Miniature-Ionization Vacuum Gauge Module
Miniature-Ionization Vacuum Gauge Module
B-RAX 3000 Multi-Guage Controller
Convection Vacuum Gauge Module
Convection Vacuum Gauge Module
Convection Vacuum Gauge Controller
Convection Type Vacuum Gauge Tube

Application Note: Using thermal vacuum gauges with different gases

Principal of Operation

Thermal conductivity gauges measure pressure indirectly, by sensing the loss of heat from a sensor to the surrounding gases. The higher the pressure of the surrounding gas, the more heat is conducted away from the sensor. Pirani thermal conductivity gauges maintain a sensor (usually a wire) at some constant temperature, and measure the current or power required to maintain that temperature. A standard Pirani gauge has a useful measuring range of about 10-4 Torr to 10 Torr. By taking advantage of convection currents that are generated above 1 Torr, convection-enhanced Pirani gauges increase the measuring range to just above atmosphere.

Output Signals and Displays

A vacuum gauge sensor, by itself, does not generate any output signal or readout. Electronic signal conditioning is required to convert heat loss into a useable signal and/or readout of pressure. Note that the output signal voltage is not the same as the displayed pressure readings. Table 3 below shows the displayed readings at various pressures for selected gases.

Non-Linear Output

Granville-Phillips' first ConvectronŽ gauge controllers produced a non-linear output signal of 0.375 to 5.659 VDC for 0 to 1000 Torr of N2, roughly in the shape of an "S" curve, as shown at right. GP adopted the same output curve for most of their Mini-ConvectronŽ modules and controllers with non-linear output (though in recent years, some GP controllers produced a different S-curve).

The non-linear output from InstruTech convection gauges, modules, and controllers duplicates exactly the original Scurve of 0.375 to 5.659 VDC for 0 to 1000 Torr.

Table 1 below contains the lookup data for converting the non-linear output voltage into pressure values.

Log-Linear Output

Many InstruTech modules and controllers also provide a log-linear output signal, as an alternative to the non-linear signal described above. This output, shown at right, is a 1 Volt per decade signal that may be easier to use for data logging or control.

Table 2 below contains the lookup data and provides the formulas for converting the log-linear output voltage into pressure values.

Linear Output

Some InstruTech controllers & modules offer a userprogrammable linear analog output. In this case, the user chooses the two end points, and the signal is a linear output voltage between those two points.

A couple of examples are shown at right:

Example-A: A full scale of 10 Volts was chosen to equal 1 Torr, and 10 mVolts was chosen to equal "zero" pressure (considered here to be 10-4 Torr). [Note: These are the factory default settings, which also happen to be the same as the Granville-Phillips linear output.]

Example-B: In this case, a full scale of 5 Volts equals 100 mTorr and 1 Volt equals 10 mTorr.

Because you can choose virtually any zero and span settings, there are an infinite number of possible linear signal configurations. If needed, you must generate your own table of signal vs pressure values for your particular setup.

Some considerations in choosing the best analog signal to work with:

1. The non-linear S-curve is convenient if you are replacing a ConvectronŽ which has that output, and you don't want to change anything in the control system. For new systems, however, the formulas needed to interpret the non-linear signal are complex and difficult to use. The linear or log-linear signal is generally easier to interface with today's data acquisition or control systems.

2. The linear signal is good when you want to monitor pressure over a limited range of 1 or 2 decades. If you try to use the linear signal over too wide a range, the signal at the low end of pressures will be too small to be useable by most electronic systems. For example: If you tried to setup the linear signal to cover the range from 1000 to 10-3 Torr, the signal at 10-3 Torr would be 1 microVolt. For most systems, that level of signal would be lost in the noise and you could never see it. For this reason, and to distinguish a valid low pressure reading from a power-loss/off/fail condition, the minimum linear signal cannot be set lower than 10 milliVolts.

3. If you need to monitor pressure over a wider range, you should use the log-linear signal. This output covers the entire measuring range with a high level 1 to 8 Volt signal. The signal changes by 1 Volt for each decade, and, even at the very low end, the smallest signal is still milliVolts, not microVolts. Also, the formulas defining the relationship between Voltage and pressure are simple and can be readily programmed into most computers / data acquisition systems.

The discussions above, about signals, assumed the gas being measured is Nitrogen. The following discussion illustrates how other gases affect both readings and outputs from thermal vacuum gauges.

Effect of Different Gases on Signals and Displayed Pressure

A thermal conductivity gauge senses heat loss, which depends on the thermal conductivity of the gas surrounding the sensor. Since different gases, and mixtures, have different thermal conductivities, the indicated pressure readings and outputs will also be different. InstruTech convection gauges (and most other thermal gauges) are normally calibrated using nitrogen. When a gas other than N2 is used, correction must be made for the difference in thermal conductivity, between N2 and the gas in use. The charts and tables below indicate how different gases affect the display and output from an InstruTech convection gauge.

For N2 the calibration shows excellent agreement between indicated and true pressure throughout the range from 10-4 to 1000 Torr. (As it should, of course, since they're calibrated for N2.) At pressures below 1 Torr, the calibration curves for the different gases are similar: the difference in readings at these low pressures is a constant, a function of the difference between thermal conductivities of the gases.

At pressures above 1 Torr, indicated pressure readings may diverge significantly. At these higher pressures, convection currents in the gauge become the predominant cause of heat loss from the sensor, and calibration depends on gauge tube geometry and mounting position, as well as gas properties.

Generally, air and N2 are considered the same as far as thermal conductivity goes, but even these two gases will exhibit slight differences in readings at higher pressures. For example, when venting a system to atmosphere using N2, you may see readings change by 30 to 40 Torr after the chamber is opened and air gradually displaces the N2 in the gauge.

For most other gases, the effect is much more significant, and may result in a hazardous condition, as described below.

Other considerations when using gases other than N2

Flammable or explosive gases:

InstruTech CV gauges and modules are not intrinsically safe or explosion proof, and are not intended for the measurement, or use in the presence, of flammable or explosive gases or vapors.

Under normal conditions, the voltages and currents in InstruTech CV Gauges and Modules are too low to cause ignition of flammable gases. However, under certain failure conditions, sufficient energy could be generated to cause flammable vapors or gases to ignite or explode.

Thermal gauges like the InstruTech CV gauges and modules are not recommended for use with flammable or explosive gases.

Moisture / water vapor:

In some processes (lyophilization, for example). the gas composition may not change significantly, except for moisture content. Water vapor can significantly change the response of a thermal gauge and correction should be made, as you would for any other gas.

Other contaminants:

If your gases condense, coat, or corrode the sensor, the gauge calibration and response to different gases will change. Generally, if the gauge can be "calibrated" ("zero" and "atmosphere" settings), these changes are small enough to be ignored. If you cannot set zero and atmosphere, the gauge should be replaced or cleaned. (See the Application Note on cleaning for more information.)

Gas Correction Chart

The Y- Axis is actual pressure, measured by a capacitance manometer, a diaphragm gauge that measures true total pressure, independent of gas composition.

The X-Axis is the pressure reading indicated by the convection gauge under test.

The chart shows readings for an InstruTech CVG as well as for a GP ConvectronŽ gauge, to illustrate that the response of the gauges to these different gases is virtually indistinguishable. (Caution: Do not assume this data applies to other convection gauges, which may or may not be the same.)

If the gas is N2, when the true total pressure is 100 Torr, the gauge will read 100 Torr

If the gas is Ar, when the true pressure is 100 Torr, the gauge will read only 10 Torr.

If you are backfilling your vacuum system with Ar, when your system reaches atmospheric pressure of 760 Torr true pressure, your gauge will be reading about 20 Torr. If you continue to backfill your system, attempting to increase the reading up to 760, you will overpressure your chamber and probably blow up your vacuum system!

If the gas is He, the gauge will read 1000 Torr when pressure reaches about 10 Torr actual pressure. You probably won't blow up the vacuum system, but opening the chamber to atmosphere prematurely may present other hazards for both people and product.

What these examples illustrate is that, without proper precautions,

use of different gases can result in injury to personal and/or damage to equipment.

Suggested precautions when using gases other than N2:

Install a pressure relief valve, or burst disk, on your chamber, to protect it from overpressure.

Post a warning label on your gauge readout "Do Not Exceed ____ Torr Indicated Pressure" (fill in the blank for your gas) so that an operator using the gauge will not exceed a safe pressure.

Table 1 - Non-Linear Analog Output Voltage vs Pressure, for Selected Gases

Note: By design, these values are identical to the outputs from Helix/Granville-Phillips ConvectronŽ gauges, Mini-ConvectronŽ modules, and Controllers, so that equivalent units can be interchanged without affecting your process system or software.

Table 2 - Log-Linear Analog Output Voltage vs Pressure, for Selected Gases

The log-linear output signal and pressure are related by the following formulas:

P = 10(V - 5) V = log10(P) + 5

where P is the pressure in Torr, and V is the output signal in Volts.

The chart on the following page shows the graphical results of the table and formulas above. Pressure is plotted on the X-axis with a log scale; the output signal is plotted on the Y-axis on a linear scale.

Log-Linear Analog Output Voltage vs Pressure

This is a plot of the formulas and data for the log-linear output signal, from the previous page.

Table 3 - Displayed Pressure Readings vs Pressure, for Selected Gases

Notes:

1) OP = overpressure

2) Display auto-ranges between Torr and mTorr at 1Torr
     Readings in normal font are in Torr. Readings in blue italics are in mTorr.



ConvectronŽ is a registered trademark of Helix Technology Corp.

   Download PDF version

InstruTech Vacuum Instrumentation

Utilizing the most recent advances in vacuum gauge sensor
and controller design.

 info@instrutechinc.com

Introducing Instru Tech's
B-RAX 3000- The new Multi-Gauge Vacuum Gauge Controller

 B-RAX 3000 Multi-Gauge Vacuum Gauge Controller Features