Photoacoustic infrared technology provides an accurate leak detection method for chillers and other process cooling systems.

According to ASHRAE 15-2007, Section, “Each refrigerating machinery room shall contain a detector, located in an area where refrigerant from a leak will concentrate, that actuates an alarm and mechanical ventilation….” The U.S. Occupational Safety and Health Adminstration (OSHA) also mandates the use of monitoring devices to identify leaks of hazardous chemicals, as noted under 29 CFR 1910.119, “Process Safety Management of Highly Hazardous Chemicals.” While many different types of instruments are available to measure refrigerant gas concentrations and detect leaks, a number of plants are turning to photoacoustic infrared (PIR) devices for their ability to provide accurate leak detection at low concentration levels.

Infrared technology is based on the scientific principle that specific gases absorb infrared light at specific wavelengths. This principle allows gas volume to be identified and quantified by measuring infrared light absorption. Specific gases can be targeted selectively by narrowly filtering the wavelength of the infrared light introduced, thereby avoiding any infrared absorption by other non-targeted gases that might be present in the sample. While all infrared technology uses this principle, not all infrared systems measure gas concentration in the same way.

Non-dispersive (NDIR) or absorptive infrared systems determine gas concentration by comparing an air sample from the machine room to a sample of inert gas (usually nitrogen) stored in the monitor. The room sample and reference gases are individually irradiated with infrared light. The light absorption measured in each gas sample is compared, and the difference between the two measurements determines the concentration. Inaccuracies with this technology can occur when the assumed zero baseline measurement of the reference gas drifts due to changes in room temperature and atmospheric pressure, the aging of the infrared light source, or contamination of the reference cell. Systems must be pulled offline to recalibrate the instruments, and such downtime can be costly.

Photoacoustic infrared avoids the problems associated with zero drift by eliminating the need to compare results to a reference sample. Instead, PIR devices operate on the principle of sound. When a gas is irradiated with the light of a frequency that corresponds to a resonant vibration frequency of the gas, some of the light will be absorbed. This absorption causes some of the molecules of the gas to be excited to a higher vibration energy state and increases the pressure inside the instrument. This pressure creates an acoustic wave, which can be detected by an extremely sensitive microphone (figure 1).

Photoacoustic infrared devices directly measure the changes in pressure that occur when infrared light is absorbed by the refrigerant present in the sample. The greater the pressure, the greater the concentration. Conversely, when no pressure pulse occurs, then no gas is present. Because comparison to a reference gas is not used, PIR devices eliminate the possibility of zero drift and the associated downtime required for recalibration.

Figure 1. Photoacoustic infrared directly measures pressure changes caused by infrared light absorption. The higher the pressure, the greater the concentration of measured gas.

Sensitivity and Selectivity

The PIR measurement principle provides a high level of sensitivity and selectivity. The sensitivity of a refrigerant monitor is quantified in parts per million (ppm) volume. While a wide-band multi-gas infrared instrument has a sensitivity of 20 ppm, a photoacoustic infrared instrument can detect trace concentrations as low as 1 ppm. To put these numbers in perspective, 1 lb of refrigerant will evaporate to occupy 3 to 4 ft3 of volume, thus raising the concentration of a 30,000- to 40,000-ft3 room to 100 ppm. For some of the most commonly used refrigerants, such as ammonia and R123, a low leak detection level is crucial to ensure a safe working environment.

Selectivity is the ability of the instrument to differentiate between refrigerants. When only one refrigerant is present in the mechanical room and the likelihood of interfering vapors is remote, selectivity might not be an important issue. This rationale might also apply when there are multiple refrigerants,but they are all of Safety Group A1 (as defined by ASHRAE).

However, when both Group A1 and non-Group A1 refrigerants are present in the same space, or when other halocarbons or volatile hydrocarbons might also be present, selectivity can be important. Photoacoustic infrared offers a higher level of selectivity than any other instrument type.

For many process cooling applications, the leak detection capabilities provided by standard non-dispersive or absorptive infrared instruments are adequate. However, for operations that require a high level of accuracy, sensitivity and selectivity, photoacoustic infrared instruments provide another option in refrigerant gas concentration measurement. PC

Lisa West is the marketing manager for Thermal Gas Systems Inc., Roswell, Ga., a manufacturer of photoacoustic infrared leak detection instruments. For more information, call (800) 896-2996 or (770) 667-3865, or visit

SIDEBAR 1: System Design Considerations

Due to the wide variation in equipment room layouts, each situation must be considered individually. The following information is a general guideline for infrared leak detection systems.

Remote Notification. ASHRAE 15-2007 requires that audible and visual alarms be present inside the mechanical room, as well as outside each entrance to the mechanical room. System-compatible remote strobe lights and audible alarms should be used to meet these requirements. Display panels can provide gas concentration and diagnostic information as well as visual and audible alarms outside the mechanical room. Modern instruments also can communicate to building management systems.

Number of Sensors/Sample Points. How many sensors do you need? A good rule of thumb is that there should be one sensor or sample point for every 20,000 to 30,000 ft3 of room volume, or no less than one sensor/sampling point fewer than the total number of chillers, whichever is less; provided that there is one sensor for each refrigerant safety group used in the room.

Location of Sensors/Sample Points. The ability of a monitor to measure the refrigerant concentration depends on the location of the sensing point. The sensing point might be remotely located up to several hundred feet from the controller. The controller and sensor/sensing point should be rigidly mounted indoors. The controller should be located in an area where the display can be viewed from most parts of the room and where it can be easily accessed for occasional calibration and service. The sensor/sampling point location should be approximately 18" above the floor in an area where refrigerant vapors are most likely to accumulate. Sensors/sampling points should be located in low lying areas for occupant safety or near each potential leak source if refrigerant conservation is a high priority.

Airflow Patterns. Smoke tubes can be useful in determining the ventilation patterns. If there is a continuous draft in the room, a sensor/sampling point should be located downstream from the last potential leak source. Airflow patterns also can cause areas of the room to become stagnant and allow refrigerant vapors to accumulate. The sensor/sampling point location should be between the refrigerant leak source (chiller) and the ventilation exhaust.

Equipment Configuration. The equipment arrangement in the room also can affect the proper place to sample or locate a sensor. Two or more points of monitoring are recommended. As a general guideline, if there is one chiller in the room, sample at the perimeter of the unit. For two chillers, sample between them. For three chillers, sample between each pair of chillers. With four or more chillers, multiple monitors or a single monitor with a multipoint sampling system should be used.

Activities in the Room. The expected activity in the room should be taken into account when choosing sampling locations. Some activities might require locating the sensing point above or below the 12 to 18" height. Traffic patterns also can affect airflow. Consult your instrument supplier or another knowledgeable resource for assistance in determining the best locations for leak detection monitoring.