There are a significant number of possible factors that can influence research results. Some factors that are easily investigated include light/dark cycles, temperature, humidity, air flow control, and watering and feeding procedures. However, two of the lesser understood and more challenging factors to examine are how acoustics and vibrations affect animal well-being.

Published studies have shown that sound and vibration can affect the results of research in which animals are used.Due to anatomical differences, the sensitivity of animals to sound and vibration differs when compared to humans. Since the animal species utilized in research labs vary greatly in size, it is important to recognize that the frequency and amplitude of sound and vibration will impact them differently.As an example, small animals are more sensitive to higher frequencies and lower amplitudes than a larger animal or human exposed to the same source. Therefore, the impact of acoustics and vibrations on animal well-being is often overlooked and under appreciated.

Limited by our hearing capability and supported by published data, the noise and vibration levels that impact animal well-being require significant consideration during the planning and design phases of projects. For facilities where construction is completed and breeding or research issues exist, it is important to understand where to begin investigating these issues.To accurately investigate the acoustic and vibration impact on the animals, it is critical to understand the measurement equipment limitations and environmental challenges.

Measurement Equipment Limitations and Differences
Acoustic and vibration measurement equipment is manufactured by a number of different companies, each offering a slightly different capability and equipment configuration.Some of the common vendors that manufacture acoustic and vibration measurement equipment include Agilent Technologies, Bruel & Kjaer, Larson Davis, LMS, National Instruments, and Scantek. Some measurement equipment is designed for a dedicated service such the programmable vibration monitor shown in Figure 1. Alternatively, measurement equipment is designed to be flexible and expandable to meet unique measurement requirements. The measurement equipment can be designed to have one channel for a transducer (accelerometer or microphone), or it may be able to expand to accommodate multiple transducers measuring simultaneously.

Many single channel measurement devices come as a battery-operated,handheld device. One major benefit to this design is the portability of the unit for measuring in multiple, different locations in a short period of time. Additionally, the handheld device generally will have internal memory for retaining the measured data. However, for measurements in animal lab spaces where the frequency range of the device is critical, there is a significant drawback in the fact that most handheld devices have an upper measurement frequency of 25 kHz.This upper frequency limitation will not cover the full hearing range of the majority of animals as illustrated in Figure 2.¹ In contrast, most multi-channel measurement systems require a computer to operate the system instead of functioning as a stand-alone device. Some are battery powered, but most require a supplemental power source, as well. Additionally, collected data will have to be saved to the computer which is controlling the device. But unlike the handheld devices, a number of the larger measurement systems are capable of measuring frequencies up to 100 kHz when outfitted with the proper data acquisition card and transducer combination.

As a general rule of thumb, a 1/2” or larger diameter microphone is not used for conducting measurements above 25 kHz. In some rare instances the performance (response) of the microphone will be such that a 1/2” microphone can be used. In cases where measurements are to be conducted at frequencies above 25 kHz, a 1/4” or 1/8” diameter microphone is preferred. Figures 3A and 3B illustrate the comparative dynamic responseof1/2” and 1/4”microphones. Figure3A shows that the 1/2”microphone performance is reduced beyond 10 kHz for this particular model. In contrast, Figure 3B shows steady performance out to 100 kHz prior to trailing off. It is worth noting that the 1/4” microphone (Figure 3B) does not have the smoothest of response between 30 kHz and 100 kHz but is still more appropriate than the 1/2” microphone for conducting a high frequency measurement.

Unlike microphones, accelerometers are used to measure vibration levels in structures. There are a number of differences that exist between accelerometers. As an example, accelerometers commonly come in one of two configurations.The first configuration is as a single axis measurement device while the other is a tri-axial measurement device. A single axis accelerometer is limited in the sense that it is capable of measuring in only one direction or orientation at a time. In contrast, a tri-axial accelerometer will measure simultaneously in three perpendicular directions. While beneficial, a tri-axial accelerometer will also require the measurement system to have three available channels.An example of a tri-axial accelerometer can be seen in Figure 4.

One of the most important performance characteristics of an accelerometer is its sensitivity. The sensitivity is important because it provides insight to the measurement resolution. As the sensitivity increases, the measurement results become more detailed and accurate. There is a trade off associated with the sensitivity, however, in that the frequency range within which the sensor is accurate may be limited. Therefore, when selecting a vibration sensor, it is preferable to know the frequency range to which the measurement will focus. Unlike acoustic measurements, vibration data generally will not exceed 3 kHz and does not require special frequency considerations. When the source of the vibration is understood, it is often recommended to take measurements focused over a 1 kHz range. By utilizing a lower frequency range, the resolution of the measurement will increase and the results will be more detailed.

Measurement Considerations in the Lab Animal Environment
Lab animal spaces provide a unique challenge for conducting acoustic measurements. The most significant challenge is understanding the effects of the hard washable surfaces which reflect sound, thereby making it difficult to localize the undesired noise source. Microphones should not be placed perpendicular to any rigid surface so as to not measure direct reflections of sound. Additionally, high frequency noise evaluation is challenging due to shorter wavelengths resulting in sound with an increased directivity and decreased energy. To acquire accurate sound levels and appropriate frequency content for high frequency noise, it is suggested to take the measurement close to the source being evaluated. In some instances, the measurement will need to be conducted within inches of the radiating source under evaluation. In general, measurements of this nature will be targeted at the cage and the laminar flow hood as these are the locations where high frequency exposure is most likely to occur for research animals. All things considered, it is suggested to conduct multiple measurements isolating the targeted equipment to be evaluated if possible. Equipment isolation can be accomplished by sequentially starting and stopping the components of interest.

Another challenge in evaluating a lab animal facility is the quantity, size, shape, and layout of the breeding, holding, and procedure rooms. The issue presented from room to room and facility to facility is knowing how many measurements are required to accurately represent the sound level. Additionally, when dealing with larger spaces, it is important to know where exactly tomeasure.Are the measurements taken near the ceiling, near the floor, or in the middle of the room? How close to the equipment do you measure?Are they taken at the face of the equipment or within the equipment? The answer depends on the purpose of the measurement. In the instance where human exposure to noise is the concern, measurements will be taken in the general space of the room at approximately head level. In contrast, when evaluating animal exposure, the measurements will be taken in or very near the caging and laminar flow hoods. Therefore, it is important to understand the purpose of the measurements prior to executing them.

Relative to conducting animal facility measurements, the following are some additional items to be considered prior to making any measurements:

1. Is an overall sound level or a peak resonant frequency desired?
2. How long should the data be averaged? This is dependent on the noise sources in the room being evaluated.
3. What type of data filtering should be used? Again, this will depend on the type of data expected. If resonances are expected, a peak hold feature will generally be used.
4. What frequency resolution should be used: narrow band, 1/3 octave band, full octave, or overall sound level?

For the above questions related to location, duration, and filtering there are many answers that are correct. However,many times the response to these questions is dependent on the response to the first question noted above.

From a vibrations perspective, there are a number of items to consider. First, the person conducting measurements should know what is or may be the source of the vibration. Is it a local source or an outside influence?A Local source is defined as one that is located within the environment being evaluated, while an outside source is one external to the location being observed. Second, is the vibration propagating to a point where the animals are housed? These two considerations are the major keys to evaluating vibration within the animal laboratory. For localized vibration sources, measurements should be made at both the source and the point of concern (animal housing). Similar to the acoustic measurements, a number of additional considerations include measurement duration, averaging, and filtering.

Lab Animal Facility Measurement Example
Recently, data was collected regarding neonatal mortality at a large research institution that houses rodents. The collected data suggested that the neonatal mortality was associated with maternal cannibalization and/or neglect of neonates which may be due to external stressors, including noise. Recognizing that an interaction exists between acoustic and vibration levels and animal welfare, a study was undertaken to qualify the acoustic and vibration levels present within animal holding rooms under varying conditions. The focus of the study included the identification of any high frequency resonances and elevated noise or vibration levels that may exist within the evaluated rooms. The measurements were conducted within the general holding room, internal to the caging systems, and around the laminar flow hoods. Noting that the upper limit of rodent hearing is significantly higher than that of humans, the acoustic measurements observed a frequency range of 20Hz to 100 kHz. Similarly, low frequency vibration measurements were conducted in the range of 4 Hz to 3 kHz.

The vibration measurements did not show any significant resonant frequencies or elevated vibration levels on the cage rack system or within the holding room. However, elevated resonant frequencies were noted on the laminar flow hood. The resonances were present in the same measurement which evaluated the blower installed in the top of the hood.The resonances measured on the front panel were noted at 724 and 960 Hz. Later computations showed that the two resonances resulted from the motor and blower impeller. Similarly, the acoustic measurements resulted in the identification of a 45 kHz resonance again associated with the laminar flow hood. The 45 kHz resonance was associated with the lighting system incorporated over the working surface inside the flow hood. Additional acoustic resonances were recorded at 64 and 87 kHz respectively without identifying the source.

Conclusion
A number of instrumentation and measurement considerations have been highlighted in an effort to increase awareness of the impact that acoustics and vibrations can have on animal well-being. It has been illustrated that there are more influences to collecting meaningful acoustic or vibration data than an understating of the measurement equipment. Without the proper understanding of how the instrumentation operates and how to conduct the measurements, it is possible that the results will be misleading or all together false. Thus, key knowledge in instrumentation selection and operation needs to be married with the understanding of how to conduct the measurements in a manner that will provide usable and meaningful results for those working in the animal lab.

References

1. Strain, George. “Deafness in Dogs and Cats.” 12 Dec, 2007. LSU. 29 DEC, 2008. <http://www.lsu.edu/deafness/deaf.htm>

Randall Rozema is a Mechanical Engineer – Vibration/Acoustics for ACS, Inc. Randy specializes in acoustic and vibration related issues with regard to facility and laboratory design. ACS, Inc., 3330 University Ave, Suite 200, Madison, WI 53705; (608) 663-1590 ext 367; rrozema@acscm.com; www.acscm.com.