That's right, exactly. Even a large basshorn is usually acoustically small. That's why they tend to have rising response, because the radiation pattern narrows as frequency rises.
But if so optimized, when used in groups, the horn has enough mouth area to support even the lowest bass frequencies in the passband. The horn becomes uncompromised, and offers full efficiency through the entire passband. It will not change directivity due to mouth dimensions.
I designed the 12Pi hornsub to provide smooth response even when used alone or in pairs, but it is optimized to be used in groups. It has about 6dB rising response when used alone, and not much ripple. So it can be used alone or in pairs and it sounds good, smooth and very powerful. However, it is optimized to be used in groups of four or more. Lay two of them on their sides, mouths together and put two more on top of that. The mouth is now a large 90" x 60" (2.3m x 1.5m) rectangle, mirrored by the mouth reflection to 90" x 120" (2.3m x 3m). This is a full size, uncompromised basshorn.
Absolutely. That's the point. There are so many different subwoofer configurations that the acoustic center is all over the map. Most are well behind the mouth, basshorns have acoustic centers that are well behind the cabinet because their internal path lengths are so long. But some designs probably have acoustic centers much closer to the mouth. Some might even be in front of the mouth. And the acoustic center sometimes changes as frequency changes. So the path-length/acoustic-center ambiguity makes distant measurements pretty important for accurate comparisons.
I agree... time to move on.
Art has questioned my 10meter measurements in the past, and I've just let it go...
Now, if you really want a debate that adds up to serious db's... directivity in a single cabinet is a great debate, aka... a TH has it.
Jim,
The questions I have had regarding your measurements have always been for clarification of the way the measurements were taken. I have not questioned the measurements themselves.
For instance, in the case of a high power low frequency single sine wave test below Fb or Fc using a dB meter there would be a question of whether the second or third harmonic is louder than the fundamental tone (as is often the case).
If you put a 28.3 volt 40 Hz tone in to an 8 ohm speaker, and the dB meter reads 98 dB at 10 meters, one may think the sensitivity of the speaker is 98 dB one watt one meter, while in reality the dB meter is reading a 80 Hz harmonic that is 6 dB louder than the fundamental, so the sensitivity is actually 92 dB.
The dB meter can be 100% accurate, but the measurement result is ambiguous.
You have usually answered my questions.
You never did tell me the drive level of the SS15 that showed over 100% distortion, but that thread got cut and pasted with several others, it became difficult to respond to anything, as the post #s all changed 🙁.
Measuring polar response of a cabinet at a distance approximately equal to the cabinet dimensions is very problematic, near field diffraction patterns would not necessarily correlate with far field directivity.
With the data set you presented, there are several questions that need to be answered to see if your conclusions are meaningful.
1) The upright vs side measurements are virtually identical, deviations may be from wind, what were wind conditions?
2) What was the position of the microphone in the upright vs side measurements , centered on cabinet face, or centered on the mouth ?
3) For the directivity test, was the cabinet rotated, if so, what was the point of rotation in the “upright” and “side” tests?
4) If the microphone was moved rather than the cabinet rotated, what are the distances from each side measured with a flexible tape measure from the center of the mouth, around the cabinet to the microphone ?
5) I thought you had done some 10 meter off axis tests, how do they compare to the one meter tests?
Art Welter
Ah, using them in groups is something I didn't consider, I was looking at the mouth size of a single unit on it's own, and without the doubling of effective mouth area from the ground reflection.
So if we consider a single unit by itself, as well as directivity increasing with frequency resulting in an upwards sloping on axis response, do you think it would be fair to say that a single unit would see a significant shift in the apparent source location of inverse square law fall off with frequency due to the change in directivity ?
In other words at 30Hz where a single unit is not directional at all, will the apparent source (of inverse square law fall-off) be right near the mouth, since most of the wavefront expansion is occurring there, but with the apparent source moving further back towards the driver as you go higher in frequency towards 150Hz ?
The higher in frequency you go, the bigger the mouth becomes in wavelengths, and the less the signal will "spread out" after it leaves the mouth, until at some point (perhaps above the operating range of this sub) it won't spread out significantly beyond the angle that the horn initially bound it to.
Another way to look at it would be to say that the location of the apparent source for inverse square fall off is affected by the proportion of the directivity shift occurring when the wavefront exits the mouth and expands. (or doesn't, depending on frequency)
Now if we stack a whole bunch of your cabinets together as you suggest, along with the ground reflection the mouth area increases dramatically, making the horn un-compromised, and presumably increasing the directivity at the low end so it's similar to the high end.
Would you then say that there would be less shift in the apparent location at lower frequencies with the stack of units than one unit alone, due to less shift in directivity ? In other words, with the stack, even at low frequencies the apparent source of inverse square law fall off stays much further back behind the mouth. Eg at the bottom end ~30Hz the apparent source of a stack of 12pi's would be further back than a single 12pi on it's own ?
So not only does frequency affect it, but whether you're testing one unit on it's own, or a stack also affects it because of the directivity change. Yet another source of error when measuring close. Is it even valid to measure a compromised horn in it's compromised frequency range up close, when it's apparent source location will be different than when used as a stack ?
I wonder how many of the measurements presented in this thread have been of a single compromised horn within the frequency range at which it's compromised - which would put the apparent centre near the mouth, but not necessarily be representative of how the unit would perform as a stack and/or at a distance.
Again, the only unambiguous way to sidestep all these potential measurement issues with a large bass horn seems to be measuring relatively far away...
Last edited:
Wayne,
The point of my and Phil’s data presented in post #237 specifically was to test the hypothesis that horn path length makes a difference in the point where the inverse square law starts.
We found that the path length difference made no difference in the rate of fall off, both front loaded and bass horn cabinets conformed to the inverse square law from 1 to 32 meters.
This proves that the path-length/acoustic-center ambiguity makes the distance of measurements irrelevant for accurate comparisons.
Unlike path-length/acoustic-center location, our test results are unambiguous 🙂.
Art Welter
I'm pretty sure I answered these before in the singlesheet thread.... and I showed you distortion charts in another thread at 28v and even up to 46v that were really low in %. The over 100% distortion chart was a quick measurement at maximum of my amp... (an insanity test) and I didn't let it go long enough to get a good volt reading.
1. No wind -- At the end of a 100' cord from my house, that is in the middle of a 20 acre field. cabinet was facing north, and house was west of cabinet location. I don't feel the upright vs side are virtually identically at all. Check the on axis readings as well as the 90 degree off axis drops.
2. Position of mic was in my hand at the end of a 10' meter string.
3. not sure I understand... cabinet was set upright, and on it's side...
4. 10m string
5. I always do 10meter tests unless someone really wanted a 1 meter because the 10meter was in question.
I know this is off topic in regards to wayne's 12pi. To get back on topic, it would be interesting if off axis directivity tests have been done on the 12pi? Either in single, double or quads. I'm very sure directivity increases as cabinet count increases.
It's just one of those curiosity questions, and is kind of related to the current topic of 10m measurements, as any subwoofer cabinet with directivity has an advantage at 10m vs 1m.
hope that clarifies.
Hmm, how do I put this to you?
First, if the acoustic center were behind the mouth, then the falloff from the mouth would simply continue that trend. But the output one meter from the mouth would be more than one meter from the acoustic center, so the devices you tested with the most distant acoustic centers would be at a disadvantage. Ironically, those would actually be the ones that were probably most efficient.
Second, I hate to bring it up again but I question the validity of your tests. Some of it doesn't pass the sniff test to me, and you yourself even commented that you were unsure of it.
From PSW Sound Reinforcement Forums: LAB Subwoofer => JBL SRX 728 vs. Ported Lab 2x12”
I have attempted to calibrate my Smaart readings, but am not sure if they are absolutly on.
I have not attempted to calibrate them with the RTA 420 microphone I was using in the above test.
I have written several times that I can’t get my 3 meters to agree at more than one frequency.
So my confidence in your data is low. Sorry to say so, but there it is.
Yes, I would expect so. One way of looking at it is the directivity shift, another is the acoustic center shift. Both are related, sort of flip sides of the same coin in the case of an acoustically small horn.
However, difraction creates an apparent source, which is not the same thing as the acoustic center. The apparent source is delayed by distance from the acoustic center, and it also is attenuated by the inverse-square law from the expansion of the wavefront from the piston forward. It's like a secondary re-radiation, that has another wavefront. This secondary wavefront reacts with the source wavefront, creating ripples in response. The interaction is somewhat complex.
Good questions, Simon.
In a single unit, the inverse square law will proceed as normal, from 1 meter to whatever distance can be measured above the noise floor.
Delay should be set for the best phase match with the top cabinets, which will be close to the path length plus filter time offset.
When subs are in a large array, the problem becomes what audience position to align tops and subs to.
If one decides to align at a far distance (say 50 meters for a 3 meter wide array with top cabinets directly above them), the delay time will be the same as for a single, and the phase response of the array will look like it does for a single cabinet.
If one decided to align a row of subs that covered the front of a stage to flown L/R top cabinets for the expensive seats up front, the alignment would be wrong for the cheap seats.
A large array will change the point of inverse square attenuation. There will be a near field interference zone where upper sub frequencies are canceled to some degree, while low frequencies within 1/4 wavelength or so are added. At any given distance and off axis angle, the frequency response of the array changes.
Measurement of arrays are best done at multiple locations in the audience area, as different distance and angles will all yield very different results.
Trying to get even coverage at various frequencies with large sub arrays is quite challenging, and there are dozens of different approaches.
Art Welter
Wayne,
Testing the test gear is an important part of testing 😀.
Calibration is somewhat illusory, a microphone test calibration with a 1K tone insures the system is accurate only at 1K, assuming the calibration unit is well fitted and precisely made for the particular microphone, and the calibrator itself has not changed over time.
Those are a lot of assumptions 🙄.
When one compares published specifications of cabinets with one’s own tests, if they agree within a couple dB or so, that is a good thing 😉.
Both your and my test of the JBL SRX 728 agree within that range, so we can be confident our test gear is giving reasonably accurate results.
I use an inexpensive test microphone (RTA 420) for much of my outdoor testing to save my B&K 4004 from harsh conditions, and I know they have slightly different frequency response. Even the two B&K 4004 have slightly different frequency response, and they are pretty much at the top of the heap as far as test microphones and are in pristine condition.
That said, my comparative Smaart tests of two front loaded subs at two meters, and my honesty questioning the absolute accuracy of my machine calibration has absolutely nothing to do with Phil’s results using a standard dB meter to test the hypothesis that horn path length makes a difference in the point where the inverse square law starts.
The more efficient horn subs showed no near field disadvantage as you hypothesize, all the result numbers are within 1 dB between the different cabinet types.
We found that the path length difference made virtually no difference in the rate of fall off, both front loaded and bass horn cabinets conformed to the inverse square law from 1 to 32 meters.
This proves that the path-length/acoustic-center ambiguity makes the distance of measurements irrelevant for accurate comparisons in single units of the size range of cabinets discussed in this thread.
Any condition that would affect his readings would have affected both types of cabinets equally.
If you find some problem with the methodology of our 1 to 32 meter tests, please point them out.
Art Welter
Attachments
See, there's the rub. You would have us believe that a Radio Shack meter and other uncalibrated instrumentation proves false the "hypothesis" that horn path length makes a difference in the point where the inverse square law starts. Yet this is basic horn theory. The path length absolutely influences the position of the acoustic center. It is not a hypothesis, it is well accepted fact, described in all the literature.
There are some relatively new papers on improved techniques to accurately measure the acoustic center, but the concept of the acoustic center is nothing new. It is somewhere in the throat, generally close to the apex or diaphragm. In the case of folded horns, that would be a point behind the horn roughly the same distance as the path length to the acoustic center. This is the position from which the sound pressure begins to vary inversely as the distance. There is area expansion from the throat to the mouth, and there is attenuation along the path as a result. This attenuation is SPL varying inversely as the distance, the very feature we're looking for.
So, no, I would not conclude you have proven the well established principle of the acoustic center to be a myth. I also would not conclude that you have proven path length has no influence on the position of the acoustic center.
The brand of dB meter we used is irrelevant, as calibration is completely unnecessary for the comparative tests we performed.
The fact that the measurements conform closely to the inverse square law is evidence the dB meter is working properly, but if you find some problem with the methodology of our 1 to 32 meter tests, please point them out.
If the meter was not correct in absolute dB SPL terms, that is it read a dB or two high or low, it would simply read high or low for both subs.
You have no data to prove your hypothesis, and give no valid reason as to why you don’t accept Phil’s results proving that the path length difference made virtually no difference in the rate of fall off, both front loaded and bass horn cabinets conformed to the inverse square law from 1 to 32 meters.
The data proves that the path-length is irrelevant for accurate comparisons in single units of the size range of cabinets discussed in this thread.
Any condition that would affect our readings would have affected both types of cabinets equally.
If you have any actual test data supporting your hypothesis, please post it.
Art Welter
Attachments
Please do not post on this subject again. People are tired of hearing about it. If you believe the acoustic center is always at the mouth, so be it. I don't care what you believe.
Your continuing to say "I have no valid reason to accept what you say" is childish. I have told you several times now the reasons I don't agree with you. Your position doesn't square with accepted facts, the laws of acoustics that are in basic texts.
It's another flat-earth argument and I'm tired of having that kind of discussion with you.
Originally Posted by weltersys
You have no data to prove your hypothesis, and give no valid reason as to why you don’t accept Phil’s results proving that the path length difference made virtually no difference in the rate of fall off, both front loaded and bass horn cabinets conformed to the inverse square law from 1 to 32 meters.
The data proves that the path-length is irrelevant for accurate comparisons in single units of the size range of cabinets discussed in this thread.
I never wrote “the acoustic center is always at the mouth”.
I post data you have no valid argument against, you post no data supporting your theory and call me childish, that is a good one 😀.
I have made mistakes and admitted them, but the test you seem to dismiss simply because it disagrees with one of your concepts is not flawed.
Since you can’t argue the facts, your "discussion" has largely been pot shots, name calling and vague references to textbooks .
Please don’t post again until you can show actual tests confirming your hypothesis 🙂.
Art Welter
You have no data to prove your hypothesis, and give no valid reason as to why you don’t accept Phil’s results proving that the path length difference made virtually no difference in the rate of fall off, both front loaded and bass horn cabinets conformed to the inverse square law from 1 to 32 meters.
The data proves that the path-length is irrelevant for accurate comparisons in single units of the size range of cabinets discussed in this thread.
Again, you show that you are not reading what I wrote.
I never wrote “the acoustic center is always at the mouth”.
I post data you have no valid argument against, you post no data supporting your theory and call me childish, that is a good one 😀.
I have made mistakes and admitted them, but the test you seem to dismiss simply because it disagrees with one of your concepts is not flawed.
Since you can’t argue the facts, your "discussion" has largely been pot shots, name calling and vague references to textbooks .
Please don’t post again until you can show actual tests confirming your hypothesis 🙂.
Art Welter
You did. That's your basic premise. You can start backpeddling and playing with semantics, but your argument boils down to the idea that the mouth can be considered to be the acoustic center even at very close distances.
This isn't "my theory". This is what I learned from reading basic texts. I just applied what I learned.
I'll ask you one more time not to post about acoustic centers any further. This horse is so dead it's dust.
Quote:
Originally Posted by weltersys
I never wrote “the acoustic center is always at the mouth”.
Wayne,
"Playing with semantics"?
Phil's test did not measure the acoustic center of the cabinet.
The test simply proves that cabinets in the size range of what have been discussed in this thread, with differing acoustic origin distance, with frontal area around one meter or less, when tested in one or two units, conform to the inverse square law as measured from 1 to 32 meters from the cabinet front, not some distance emanating from within the cabinet.
I have read many acoustical texts over the years, perhaps many of the ones you have.
I don’t know of any text that shows results of a similar test, which is why I chose to do it in the first place.
Many individuals that actually make a living designing, selling, operating and providing consultation for professional sound systems were quite interested in the test results.
Some were a bit surprised, others realized the test confirmed what they already knew.
Thus far, other than you, no one has dismissed the test results.
The test is so easy to do, you could have done it 10 times in the time you have spent trying to dismiss it.
I welcome you to try it.
Art Welter
Originally Posted by weltersys
I never wrote “the acoustic center is always at the mouth”.
Wayne,
"Playing with semantics"?
Phil's test did not measure the acoustic center of the cabinet.
The test simply proves that cabinets in the size range of what have been discussed in this thread, with differing acoustic origin distance, with frontal area around one meter or less, when tested in one or two units, conform to the inverse square law as measured from 1 to 32 meters from the cabinet front, not some distance emanating from within the cabinet.
I have read many acoustical texts over the years, perhaps many of the ones you have.
I don’t know of any text that shows results of a similar test, which is why I chose to do it in the first place.
Many individuals that actually make a living designing, selling, operating and providing consultation for professional sound systems were quite interested in the test results.
Some were a bit surprised, others realized the test confirmed what they already knew.
Thus far, other than you, no one has dismissed the test results.
The test is so easy to do, you could have done it 10 times in the time you have spent trying to dismiss it.
I welcome you to try it.
Art Welter
OK, I'll give this one more try. I wasn't going to respond, but this just begs for closure.
How can you say this, but then say "I never wrote the acoustic center is always at the mouth”?
If you claim the inverse-square falloff is uniform from the mouth, then you are effectively saying the mouth is the acoustic center.
So which is it? Where's the source location, according to you?
How can you say this, but then say "I never wrote the acoustic center is always at the mouth”?
If you claim the inverse-square falloff is uniform from the mouth, then you are effectively saying the mouth is the acoustic center.
So which is it? Where's the source location, according to you?
Last edited:
Do y'all understand that you're talking about two different things. Best i can figure, one of you is talking about time delay imposed by the path length and the other is talking about gross radiation. Inverse square starts just outside the box where the horns terminus occurs (about 1/2 the diagonal of the opening away from the box), time delay can be described as an illusionary "acoustic center" occuring at some point behind the box.
unless i'm confused...
unless i'm confused...
Last edited:
I guess it's a common misconception that the inverse-square law starts at the mouth or just outside. This is not true. The acoustic center is in the throat, both with respect to delay and in terms of radiated power.
A horn creates an impedance transformation by matching the wavefront expansion with its physical area expansion. The cross-section of the horn forms a pie-slice that is the radiating angle. So it is already developing an expanding wavefront before it even leaves the horn.
At the horn throat, pressure is highest, and it is gradually reduced as the wavefront travels down the horn. In fact, in a conical horn, the pressure attenuation down the horn is precisely the same as a spherical wave in free-space, exactly following the inverse-square law. So you can easily see that the acoustic center of a traditional horn is pretty much where the radiating diaphragm is. It actually moves with respect to frequency, but it is rarely at the mouth. If the horn is folded, the acoustic center may actually be behind the cabinet some distance.
A horn creates an impedance transformation by matching the wavefront expansion with its physical area expansion. The cross-section of the horn forms a pie-slice that is the radiating angle. So it is already developing an expanding wavefront before it even leaves the horn.
At the horn throat, pressure is highest, and it is gradually reduced as the wavefront travels down the horn. In fact, in a conical horn, the pressure attenuation down the horn is precisely the same as a spherical wave in free-space, exactly following the inverse-square law. So you can easily see that the acoustic center of a traditional horn is pretty much where the radiating diaphragm is. It actually moves with respect to frequency, but it is rarely at the mouth. If the horn is folded, the acoustic center may actually be behind the cabinet some distance.
Last edited:
Definition for acoustic origin:
"The point in time at which the signal originates."
Definition for acoustic center:
The point in space of the origin of sound. For a sound emitting transducer (e.g., a loudspeaker), the point from which the spherical waves appear to diverge as observed at remote points.
One can easily measure the the point in time at which the signal originates using dual FFT systems like Smaart.
I’m familiar with procedures to follow for time and phase alignment, though I’m not sure how one would measure the the “point in space of the origin of sound” for a horn sub, as it is obvious to me that the acoustic origin and the point at which the sound drop off according to the inverse square law are two different points in space.
The interesting thing about my test is it shows that with differing acoustic origin distance, cabinets with frontal area around one meter or less, when tested in one or two units, conform to the inverse square law as measured from 1 to 32 meters from the cabinet front.
I am sure that the same would be true for a single 12Pi.
If the cabinet radiating area or array size gets to be significant relative to the measurement distance, there will of course be a deviation from the inverse square law, as the inverse square law assumes a single point source, which no loudspeaker actually is.
I can understand your reticence to accept information that seems to be contradictory in some regards to what we know , but I subscribe to the Synaudcon view, measure reality, then figure out which theory applies .
It does not seem to me my measurements have broken any laws of physics, or conflicted with any theories, though for whatever reason, no one seems to have bothered to specifically test what we did.
I’d really like you or anyone else to repeat the test, and see if there are any subs that don’t conform with the inverse square law as measured from 1 to 32 meters from the cabinet front.
Art Welter
You don’t seem confused at all, you are absolutely correct, but Wayne disagrees with what you, me, Josh Ricci, and many others agree upon.
Hence my repeatedly bringing up a test that proves what you just wrote.
If you have any "big name source" that illustrates what you wrote, that would be helpful for “closure” on the subject.
- Status
- Not open for further replies.