Updated: Oct 12
There have been some interesting comments on our recent release of the test data of our BassMod line of bass traps.
It's understandable given that the tests,
run at NWAA Labs in Elma, Washington,
was not done in a conventional fashion i.e. per standard ASTM C423 standards. Why is that? the answer is simple, yet complex at the same time.
Both NWAA Labs owner Ron Sauro and I are long-time members of the ASTM E33 Acoustics Committee for the United States. We both also sit on the C423 sub-committee and have for a number of years. For those who may not be aware, the C423 standard has a frequency limit of 125Hz - 4kHz, per the standard, even though we see it expanded quite often, Why is that? Because in its standard use, it takes place in the diffuse field of the reverberation chamber, which is defined as .75 meters (29.52") from any wall surface.
The reason for that is that the products tested in the diffuse field were always intended to be defined as velocity-based. This means that products like fiberglass, mineral well, PET and other substrates, which absorb based on the nature of the absorbent material, can be tested per the parameters of the standard. The diffuse field is designed to homogenize the sound field and make its distribution, and decay, more even.
The problem is that neither Sabine, nor anyone else originally involved, imagined testing lower frequencies, much less diaphragmatic devices. The absorption nature of the diaphragmatic bass traps begins in the 125Hz rand and goes down from there. The action of the device requires waveforms low enough to excite the diaphragmatic plate, in multiple cycles, to achieve the test results.
In order of this to take place, the devices must be placed not in the diffuse field, but in the modal field. This is the area opposite of the diffuse field, against the walls. This also directly reflects the actual areas where these devices are used, i.e. in-situ. While some may consider this not in compliance with the standard, and it's not, there is no standard for low-frequency absorption testing. We are entering new territory here, even though some labs have used forms of this method before, including Riverbank and other well-known labs.
After 3 years of designing and testing our BassMod line at NWAA, we have arrived at some significant conclusions as to how to test low frequencies in a reverberation chamber. It should be noted that these tests are conducted using the standard C423 parameters, with the exception of the location of the test samples. Interrupted noise, mic positions, decay times, programs etc. are essentially all the same. However, here are the discoveries we have made along the way.
Frequency range: For obvious reasons, the frequency range of the test has to be expanded into the lower range. Due to the qualification size of NWAA Labs, it begins at 20Hz and extends, in this case, to 10 kHz.
Test Samples: Unlike standard C423 placements (A-Mount especially) the units are mounted against opposing walls, with the units flush against the wall and no air gap in back. This allows the units to be in the closest area of the modal field. Corners may be used for corner (angled) traps, but care must be taken to seal all back areas to eliminate false readings. The area of the devices was also kept as per C423 standards so that the empty room/full room balances were per C423 standards.
Spacing: One of the main concerns in testing bass traps is ensuring that there are no coupling effects. Unlike velocity-based absorption, testing diaphragmatic devices coupled can potentially cause a false narrative as to the performance of the individual units. In most cases, bass traps are not used in multiples. It was determined that spacing these units a minimum of 1 meter apart eliminates both coupling and edge diffraction effects. For the record, tests were performed using the standard "A mount" configuration per C423. This proved to be problematic and unsuccessful as the generation of the appropriate LF energy was minimalized and not reflective of the actual use of the products in-situ.
Microphone positions: The microphone positions, located in the diffuse field of the lab as per standard C423 practices, were kept as is. It is felt that the mics remaining in the diffuse field is appropriate for the overall response of the devices and the room combined. Putting the mics in the modal field would diminish the accuracy of the tests.
Size of Devices: Of special note is the premise that we should all recognize, but somehow have failed to acknowledge, that is "size matters". For years we have seen companies make claims about smaller format bass traps (2' x 2', 2' x 4' etc.) that can go down to 40Hz. Frankly, this has never made sense to me. The size of the device is directly proportional to the size of the waveform it is being asked to control. This has been proven in our testing. Because these units are tested uncoupled, we know what they can do individually. The diaphragm, and the associated dampening systems in the units, are important, no doubt, but the size and geometry of the unit will determind how low it can go. For instance, our 3248-8 unit (32" x 48" x 8" deep) test at a 63Hz (center frequency). Our 4848-8 tests at a 50Hz center frequency. While volume and other factors play a part in this, the fact is that these devices must have the ability to diaphragm at the modal frequency of choice. That cannot be manipulated in any other way. Physics does not change because we want it to, its demands are fixed by nature, not hyperbole.
The fall edition (2023) of Acoustics Today (ASA magazine) features a cover article titled Measuring Sound Absorption - The Hundred-Year Debate on the Reverberation Chamber Method. This article bears witness as to the many deficiencies in the C423 (and ISO354) standards and the difficulty of obtaining accurate results. It is this authors impression that this test methodology for diaphragmatic bass traps may actually be the most accurate use of this test method simply because it reduces the diffuse field anomalies between labs. Further testing using round-robin testing will need to be done to determine the potential changes in performance based on lab size. The lab size differences will have to be addressed to ensure the overlapping modes meet standards for the minimum frequency a lab can test.
It is my opinion that this method is accurate and repeatable given the proper setup and room conditions. We look forward to further research in the near future.