WHAT IS ULTRASOUND?
Sound has been defined as vibration of an air column to which a human ear would respond. Ultrasound is those frequencies that are above human hearing. Humans can usually begin to hear low frequency sound at about 30 Hertz (cycles per second). This is comparable to the low bass frequencies in a good Hi Fi system. The upper range for high frequency hearing is usually in the neighborhood of 15,000-20,000 Hertz or 15 to 20 kilohertz. Near either end of this range, hearing sensitivity has decreased significantly; sound has to be louder to be heard.Maximum human hearing sensitivity is in the 2.000 to 4,000 Hertz range, which includes most speech frequencies. As children age toward adulthood, they gradually lose some hearing sensitivity. Deterioration generally affects the high frequency hearing ability most, and affects males sooner than females. The discrimination might relate to the size (and consistency) of the male eardrum as compared to the female.
Even though we humans don't have much sensitivity to the higher (or to us, ultrasonic) frequencies, these sounds exist in nature. Bats use high frequency sonar to zero in on flying insects at and after dusk. Crickets rub their legs together and generate considerable high frequency noise as part of the chirp that we hear. It is interesting to listen to crickets or bats with an EFI leak detector. Some species of fish use very low frequencies for communication of a sort, and others use high frequencies. Rodents can communicate at frequencies above our hearing range. Dogs and cats have high frequency hearing sensitivity above our limits. The so-called "silent dog whistles" generate high frequency sound that dogs can hear but humans cannot.
As the frequency (or pitch) of sound rises toward and beyond our upper hearing limit, its characteristics change considerably. If the source is "directional", such as inside a room with an opening, the low frequencies tend to spread out from the opening (source) of sound. High pitch sounds tend to form more of a beam, and the effect is more pronounced as the pitch rises. Higher frequencies are more directional. We know that if a sound source is distant, it does not sound as loud (the intensity is lower).
Sound is absorbed as it passes through air. High pitched sound is absorbed more than low-pitched sound. Think about distant thunder. We hear a low pitch rumble. If lightning strikes nearby, we also hear a crackle with the rumble. The crackle is the high frequency portion of the noise. The higher frequencies are often absorbed before reaching us.
Relative humidity is a factor in sound absorption. Absorption is much greater in a relative humidity of ten to twenty percent. As relative humidity rises toward or beyond 40%-50%, the absorption for all frequencies is considerably less. The relative humidity factor is emphasized for higher frequencies. High frequency absorption in low humidity can be several times greater than for low frequencies. The bats at Carlsbad Caverns in New Mexico have to be closer to their flying prey to locate them, as compared to more humid locations. Please remember that we are explicitly referring to sound in air.
A PRACTICAL USE FOR ULTRASOUND
There was little widespread commercial use of ultrasonic sound until the 1950's and 1960's when many televisions were equipped with ultrasonic receivers and external ultrasonic transmitters for remote control. Devices using infrared light later replaced this type of remote control. Research on devices using ultrasonic sound in air became relatively dormant after the loss of the only major product in that field. Only a few engineers who were especially interested in the field continued to work with high frequency airborne sound. Few other commercial applications have surfaced since the remote control era. Of course, if you see a bat flying toward you, you can frighten it away by giving it a blast of ultrasonic sound, but one does not see a bat flying toward one very often. Production of an ultrasonic Bat Chaser doesn't seem practical. A very loud source of high frequency sound energy can frighten mean dogs away. This is another practical application.
ULTRASONIC LEAK DETECTION
EFI started production of its line of leak detector tools in the 1960's. Then, as even now, the only generally practical alternative to these tools for locating leaks is the slow, inconvenient painting of suspected leak areas with soapy water, then watching for bubbles to appear. These detectors are extremely sensitive to the ultrasonic noise generated by leaking pressurized air (or other gas, including steam), but they are as insensitive to ordinary noise as our ears are to ultrasonic noises. The detectors convert ultrasonic sound waves down to sounds that we can hear with our ears. This is similar to the way radios convert radio waves to audible sound.
We all know that a gas/air leak generates noise; sometimes we can hear it. Whether we can hear it or not depends on, the amount of other noise in the vicinity to mask it, the size of the leak, characteristics (especially the size and shape) of the hole that allows the leakage, the type of material, and the pressure that forces the gas out of the hole.
Leaks generate ultrasonic noise as well as noise that we can hear with our ears. The problem with listening for leaks is that our ears are not sensitive enough, especially for small leaks. They are not very directional due to their design and purpose. Another problem is that there is almost always some noise in the vicinity to mask the leak source sound.
Using a leak detector to listen for the ultrasonic noise component is far superior to using our ears as detectors to locate leaks. First, the characteristics of ultrasonic sound have advantages. There is not generally much ultrasonic noise to mask leak noise. If there are other sources of ultrasonic noise, they are usually widely enough scattered to be unimportant, since ultrasonic noise is absorbed much more as it passes through a distance in air than is normal noise.
Ultrasonic detectors can be made much more sensitive than the ear to detect leaks. Ultrasonic noise is also much more directional, making it easier to find the leak vicinity, more closer, and then pinpoint the source. The detectors listen for noise in the 40-kilohertz region. This is a compromise. Lower frequencies are more susceptible to common low frequency noises. Higher frequencies are absorbed to a greater extent as they pass through the air.
To better visualize leaks, consider a hose attached to a bibb or faucet. The water is turned on, and there is a shutoff nozzle on the end of the hose. You might find a damp spot somewhere along the hose where a leak will soon be apparent. There might be another spot with a very mall hole that gradually forms a drip. Surface tension of water tends to hold it together until there is sufficient mass to pull it loose as a drop. There might also be a small hole in the hose that squirts a very fine stream of water. Finally, there might be a leak in the valve, with water coming out around the turn-off valve stem. If the water pressure were quite low, all of these problems would be less apparent. When the pressure is increased, each would be emphasized and become more prominent.
Suppose now that the same system is drained and dry, and is pressurized with air. The spot that was damp might or might not generate ultrasonic noise. Increasing pressure would increase the likelihood of detection. The source of the drip would probably be easy to detect. The air does not have surface tension, and there would be a small steady leak from the hole. As the air escapes from the holes, there would be turbulence and noise generated. As the pressure increases, noise increases. The hole that had a very fine stream of water should generate lots of noise and be detectable at a considerable distance.
Remember that this noise is generated at the point where gas is liberated from the higher pressure area (inside the hose, through the hole) and out into the lower pressure area. As the higher pressure leakage passes into lower pressure outside, turbulence results, which creates the noise.
In the case of the faucet (valve), and sometimes with threaded fittings, a different condition can exist. A high- to lower-pressure transition can occur inside the structure. The primary leak might escape into an area that is still slightly pressurized due to the leak itself, but with a lower pressure. There is a hole leaking into a leakier cavity. Since the pressure differential is less between the real source leak and the cavity it leaks into, less noise is generated
The noise from the source leak then passes along a somewhat devious course to the outside air. At this final opening, a little more sound might be generated to mix with the first (now diminished) sound. The result is often less noise than would be expected, requiring the detector to be closer to the leak.
Folded or rolled metal seams/joints on containers or high quality ductwork can cause similar problems. A cross section of this type construction is diagrammed below.
A situation somewhat related to the faucet/threaded-fitting condition is found in searching for leaks in a vacuum system. More noise generating turbulence exists inside the chamber than out, because the high-pressure area is outside. The outside air is forcing its way through the hole into the lower pressure area inside. Detectors on the outside can still be very useful even in this adverse situation.
A quick example of the sensitivity of the leak detectors is to put on the headset and turn on the detector. Hold your free hand a few inches away from the front of the detector, and gently rub your thumb and forefinger tips together. The noise you hear is friction generated ultrasonic noise converted down to the audible sound.
A similar example is to hold the detector opening a couple inches from one eye and then blink the eye rapidly. The friction of the eyelashes rubbing together generates the noise. Yet the detector does not react to ordinary sound in the vicinity. You do not hear people talking or machinery running.
INTERFERENCE SOURCES
If there is ultrasonic noise interference in the vicinity of leak searching, reduce the sensitivity of the detector and work closer to the item being tested, or try one of the following:
Common Sources of Interference:
Expansion Valves in high pressure refrigerant lines
As the high pressure liquefied refrigerant expands to gas, it can sound as if the entire surface of the expansion valve is leaking. To find leaks in these systems, it is generally better to shut the system down. While the high pressure is in the lines, make an immediate search for the leaks.
Glass bottles on conveyers
Bottles rattling against each other on a moving conveyer line can generate very high intensity ultrasonic noise. When searching for leaks in the vicinity of these lines, it is often better to wait until the line is stopped or to make a "sound proof" test area by enclosing the test area with visquine.
Pneumatic equipment and tools
Some air operated equipment regularly exhausts air, which amounts to a large leak. Try steps 1, 2, 3, above, if this does not help you may have to wait until the interfering equipment is not in use.
Grinders
Grinding machines and cutting wheels, especially when cutting metal, create large amounts of ultrasound.
Computers
Computers are now found almost everywhere; they are often in maintenance offices, and sometimes on factory floors to control equipment or processes. Some computers and some computer displays (monitors) generate ultrasonic noise. Transformers in both sometimes operate at an electronic frequency comparable to that of the leak detectors. The transformer structures can vibrate at the operating frequency and generate ultrasonic noise. A noisy high voltage (fly back) transformer is an example. This noise when heard using a leak detector does not sound like a leak, it usually sounds more like a tone, buzz, or whistle. This is a good example of where the meter reading would only tell part of the story. The meter would read as if a leak were being detected but by listening you would be able to identify the type of sound and know it was no a leak.
Arcing
Electrical arcing creates turbulence in the air which creates ultrasound. Again the signature is different from that of a leak. Low level arcing, called corona, would sound like a 60 cycle buzz. Arcing or sparking as around a defective spark plug wire would sound more like crinkling cellophane.