table 1: Units of Measure.

The bar is also widely used, although not an SI unit. 1 bar is equal to 100000 Pa or 100 kPa. 1 bar is also nearly equal to one atmospheric pressure (one atmospheric pressure is equal to 1.01325 bar). 1 atmospheric pressure is sometimes measured in mm of Mercury (Hg), for example atmospheric pressure is equal to 760 mm Hg. The normal measuring range for pressure is from around 1 bar to 7000 bar.

3. The Nature of Pressure

Pressure is the action of one force against another over, a surface. The pressure P of a force F distributed over an area A is defined as: P = F/A Fluid pressure depends on the number of molecules striking a wall per unit time and on their average velocity. For an ideal gas:
ITI104 - Session 11 3/21 Pressure Measurement

Where: n is the molecular density (molecules per unit volume) m is the molecular mass (mass of one molecule) v is the root mean square molecular velocity The root mean square velocity depends on absolute temperature of the fluid as follows

where: k is the Boltzmann constant (1.3803x10-23 JK-1mol-1) T absolute temperature in Kelvin m is the molecular mass (mass in kg of one molecule) The mean free path is the average distance travelled by a molecule between collisions.

Where: r is the effective radius of the molecule n is the molecular density (molecules per unit volume) There are three laws that are worth remembering when dealing with pressure measurement. These are: Boyles law states the pressure is inversely proportional to volume at constant temperature. Charless law that states the volume is directly proportional to temperature at constant pressure. Pascals law states that a change in the pressure of an enclosed incompressible fluid is transmitted to every part of the container containing the fluid. Boyles law and Charless law can be combined into the ideal gas law, that states that: Where:
ITI104 - Session 11 4/21 Pressure Measurement

p V n R T

is the pressure of the gas is the volume of the gas is the amount of substance of the gas in moles is the gas constant (8.314472 JK1mol1) is the absolute temperature of the gas

4. Dynamic versus Static Pressure Measurement

It is important to note that most pressure measurements are carried out on static readings. Measuring the dynamic pressure is complicated. This is due to the fact that the measurement system itself will distort and affect the measured variable as a result of the dynamic nature of the pressure and how it affects the measurement system itself. A general arrangement of a pressure measurement system is shown in Figure 3 below. It consists of a tube that is used to transmit the pressure using a fluid, which transmits the pressure to a chamber that has a volume, V. This then affects the transducer/sensor. In general two factors affect the dynamic response of a pressure measurement system: 1. The transducer response. 2. The response of the pressure transmitting fluid and tube.

Figure 3: General arrangement of a pressure measurement system.

The transfer function of the system above can be derived. The input to the system is the pressure at the input of the tube. The output of the system is the pressure affecting the transducers/sensor. The dynamic relationship between these two variables is described by the transfer function. This takes on the following general form:

ITI104 - Session 11

5/21

Pressure Measurement

The damping ratio is defined as:

And the natural undamped frequency is defined as:

Where: V is the volume of the Chamber in m3 r is the radius of the tube in m c is the velocity of sound in the fluid in ms-1 is the dynamic viscosity of the fluid in Pas is the density of the fluid in kgm-3 L is the length of the tube in m If we reduce the radius of the tube (e.g., use a capillary tube) this increases the damping ratio reduces the natural un-damped frequency. This section illustrates the concept of the effect of the measurement system on the signal being measured. The pressure measurement system in this case has a second order transfer function. The system cannot faithfully measure input variable that has a frequency near the resonant frequency of the system. As a general rule of thumb, the maximum frequency of the input pressure must one decade below the resonant frequency of the measurement system.

5. Medium pressure measurement devices

There are three types of pressure measurement devices for medium pressure range: 1. Manometers: These consist of mercury filled U-tubes. The absolute or differential pressure can be read off the difference in height between the two columns of mercury. If one end is sealed and evacuated, the manometer will read absolute pressure.
ITI104 - Session 11 6/21 Pressure Measurement

Figure 4: Three forms of manometer: (a) U tube, (b) well type, and (c) inclined type.

2. Mechanical Effect Devices causing displacement: These are devices that convert the pressure measurement into a displacement. The following are examples of these devices: a. The Bourdon Tube: The Bourdon tube is a tube that has an oval cross section. It is coiled or curled under normal pressure but tries to uncoil/uncurl as the pressure increases. The movement can be displayed by connecting its tip to a pointer. The movement of the pointer is usually linearly proportional to the pressure. The Bourdon tube is usually of C shape. However its sensitivity can be increased by using a helical tube or a spiral tube. Shows graphical representations of the three types.

ITI104 - Session 11

7/21

Pressure Measurement

Figure 5: Different configurations of Bourdon tubes used for pressure measurement.

b. Bellows: Bellows are used to detect pressure and convert the pressure into displacement. They cannot generally be used for high pressures.

Figure 6: Bellows.

c. Diaphragm: Diaphragms can be used to convert pressure into displacement. They are of two types: flat and corrugated. The diaphragm is widely used in pressure sensing. In this section we examine the equations for the displacement and strain. We will examine circular diaphragms that are supported at the circumference. There are two types of strain at any point: radial and tangential. These are shown at a certain point away from the centre in Figure 5.

ITI104 - Session 11

8/21

Pressure Measurement

Figure 7: Circular diaphragm showing the radial and tangential strain at a certain point from the centre.

The pressure is uniformly distributed over the whole area of the diaphragm. At the centre of the diaphragm, the tangential strain equals the radial strain and can be calculated as follows:

Where: P is the pressure to which the diaphragm is exposed in Pa is the Poisson ratio of the material of the diaphragm (dimensionless) a is the radius of the diaphragm in m t is the thickness of the diaphragm in m E is the modulus of elasticity (Pa) The maximum deflection of the centre of the diaphragm can be calculated as follows:

The deflection is linear with the change in pressure provided that the maximum deflection is no more than 1/3 of the thickness of the diaphragm. Example 1 Calculate the maximum deflection and the maximum strain for a diaphragm that has the following parameters: a=25 mm
ITI104 - Session 11 9/21 Pressure Measurement

t= 1 mm Material is Stainless steel (E=210 GPa) =0.3 The measured pressure= 1 MPa Solution Subsituting in the equation for the maximum deflection gives:

This meets the 1/3 rule. The maximum strain at the centre can be found using the equation above:

The natural frequency of the diaphragm can be calculated as follows:

Where is the density of the material from which the diaphragm is manufactured (kgm-3) All other variables have been defined earlier in this section. Example 2 Calculate the frequency of the diaphragm in Example 1. Assume the density of Stainless steel is 7800 kgm-3. Solution Using the equation for the frequency:

It is important to ensure that the measured pressure signal has a frequency that is below this natural frequency of the diaphragm. There are two type of diaphragm: flat and corrugated. The diaphragm is corrugated to allow lower pressures to be measured. Corrugation has the
ITI104 - Session 11 10/21 Pressure Measurement

advantage that it allows more deflection and reduces the stress within the diaphragm material. A variation on the diaphragm is the bellows that produce a displacement with the change in pressure. But these are not suitable for higher pressures.

Figure 8: Schematic representation of a diaphragm pressure sensor.

The mechanical devices above will produce a displacement. This displacement can either be detected in one of the following methods: Visually: The displacement affects a pointer, as is usually done with the Bourdon tube devices. Electrically: The displacement is converted into an electrical signal using an electrical sensor. In such a case the mechanical device acts as the primary sensor converting the pressure into a displacement and the electrical device acts as secondary sensor converting the displacement into an electrical variable. This can be one of the following three devices: linear variable capacitor (LVC) in Figure 9 , potentiometer in Figure 10 and linear variable differential transformer (LVDT) in Figure 11.

ITI104 - Session 11

11/21

Pressure Measurement

Figure 9: Diagram showing the use of a linear variable capacitor to detect the displacement in a diaphragm

ITI104 - Session 11

12/21

Pressure Measurement

Figure 10: The use of a potentiometer with bellows to detect the displacement caused by pressure

Figure 11: Use of an LVDT to detect displacement from a pressure sensing capsule.

3. Electrical effect devices: a. Diaphragm strain-gauge devices: In addition to using a diaphragm to produce a displacement, a strain gauge can be
ITI104 - Session 11 13/21 Pressure Measurement

used to detect the strain. The strain gauge is then connected in a Wheatstone bridge circuit. The strain gauge can either be bonded to the diaphragm or diffused into it. As discussed above, the diaphragm can either be flat or corrugated. b. Piezo-electric devices: These devices rely on the principle that pressure on a piezo-electric device produces a voltage. c. Micro-mechanical piezo-resistive pressure transducers: These devices are micro-machined with the Silicon wafer and contain a piezo-resistive sensor that changes its resistance with the change in pressure. The piezo-resistive sensor(s) is(are) then connected in a bridge arrangement to produce a voltage output. These devices are extremely small in size, and are cheaper to produce making them very attractive compared to conventional devices. The devices discussed so far are only suitable for the measurement of medium pressures. For high and low pressure, other methods are needed. These are discussed below.

6. Low Pressure Measurement

Measuring low pressures is more difficult that measuring medium pressure. Pressures below one atmospheric pressure are sometimes referred to as vacuum. There are three methods for measuring low pressures: a) The most widely used method is the McLeod gauge. Figure 12.a shows the general form of a McLeod gauge in which low-pressure fluid is compressed to a higher pressure that is then read by manometer techniques. In essence, the gauge can be visualized as a U-tube manometer that is sealed at one end and where the bottom of the U can be blocked at will. To operate the gauge, the piston is first withdrawn. This causes the level of mercury in the lower part of the gauge to fall below the level of junction J between the two tubes marked Y and Z in the gauge. Fluid at unknown pressure Pu is then introduced via the tube marked Z, from where it also flows into the tube of cross-sectional area A marked Y. Next, the piston is pushed in, moving the mercury level up to block junction J.
ITI104 - Session 11 14/21 Pressure Measurement

At the stage where J is just blocked, the fluid in tube Y is at pressure Pu and is contained in a known volume, Vu. Further movement of the piston compresses the fluid in tube Y and this process continues until the mercury level in tube Z reaches a zero mark. Measurement of the height (h) above the mercury column in tube Y then allows calculation of the compressed volume of the fluid, Vc, as Vc=hA. Then, by Boyles law: PuVu=PcVc.

Also, applying the normal manometer equation, Pc=Pu+hg, where is the mass density of mercury, the pressure, Pu, can be calculated as

Compressed volume Vc is often very much smaller than the original volume, in which case the above equation approximates to

Figure 12: Low-pressure gauges: (a) McLeod gauge and (b) ionization gauge.

b) Ionisation gauge: The ionization gauge is a special type of instrument used for measuring very low pressures in the range 10-10 to 1 mbar. Normally, they are only used in laboratory conditions because their calibration is very sensitive to the composition of the gases in which they operate, and use of a mass spectrometer is often necessary to determine the gas composition around them. They exist in two forms known as a hot cathode and a cold cathode.
ITI104 - Session 11 15/21 Pressure Measurement

The hot cathode form is shown schematically in Figure 12.b. In this, gas of unknown pressure is introduced into a glass vessel containing free electrons discharged from a heated filament, as shown in Figure 12.b. Gas pressure is determined by measuring the current flowing between an anode and a cathode within the vessel. This current is proportional to the number of ions per unit volume, which in turn is proportional to the gas pressure. Cold cathode ionization gauges operate in a similar fashion except that the stream of electrons is produced by a high voltage electrical discharge. c) Pirani gauge: A typical form of Pirani gauge is shown in Figure 13.a. This is similar to a thermocouple gauge but has a heated element that consists of four coiled tungsten wires connected in parallel. Two identical tubes are normally used, connected in a bridge circuit, as shown in Figure 13.b, with one containing the gas at unknown pressure and the other evacuated to a very low pressure. Current is passed through the tungsten element, which attains a certain temperature according to the thermal conductivity of the gas. The resistance of the element changes with temperature and causes an imbalance of the measurement bridge. Thus, the Pirani gauge avoids the use of a thermocouple to measure temperature (as in the thermocouple gauge) by effectively using a resistance thermometer as the heated element. Such gauges cover the pressure range 10-5 to 1 mbar.

Figure 13: (a) Pirani gauge and (b) Wheatstone bridge circuit used to measure output.

ITI104 - Session 11

16/21

Pressure Measurement

7. High Pressure Measurement

For high pressure measurement special methods are needed. The most widely used method is called the Bridgeman Gauge. The resistance of fine wires changes with pressure according to the following linear relationship. R = R1 (1 + p) Where R1 Resistance at 1 atmosphere ( 100 KN/m2) in ohms Pressure coefficient of resistance in ohms/100 KN M-2 p gage pressure in KN/m2. The above said resistance change may be used for measurement of pressures as high as 100,000 atm., 10.00KN/m2. A pressure transducer based on this principle is called a Bridgman gauge. A typical gauge uses a fine wire of manganin (84% Cu, 12% Mn, 4% Ni) wound in a coil and enclosed in a suitable pressure container. The pressure coefficient of resistance for this material is about 2.5x10-11 Pa-1. The total resistance of the wire is about 100 and conventional bridge circuits are employed for measuring the change in the resistance. Such gauges are subjected to aging over a period of time, so that frequent calibration is required. However, when properly calibrated, the gage can be used for high pressure measurement with an accuracy of 0.1%. The transient response of the gauge is exceedingly good. The resistance wire itself can respond of variations in the mega hertz range. Of course, the overall frequency response of the pressure-measurement system would be limited to much lower values because of the acoustic response of the transmitting fluid.

Figure 14: Bridgeman Gauge.

ITI104 - Session 11

17/21

Pressure Measurement

8. Calibration: The Dead Weight Tester

The dead weight tester is a laboratory device that is used to calibrate other pressure measurement devices. A simple hand sketch is shown in Figure 15. The piston is first removed. Then a fluid is gradually filled in the chamber (with the plunger fully pushed in. Then, as the fluid is further filled, the plunger is gradually withdrawn to allow the fluid to fill all the space. The piston is then returned and the valve to the pressure gauge is opened. Weights are then placed on the piston. By dividing the weight of the weights by the area of the piston, the actual pressure can be known. This can compared to the reading of the gauge and calibration can be carried out if needed. The dead weight tester is an example of a source calibration system. The sources of error in this process are: 1. The friction between the cylinder wall and the piston. This can be reduced by continuously rotating the cylinder in order to reduce the friction. 2. Uncertainty in the knowledge of the area of the piston.

Figure 15: Schematic representation of a dead-weight tester.

The dead weight tester is not used in practice as a pressure measurement device. It is only used for the calibration of gauges.

ITI104 - Session 11

18/21

Pressure Measurement

9. General comparison
A general comparison chart for the ranges of pressure for various devices is shown on the next page in Figure 16. A chart showing the different methods of pressure measurement is shown in Figure 17.

Figure 16: Comparison chart for pressure measurement devices.

ITI104 - Session 11

19/21

Pressure Measurement

Figure 17: Overview of the different pressure measurement technologies.

ITI104 - Session 11

20/21

Pressure Measurement

10. Problems
1) Explain the difference among absolute pressure, gauge pressure, and differential pressure. When pressure readings are being written down, what is the mechanism for defining whether the value is a gauge, absolute, or differential pressure? 2) What are bellows pressure sensors? How do they work? Describe some typical applications. 3) How does a Bourdon tube work? What are the three common shapes of Bourdon tubes and what is the typical measurement range of each type? 4) Describe the three types of manometers available. What is the typical measurement range of each type? 5) Discuss the range of instruments available for measuring very low pressures (pressures below atmospheric pressure). 6) How are high pressures (pressures above 7000 bar) normally measured?

11. References
[1] Alan S. Morris and R. Langari, Measurement and Instrumentation. Theory and Application, Ed. Elsevier, 1st Edition, 2011. Some text is reproduced with authorization of Dr. Lufti Al-Sharif ([email protected])

ITI104 - Session 11

21/21

Pressure Measurement