Seperate drivers for LF and ULF or one driver? - Page 7

"Point me to a relevant chapter."

i recall it being referred to a few different ways in that source, but section 2.2.3. on compactness and 4.1 on plane waves (read through to the first footnote, about one page) discuss how below the critical frequency, air, which is normally a compressible fluid, behaves as an incompressible fluid and pressure is pretty much uniformly increasing and decreasing everywhere at the same time.

given the extent and sophistication of the treatise, i am inclined to take them at their word.

http://www.win.tue.nl/~sjoerdr/papers/boek.pdf

I think it'd be doable, besides, the primary focus of Harman's "speaker swapper" was to facilitate quick/timely comparisons of full range loudspeakers in essentially the same physical position. The inherent wavelengths involved for subwoofer comparo diminish the magnitude of any positional differences that would be encountered.

----

It's at their Northridge campus. They would never allow any individuals outside the Harman group utilize their listening facility for their own use. I believe outsiders have likely participated in testing, for more and more data acquisition, studies, etc. But you know, it would be very cool to check it out and allow someone to demo their presentation to a group of enthusiasts. This is an aspect of AVS that's missing, IMO. Field trips, excursions, whatever, but collectively going to a facility such as this, IMAX backstage, behind the scenes, driver testing/mfr,ing facility, I think it would be so fun to get like minded a group together for such an event.

---

Bosso, what is this? What parameters changed?

---



---

+1

Absolutely, the methodology must be well vetted in advance, and executed precisely,....otherwise, what's the point...

Long overdue indeed. Please, I'm quite late to the party in this specific thread and do not intend to re-engineer the wheel here. However there are a handful of ongoing points of interest, whereby a collective effort to make some progress toward resolving some of these issues would be very worthwhile.



Thanks

didn't jbl close northridge or was that just manufacturing there?

"The inherent wavelengths involved for subwoofer comparo diminish the magnitude of any positional differences that would be encountered."

the test would be interesting. i still assert that some sort of "sub swapper" would be required, as the sound at the listening position indoors can change greatly even a couple of feet apart because of room effects, whether that is from having one sub and two listening positions separated by a couple of feet or one listening position and two subs separated by a couple of feet, the amount of the difference can be similar. for a fair test, you'd need one position for the sub under test and one position for the listener evaluating. maybe the subs could just be put on moving dollies and rolled around--a manual sub swapper.
Edited by LTD02 - 8/12/12 at 10:57am

Sub 'A' is placed in the front right corner of the room.
Sub 'B' is placed in the front left corner of the room.
Microphone is at the LP and its position remains unchanged.
Sweep is played through sub 'A' sub only (RED TRACE)
Sweep is played through sub 'B' only (BLUE TRACE)
Sweep is played through both subs simultaneously (BLACK TRACE)


The pressure pot theory dictates that room gain below the modal region occurs because the drivers cone moving into the room makes the rooms volume smaller, thus increasing the ambient air (barometric) pressure of the room. There is no escaping the fact that when both subs are playing at the same time, there is an exact doubling of cone displacement of the rooms volume.

Therefore, the summed sub black trace measurement should show a perfectly uniform increase in dBSPL in the pressure pot region, if there were such a thing.

Instead, we see evidence that there is no such phenomenon. As well, we see that the law of reflection holds true across the entire sweep bandwidth.

I believe I'm the only person on earth to have performed this simple exercise with accurate measurement capability to 4 Hz. I did this measurement nearly 4 years ago, posted the results and commented that phase still matters, regardless of frequency vs the modal region.

When I open a door or a window, it changes the pattern of wave propagation at the LP in a positive way. Opening a window may just as well cause a destructive pattern of propagation at the LP, depending on where the window is, how big it is, etc. Changing the listening position or the subwoofer positions can also place the listener in a more constructive or destructive pattern.

This is all only relative to a measurement and not so much the listening experience because a) if the subwoofer system is correctly scaled to the room and desired listening levels it is presenting an accurate playback and those frequencies in discussion are inaudible and b) that slice of bandwidth that is below the modal region is almost exclusively sensed by other than hearing and mostly through tactile responses generated by the reaction of the room (or seat, or your body) to the pressure waves, so that it does not matter where you sit in relation to the propagation pattern.

The 'both' graph, the 'window open vs closed' graph, the 'door open vs closed' graphs, the 'mic at progressively farther from the sub' graphs and the 'room gain profile' graphs I've posted over the years are all strong evidence against the pressure pot theory and serious clues as to how to attain optimal ULF playback and they were always accompanied by accurate in-room FRs from the LP showing flat-to-3 Hz or 4 Hz results.

The response has been:

Your measurements are wrong.
It's the measurement rigs noise floor.
You're using the wrong measurement window and the result is all noise.
It must be you have a magic room.
It must be your magical subwoofers.
LOL, that's not how it works at all.
That's because when you have no topical reply, you change the discussion to something else and pour on reams of verbiage and data.

The point is that I get results. But, I could very well not get results (SEE THE BLACK TRACE) and post that my room has no gain below 20 Hz but that I don't miss it anyway or some such typical sentiment.

Obviously, I didn't do that.

i explained several times the open window or open door effect as helmholtz resonator effect, but you chose to ignore it.

in #181, i answered your question, but you chose to ignore it.

that is in part why the responses that you have posted are as they are.

Question on the single location sweeps. Can the unused subwoofers located in the room act like a bass trap when the subwoofers are swept only at one location as compared with the two locations?

As far as ambient air pressure (barometric) is concerned, how can ambient air pressure "increase" when an AC signal is being used (except when measured on a momentary basis)? There is a modulation (increase/decrease) of ambient air pressure, but that is similar to looking at ripple on a DC power supply.

"how can ambient air pressure "increase" when an AC signal is being used (except when measured on a momentary basis)?"

that is just on the positive side. there is an equal decrease on the negative side. you have it right. by "pressurize" i mean the whole room is experiencing roughly the same increase or decrease, not that it is always going up.

your question about the other sub acting as a bass trap is a good one and the effect depends if it is powered or not. an unpowered sub will be a much better bass trap, while a powered sub will "resist" attempts to perturb it and so its bass trap effect will be lower, much lower. i was thinking about the same for the subwoofer comparo...what to do with the dead sub. it seems that it should at least be powered and at best sealed up and/or removed from the arena.

If schedule allowed, I'd seriously consider attending as well.

JSS

how about a tcsounds lmsr 15 up against an ae speakers td18h, both in 3 cubic feet sealed?

that way you get a heavy moving mass, high qe, and high inductance motor against a low moving mass, low qe, and low inductance motor...where both are of high quality.

I tend to ignore irrelevant explanations. You answered nothing. You cited a paper with irrelevant information, which you tend to do with every argument. That, or cite subjective preferences, another weak stance for a debate. "Hey, Mikey likes it so it must be good".

Sound waves are exclusively longitudinal, never plane. Plane waves are irrelevant.

incompressible potential flow also has nothing whatever to do with room gain.

In a word, no.

If they did, the separate sub sweeps would show the decrease in output, not the summed sweep. The opposite occurred.

The premise for pressure pot is simple; the cone moving into the room changes the volume of the room, thus increasing the barometric pressure everywhere in the room. Seaton said it, Noah said it, LTD said it.

It follows that if you double the cone displacement moving into the room, there must be a uniform increase in pressure across the entire bandwidth below the modal region, regardless of microphone placement. There can be no out for this assumption if the premise is a correct one.

Instead of seeing the uniform increase, there was a significant decrease. To be clear, there isn't always a decrease. With phase, placement and seat location experimentation, the differences between separate sub results (which were always practically identical) and summed sub results vary accordingly. This is pretty concrete evidence against the pressure pot premise. I can't see a leg to stand on. I didn't when I observed the results, which is why I experimented further.

Danley brought up the 'asympathetic resonance' theory when Josh showed a ULF suck out in his in-room measurements. Virtually everyone jumped on that bandwagon and some actually began to suggest bracing his friggin' walls. As far as I know, there's no such thing as asympathetic resonance. When a resonance is excited by an external force there is always an increase at the resonant frequency and never a suck out, especially if its a wall with a Sd of some 150,000 cm^2.

IMO (and the evidence) the idle sub as a bass trap below the modal region idea is as unlikely a relevant phenomenon as the drivers cone changing the air pressure in a room.

What about the case of a ported subwoofer where the port was directed outside the room? Resonances can show up as more energy or a sponge of energy at their resonance.

"The premise for pressure pot is simple..."

if it were so simple, everybody would know how it works. the dynamic transition of air as a compressible fluid to a non-compressible fluid in which the speed of sound goes up dramatically below the critical frequency is far from simple.

"...the cone moving into the room changes the volume of the room, thus increasing the barometric pressure everywhere in the room. Seaton said it, Noah said it, LTD said it."

...and so did the authors (drs. rienstra and hirschberg) of the treatise on acoustics who are at one of the most prestigious technical universities in europe. one is a physics and math professor, the other is a physics and fluid/gas dynamics professor.

"When a resonance is excited by an external force there is always an increase at the resonant frequency and never a suck out..."

in room treatment, various forms of resonators are commonly used for the purpose of pulling energy out of the room.
Edited by LTD02 - 8/13/12 at 10:11am

Scott,

Do it

Hey... I'm down for this epic meet. Somebody needs to coordinate such a visit. I just won't be able to bring anything. I doubt one of my LLT's will be considered luggage .... or able to fit in an overhead compartment.

Hmmm. Maybe I can convince Edogg to take us out there with it in the back seat.

*strokes chin*

That's not exactly what I said. The answer is all of the above plus a few other characteristics as well. I am most interested in which ones should be the highest priority below 120Hz as far as their effect on the sound.


Right. No matter what type of system whether pro audio, commercial theater, car audio, HT, or whatever the sub bass system is almost always low pass filtered somewhere near or below 100Hz. The LFE channel for HT is rolled out at 120Hz so I like to consider that as a good stopping point. Even with a steep 24dB octave low pass as is commonly used it is worthwhile to look at the energy for the next octave past the LPF and the transition through the crossover region.

A low pass filter applied will filter the upper harmonics and frequencies of the signal being sent to the amplifier/speaker as that is a function of its very design, but it will not reduce the harmonic distortion generated by the speaker system and or driver. An acoustic band pass will affect the harmonics and upper frequency noise generated by the driver and will also affect the upper frequency content present in the signal. However it will not decrease the load on the amplifier and speaker like the electronic filter will.


Super dumb question time. Are you sure that these 2 systems were wired in electrical phase with each other?





Ideally this would best be done outside but that may not be practical.

I believe it could work in room. It is not as difficult as thought. Yes the room will color the sound of each system but it should apply equally to each of them as long as they are placed exactly the same, the listener positions remain the same as well as the equipment. If you have seats A,B,C and D and we equalize the first sub to the best response across the seats that is possible, each seat will have a different response yes but as long as each sub is placed the same and equalized as closely as possible and each listener sticks to their same chair that won't matter because the response at each individual chair will stay matched. For this reason and simplicity I would much prefer to use a single sub placement. Mark it with masking tape on the floor. EQ the first system to the best response possible. Move it out and use a separate DCX channel to set-up #2 and match it as closely as possible in FR shape and level to #1. Wash rinse repeat. Then start the blind listening. All it would take is moving one sub out and the other in and changing the channel routings on the DCX. Perhaps a quick verification measurement just to be sure things are right. Each listener occupies the same seat the whole time and takes notes.

I would also want to see the same enclosure type and size used for each driver if not the same exact enclosure ideally. This would remove any effects of the enclosure itself acting differently on the room acoustics as well. I don't care about comparing a big vented cab to a small sealed one or a horn or completely disparate tunings. That is a different test. I want to hear the differences in driver design and that muddies the water far too much. We could do the alignment comparison later where we keep one single driver as a constant. I would want to use sealed enclosures for this test due to size and simplicity.

Another thing is that I want to hear the comparison with all drivers being operated well within their limits. If there are overload noises that is an obvious demerit and clue. I don't think anyone here operates their drivers at a point where they are losing composure on a regular basis. If we run into that, we get more subs and power until it isn't an issue. To that end this will require enough drivers to produce enough headroom for moderate to generous volumes but not so many that it unnecessarily complicates things (maybe 2 of each driver would be "enough"?). I would also like to use more than just 2 different drivers. I would like 4 of varying design and strengths if we were going to do this. Perhaps 2 drivers at the very polar opposites and 2 others somewhere in the middle.

If this meet is driveable from Rochester, I'm in. I'll bring a few "toys" to submit for listening...




Alot of black and white discussions and theories in this thread. Sound is analog therefore full of infinite shades of grey. Black and white is far too digital to be 100% right.

if the room has resonances and most rooms do, might it get difficult to distinguish speaker damping from room damping?

Ok what I have and is usable for testing of this nature:

Subs:

-two RE XXX18's sealed in 7 cuft.
-two jbl 4648's which we can seal the ports with pillows if needed
-two LLT's with Tempest x-2's tuned to around 11-12hz

Amps:

-LG 14K
-epx4000
-xpa-5 (just in the event it may be needed)

EQ gear:

- DCX2496
-berry mic 220
-sms-1
-denon 4311 w/ xt32
-REW on my laptop
-OmniMic
- mic boom

Additional:

-plenty of cables, speaker wire (12ga, 10ga.) etc.
-easy access to rear of amp rack
-no kids, wife, or neighbors close enough to piss off
-possibility of backyard use for outdoor measurements with removal of racks and such but it's not the BEST spot.
-couple of guest rooms and plenty of floor space for those that need a spot.

I think this should limited on the # of toys being tested, to keep it focused on the objective at hand. I am more than willing to offer my spot assuming you guys think it is a good space. check my sig for some pics, it HAS changed a good bit since then though, but the walls are still in the same spot...

You're showing a -6dB or so result, are you claiming acoustic destructive interference?

What happens in the pressure region to quarter wave cancellation effects? At 5hz the quarter wave point for any destructive interference is ~56 feet. Empirical evidence typically beats theoretical every time. Your measurement rig and sub system possess adequate resolving power into the range of interest. The measured evidence you present trumps theory,....but you've not convinced me of causality. I mean, say your evidence is solid, but what is it actually representing? Your position is it's acoustic. Ok, how? How are you going to get quarter wave interference at 5hz in your room? Maybe, I don't know, I don't see that.

To me there appears to be a component that's not purely acoustic in nature, because I've no idea how one could get the destructive interference.

I recall having briefly discussed such a similar resonant suck-out with both Jordan (K-Dub), and Josh. I've seen it and measured it in my room. Now, I never placed the suck-out / sympathetic resonance scenario under the scrutiny it likely deserves, but I felt it had some merit. Little else came to mind regarding such intermittent losses. Intermittent, because depending on the scoop of data collected, you could either catch it or not. In my notes I added there appeared to be an additional component as a function of time and drive level. Yeah I know,...but it's what I observed and noted.

I dug this one up. This particular measurement, naked-no EQ, and wanting to measure the effect of my seated position within the chair. Leaning right-blue trace, leaning left-red trace. That's the effect of a lateral head (mic) change of only about two feet within my over-stuffed easy chair. The upper bass nulls are so sharply defined because the left wall is un-treated brick fireplace, right side untreated drywall, at the time of this measurement (maybe a year ago or so). Shortly thereafter I lessened this lateral positioning issue with treatment.

no EQ LP leaning right blue, leaning left red.bmp 1811k .bmp file


Actually, the audibility of those nulls wasn't as bad as one may think. I changed the seat front to back position also, which helped as well. I've been experimenting for over a year, collecting tons of interesting data. It's been on hold due to illness.


I can elicit a 15dB hit at 12hz or so. I intended to brace the ceiling span (rafter 2x10, secured to truss top) to experiment and see if I can damp the null. I don't like the idea of a permanent brace there, as drywall ceiling could crack from redistributed roof loading. I've just always wanted to temp damp it and see what happens.

Whatever is going on, it's quite interesting.

"Danley brought up the 'asympathetic resonance' theory when Josh showed a ULF suck out in his in-room measurements. Virtually everyone jumped on that bandwagon and some actually began to suggest bracing his friggin' walls. As far as I know, there's no such thing as asympathetic resonance."

there is a space in between the "a" and the "sympathetic". a sympathetic resonance occurs when you have one object resonating at a frequency near another object with a natural resonance at the same frequency--for intuition of the term if a person near you feels bad, you will begin to feel bad, we call that sympathy...it's just one responding to another. the classic example is if you take two of the same design tuning forks, strike one such that it vibrates and then hold it near the second one, the second one will begin to vibrate almost as much as the first one.

if the suckout in josh's room around 15hz is related to the natural resonance of one of his boundaries (wall, floor, ceiling), the sympathetic resonance could cause the suckout just as danley mentioned, as the energy is going to vibrating the boundary and is lost as heat instead just vibrating the air.

foh, just a minor bit. if you have the option to save as a .png image, it can chop the size of the file down, sometimes greatly (in this case by 98%). .png format is lossless compression iirc.

that is what i was talking about with respect to the sub test. just a couple of feet can lead to a vastly different impression.


Edited by LTD02 - 8/13/12 at 9:50pm

I don't recall seeing this disclaimer in any of your "folks say the pro sound driver sounds more betterer" comments.

I don't know what you're referring to ("the case...", as though it's something I should be aware of) but I'm pretty sure it has nothing to do with the pressure pot theory. Can you show me a case where a sympathetic resonance causes a suck out during a sine sweep at 10 Hz in a room?


Polarity was correct. If I reverse the polarity of one of the subs, there is a huge drop across the bandwidth from cross down.

I said nothing about what caused the drop. I only said that if the pressure pot theory were valid it would not have happened.

And, don't take my measurements as evidence, just look at the vast majority of posted in-room measurements.

I prefer that the proponents of this pressure pot theory explain the results.

Notice how adamantly they predict a complete erasure of the 2nd order roll off based solely on room dimensions and when that doesn't materialize in actual posted measurements the suck out is explained away as the rooms dimensions causing absorptive sympathetic resonances. At least that's the only possible reason given to date (and a terribly weak one at that). And you're supposed to just accept that and walk away with the "Drat! My room is bad for the pressure pot!"

Sorry, but that all sounds far too convenient to me and it doesn't explain how I get results. My walls, floors and ceilings are built using exactly the same methods and materials and are the same size as any 3500 cubes living space in most every house in the US (National Building Code), so why is my room resonating any differently at 5 Hz than anyone else's?

Yes, like the suggestions that Josh brace his wall, it's something that never seems to actually happen in the end, but the problem has its answer nonetheless.

Your 'hit' at 12 Hz is undoubtedly below the noise floor of the measurement system. For valid experimentation at those frequencies, unfortunately, you have you crank the levels a bit or you'll get wildly varying and inaccurate results.


Yes, you've run to Google as usual. And can you post the measurement of these sympathetic tuning forks to show the suck out at resonance?


Baloney.

If one of his boundaries has a natural resonance at 10 Hz, it will also have resonance at octaves of the frequency. This makes it quite simple to investigate whether or not your sub is feeling bad because your wall has the flu.

Typically I run such through a image optimizer conversion, so yeah, gotta tighten up


That's correct. As I suggested,... a facility for such a GTG should minimize the room's relative significance and facilitate ease of load in/load out. If this is to be done, do it right. The room, methodology, gear, etc, ideally should be beyond reproach,...that should be the goal. However, unfortunately there will likely be compromises encountered.

In the case of what I illustrated in that particular graph I submitted, I suspected such a phenomenon was occurring, so schlepped out the measuring rig and correlated what I was perceiving to an actual measurement. Again, I want to make clear that the bottom octave difference isn't positional, it's related to the slice of time sampled.

Thanks

"Yes, you've run to Google as usual."

huh? the tuning fork experiment was from back in 6th grade.

"Baloney.

If one of his boundaries has a natural resonance at 10 Hz, it will also have resonance at octaves of the frequency. This makes it quite simple to investigate whether or not your sub is feeling bad because your wall has the flu."

the bbc has a good paper on this, but i'm getting tired of pointing you to resources.
Edited by LTD02 - 8/14/12 at 10:20am

---

What do you think caused the drop?

Few individuals, especially of the "audio enthusiasts w/measurement gear" crowd, have devoted as much time and energy into the studying of the bottom octaves in the home listening environment as yourself. Admittedly, you have shared with us the technical merit of your gear and your prowess. Others have devoted a significant amount of time to audio, even the bottom octaves, but perhaps not as primarily focused on the LF/ULF as yourself. All that said, and in your best guess, what could cause such a drop?


Thanks

Then don't. I'll make it through somehow.

My old room had a suckout at about 12.5Hz. The room was about 4100 cu ft with 10 ft ceilings and was 4 walls of regular wood frame construction, wood frame ceiling with small attic over top and wood floor over shallow crawlspace. One doorway. One window. This room was extremely lossy and "live". The suckout was always there. I don't believe I ever took a measurement in there where it was not present. The mic position, sub position and the amount or placement of multiple subs made no difference at all to this suckout. My best guess is that the floor resonated at this frequency and it would absorb the bass output. The floor would vibrate violently with sine tones at 12-13Hz.

I'm still open to other ideas to explain it.

Evidence...

Exactly.

Of course and this is one of my main issues with hearing damping or "speed" or "overhang" characteristics of drivers. The modification by the room acoustics will typically swamp any differences from the drivers themselves IMHO.

Question: How much of your critical listening during your lifetime has been spent outside of a room or vehicle interior? What about everyone else? 99% of serious Hifi or HT systems are used in a room of some sort including when these subjective characterizations and observations about sound quality differences between drivers are made.