Many times you will see discussions of how much extra SPL a subwoofer will gain by being placed into a room, a corner, a vehicle, or a 1/8th, 1/4th or 1/2 space environment rather than a full space or anechoic equivalent. There are often times assumptions made about when and where gain takes place in a "typical" space and by how much. These assumptions are used to make educated guesses during system design or when evaluating what the response will be or what SPL levels the system might be capable of.
You may have also seen simulations of subwoofer systems inside of enclosure modeling programs where the distance to the system or the environment that the system is operating in can be simulated as well. For example, a proposed system might be simulated at a 3 meter distance or placed into an eighth space environment and an assumption is made about the real space frequency response or output headroom based on this.
There are a few major errors in judgment commonly seen in these speculations:
Not accounting for the distance to the listening position from the subwoofer. Most system models are positioned one meter from the subwoofer, however, most listeners do not sit this close.
Not accounting for the construction of the space the subwoofer will be placed in.
Not accounting for other objects that may be close by.
Other things to consider are the volume, dimensions and shape of the space which also can have significant impact. There are programs that will help estimate the basic acoustics of a square or rectangular room but they only cover the most simple of spaces and parameters. In light of all this speculation, we decided to do a simple investigation on the effects of placing a basic subwoofer into an example of a fairly typical or average construction, shape and size room.
The Game Plan
The basic procedure for testing was as follows:
Side Note: This is the same testing procedure we used last year for the cabin gain measurement of the Jeep SUV.
A subwoofer is measured outdoors in a groundplane setting well away from any large objects with the microphone at a specified distance from the subwoofer, either one or two meters. (We used two meters in this case because two meter outdoor groundplane measurements are a good approximation of a one meter anechoic result.)
A drive level is settled on for the unit and the basic outdoor response and SPL at this drive level is captured.
This same exact drive level and subwoofer/amplifier/measurement rig is used to measure the SPL and response in the test room (more details on the room below). The difference between the two measurements will show the effect the boundaries and acoustics of the space has on the subwoofer’s basic raw response as delivered to the listening position. From this data, only the change imparted to the raw response can be subtracted.
Based on these results, we can then gauge what other subwoofers’ maximum headroom and basic response shape would be in the same placement, either through looking at outdoor groundplane results such as those presented here or inside of a simulation program.
The Subwoofer
To conduct this test example, a small 12" DIY sealed subwoofer was chosen for its predictable response and roll off into the deep bass and for its ability to be moved multiple times easily. It was measured at a two meter distance, outdoor groundplane at a known drive voltage.
The Test Room
The room is about 4,200 cubic feet, roughly 10 x 17.5 x 24ft, and is constructed with typical 2x4 studs, insulation and 5/8 sheetrock. The floor is wood over a crawlspace. There are four available subwoofer placements, a main listening position, and five additional listening positions. The subwoofer positions are in the front two corners of the room and along the back wall about a third in from either corner. The distance to the listening positions from the subwoofer placements are between 3.5 to 5.5 meters with an average distance slightly over 4 meters. The main listening position is right about 4 meters.
Testing
The response was measured at each of the six listening positions using each of the four subwoofer placements for a total of 24 in room measurements. This allows us the ability to look at the total average gain seen by a distributed four subwoofer system 1) at all listening positions, 2) at any one listening position (such as the main one), 3) at only one subwoofer placement, or 4) at a single subwoofer placement and a single position. A few different examples of this are presented here. What is interesting to note is that while some combinations do look drastically different, the overall results point toward large trends developing and showing up in the overall data set.
Results
Based on our testing, the effects of taking the subwoofer into the room results in very large changes in the response that arrives at the listening position. Looking at both the main position response and the total average response change, it is easy to see that generalizations or simulated guesses as to the amount of room gain are likely to be far off and can vary wildly with frequency.
For example, this room provides a decent amount of average gain between about 2-4dB from 190Hz until about 50Hz in both the main position average and the total 6 position average before exhibiting a huge spike between 40 and 50Hz which is related to the room’s worst resonance. However, at 30Hz there is an average loss of 4dB. Between 15 and 26Hz we again see a nice boost of about 7 or 8 dB on average but at 12Hz there is no net gain or actually a small loss before rebounding to about +7dB at 10Hz. From there, the gain skyrockets into the single digits.
What does this mean? This means that if you are the type of person who listens loudly to music or movies and want to maintain clean headroom, you're going to need a lot of power at 29-40hz because there is very little free lunch there in this room. If you are after subterranean HT bass down to 10Hz or lower at reference playback levels, it will require one beastly powerful infra bass system to provide flat response with headroom due to the area near 12Hz having no appreciable gain. On the other hand, even a small, cheap subwoofer should easily pressurize the room at 45Hz and any sub with legitimate 16Hz extension should provide good bass power until 15 or 16Hz in this room.
Conclusion
This is not a difficult process but it is a lengthy one taking about a half day to properly set-up. But, it is well worth it if you are at the system design stage, for instance if you plan to place one large subwoofer system in a single location such as an IB, or if you want to know what effect your room will have on your subwoofer’s response at the listening positions, or with multiple placements. You can look at simulations or make educated guesses but this may or may not be close to the reality of the situation in any way shape or form. It's a gamble at best. The room looked at here did not exhibit the extreme amount of gain that is often touted as occurring.
At 32Hz and 12Hz there was a net SPL loss in output at the listening position compared to anechoic one meter or two meters groundplane. Sure, there was a lot of gain seen in other areas of the frequency range but if clean headroom and flat response is to be maintained at the highest playback volumes, it is the areas with the least gain that are of concern.
The other take away here is that there is actually a large amount of gain, but that it is largely nullified due to the listening distance being four meters or slightly more from the subwoofer placement. If the subwoofer were placed in the corner and measured at one meter there would surely be much higher output and a different response shape, but how many people have their chair one meter from a corner loaded subwoofer? The gain and response seen at the listening position is what should be cared about.
If you wish to take the guess work out of the situation, this type of test can be done with nothing more than a free audio program, a volt meter, a very cheap sealed subwoofer, a good sized open space outdoors to do the baseline measurement and less than $200 in extraneous equipment. Unless your listening position is one meter from the corner the subwoofer is placed in, than any assumptions of the room response and amount of gain seen are likely little better than a guess.
Side Test: How much SPL from adding an identical subwoofer?
One other short test we conducted was to measure a subwoofer system with four separate, identical subwoofers placed away from each other in room and measure what happens to the frequency response and output when you start with one operating and add in the others one at a time until all four are operating at the same output level. This is shown in the chart below.
It is often times stated that the maximum gain that can be seen from adding an identical subwoofer producing the same output level is 6dB and that the full 6dB will only be realized if they are within 1/4 wavelength of each other or in other words, co-located or placed very close to each other. This is not correct in practice, in a confined environment, or basically anywhere other than very large anechoic chambers, outdoors, and huge spaces. There are a variety of factors that can cause a different interaction at the listening position such as the room acoustics, the distance from the individual subwoofers to the listening position, and the placement of each subwoofer in room.
You cannot simply assume a static gain of any amount over a substantial frequency range when the system is emitting from multiple points within a room or other confined space.
In the example given, the green trace is one subwoofer, red is two, blue is three and the black trace is with all four of them operating. It is easy to see that adding in the second subwoofer caused a gain larger than 6dB at the listening position in quite a few places – 12, 30, 40, 50 and 90Hz for example. Over some ranges output was actually lost (58-65Hz and 97-116Hz) most likely due to the phase relationship at those frequencies.
Looking at the gain seen from adding in the third subwoofer, note that again at 30Hz more than 6dB is gained and the system’s total radiation is not even doubling at this point by adding a third sub. At 80Hz we again see a slight loss.
By the time the fourth subwoofer is added in, the gain is between 2-4dB for the most part. However, note that the overall response has been flattened out and smoothed significantly, fitting within a 12dB window from 5-120Hz, where it was about an 18.5dB window before with much more overall variation.
Obviously, you cannot simply assume a static gain of any amount over a substantial frequency range when the system is emitting from multiple points within a room or other confined space. Put simply, gain is not assured at all at any specific frequency. Nor is the maximum gain by doubling the system capped at 6dB at any one frequency due to the complex interaction of all of the forces in play once you confine the system inside an enclosed vessel.
LinkBack URL
About LinkBacks
Re: Proyectos SI HT18 | Ideas y opiniones
Citar
Iniciado por Alex_Altea
. Como curiosidad tengo que admitir que he usado cascos porque el volumen es excesivo. También hay que tener en cuenta que la respuesta es plana y en punto de escucha en el centro de la habitación, supongo que en la pared trasera o cerca podría ser bastante mayor. La habitación tiene 20m2
