7.5Vacuum Gauges

7.5.1The Purposes (and Limitations) of Vacuum Gauges

The two most common reasons to know the quantitative level of vacuum in a system are either to know that a general level of vacuum has been achieved (so that something else can be done), or to see what quantitative effects an experiment has had upon the vacuum. The former requires a type of "go/no-go" reliability that may or may not have accuracy requirements. The latter is important with experiments such as gas creation or changes in vapor pressure. For proper analysis of experimental data, highly accurate readings are required. Regrettably, the greater the vacuum desired, the less reliable, or consistent, a vacuum reading is likely to be.

To obtain accurate and consistent low-, high-, and ultrahigh-vacuum readings requires years of experience and training beyond the scope of this book. This book explains the basic operation of common gauges and (hopefully) will give you the background to obtain the years of experience. Accuracy, however, is a somewhat loose term when used for vacuum gauges because actual (real) pressure must be experimentally determined.41 The reading that one obtains from a vacuum gauge can depend on factors as diverse as system use, experimental technique, the gauge angle of attachment, external magnetic fields near the gauge, and ambient gases within the vacuum system. The formal definition for accuracy by the ISO (The International Standards Organization) is "the closeness of agreement between the result of a measurement and the true value of the measurement."42 To maintain the best possible accuracy, recalibration is required to maintain the closest possible agreement with the manufacturer's stated accuracy. In addition, recalibration may be required each time a new gas species is introduced to a system. Unfortunately, all vacuum gauges tend to show a loss of accuracy over time, and this process continues over the life of the gauge. Recalibration cannot restore this lost accuracy, nor can re-calibration prevent error.

There are two basic ways for a vacuum gauge to "read" a vacuum: direct and indirect. For example, say that on one side of a "wall" you have a known pressure, and on the other side of the "wall" you have an unknown pressure. If you know that a certain amount of deflection implies a specific level of vacuum, and you can measure the current wall deflection, you can then determine the pressure directly. This process is used with mechanical or liquid types of vacuum gauges. On the other hand, if you know that a given gas will display certain physical characteristics due to external stimuli at various pressures, and you have the equipment to record and interpret those characteristics, you can infer the pressure from these indirect measurements. This indirect method is how thermocouple and ion gauges operate.

Each method of vacuum reading has its advantages and disadvantages. No single method of vacuum reading is entirely easy and/or comprehensive, and no vac-

uum gauge can read a full range of pressures or be wholly accurate on all types of gases. Thus, the vacuum gauge(s), controllers, and other peripheral equipment you decide to use on your system are dependent on what vacuum range you need to read, what level of accuracy you need, and cost.

Once you have decided on which vacuum gauge to use, you then need to consider where and how it should be attached. The vacuum gauge should preferably be attached near the area where the vacuum work will be performed. Pressure is determined by the number of molecules per cubic centimeter, and at low pressures a vacuum gauge located far away from where you are working may give a misleading value. Despite Avogadro's law, at different pressures there may be a random distribution of molecules within the vacuum. These differences can cause gauges to provide inaccurate pressure readings. There are three reasons for this problematic condition:

1.In molecular flow, there is a time factor before pressure can equalize within a vacuum system. Small-diameter (and/or long-length) tubing can compound this problem.

2.If the entry tube to the gauge is too narrow (and/or long), the statistical movement of molecules to enter the gauge may be inaccurately low. On the other hand if the angle of the tube to the gauge is in direct line with a molecular stream, the count of molecules could be inaccurately high.

3.Ion gauges have their own pumping capability providing lower pressures in the gauge region than in areas located some distance away (see Sec. 7.5.19).

Finally, be aware that if you read 3.0 x 10"3 torr from your vacuum gauge, that figure is not likely to be your "real" vacuum. Likewise, if you are trying to duplicate an experiment that is pressure-specific, and despite repeated attempts the experiment fails, your vacuum reading may be at fault, not you. Vacuum gauges, by their nature and design, are always inaccurate. In standard laboratory conditions, vacuum measurements that are ± 10% reliable are very difficult, and those that are ± 1% are essentially impossible.43 The reasons for this imprecision can include the following:

1.The gauge may be inherently inaccurate.

2.The gauge may be calibrated for a different gas than what is in your system.

3.The elements of the gauge may be contaminated.

4.The gauge may not be receiving a sufficient amount of time for proper equilibration.

5.The misalignment of parts within a gauge can alter the accuracy of readings.

6.Different components, or types of filaments, within the same gauge can alter the accuracy of readings.

404 Vacuum Systems

Always have a valve (or stopcock) between your vacuum system and vacuum gauge (and any other component of your vacuum system) to allow unneeded sections to be shut off when not in use and limit contact with potentially corrosive materials. The less contamination a gauge is exposed to, the longer the accuracy and reliability of the gauge will be maintained.

There are five major families of vacuum gauge design. The gauge families are the mechanical (see Sec. 7.5.2), liquid (see Sec. 7.5.4), thermionic (see Sec. 7.5.14), ion (see Sec. 7.5.19), and other (in this book I have only made passing reference to the momentum transfer gauge in Sec. 7.5.24). Each family has its own strengths and weaknesses, and many vacuum systems often will have two or more gauge families represented on a single vacuum line. This mixture is partly to allow uniform pressure readings from atmospheric to high vacuum (a single gauge cannot do both) or to provide periodic cross-reference and calibration.

7.5.2 The Mechanical Gauge Family

The easiest way to see a quantitative vacuum measurement of a system is to look at a dial and read it. This direct reading can be done with mechanical gauges. The surface of mechanical gauges that come in contact with gases within a vacuum system can be made out of metal or glass (borosilicate or quartz), both of which can be fairly impervious to chemical attack.*

Mechanical gauges rely on diaphragms, bellows, Bourdontubes, and capsules that are squeezed, pushed, pulled, twisted, and turned by the pressure on one side of the device to the vacuum that is on the other within a vacuum system. Mechanical gauges mechanically transfer this distortion to dials, mirrors, or pointers for reading. Many mechanical gauge designs can be quite accurate with proper calibration and monitoring^ For recording purposes, or to translate extremely minor amounts of distortion into readable figures, a displacement transducer can be connected to the moving part of a gauge.

The advantages of mechanical gauges are as follows:

1.Ease of use and the ability to read a positive and negative pressure.

2.The ability to provide a constant (as opposed to intermittent) reading.

3.The ability to provide continuous recording during use (if properly equipped).

4.If the proper construction material is selected, there can be limited reaction, or no reaction, between the gauge and materials within the vacuum system.

The disadvantages of mechanical gauges are as follow:

If there is any doubt whether the gases in your system may attack the materials of the vacuum gauge, ask the component's manufacturer.

1 For an unidentified type of quartz Bourdon manometer, an accuracy of about ±0.01% over 20% to

100% of its range, with regular recalibrations, has been reported.