Thanks for apology. Rare on the web!
I wouldn't raise this question on this thread but Bach On said his organ-building buddies mostly urged him to use a sealed box. But this forum tends to not like sealed boxes.
The usual comment on sealed boxes (at this forum) is that their sim says to build a box which is smaller than a ported box (that is good) but then the low frequency end is attenuated because the system resonance is driven up (that is very bad).
So my question is, "Why do sims recommend small sealed boxes which reduce the low end instead of large boxes?" That seems very odd and not the way to design a sealed box sub-woofer.
So far nobody has answered. Sad that nobody understands their sim well enough to answer.
Ben
Last edited:
I disagree, many understands well the sims. Also many tend to like sealed boxes - but only if they are adequate for the job called. High output 16 Hz (organ) loudspeaker could not be sealed type, much better solution with much higher 16 Hz SPL is a vented box or back-loaded horn.
Simulations (and the theory) of loudspeakers are very well correlated with a real-world measured boxes. Simulations at default outputs maximum flat, minimum F3 designs, with adequate box volume (which might seem low in volume, for someone). But you can input in the sims different box volumes, to see that bigger sealed box volume increase F3, but decrease F10 (for example). But, and this is VERY important, maximum SPL at 16 Hz from that bigger volume sealed box is well under maximum SPL from a good vented or back-loaded (or tapped) horn. By the way, the same came from every good or crappy simulation software I tried so far. Theory says the same, as well.
Big sealed box is not the right solution for a high output organ speaker at 16 Hz. Maximum SPL at 16 Hz with the bigger sealed box will be slightly higher than a maximum flat minimum F3 sealed design, but not enough! Sealed box with infinitely big volume always will output much lower maximum SPL at 16 Hz than a good vented box or back-loaded horn.
Simulations (and the theory) of loudspeakers are very well correlated with a real-world measured boxes. Simulations at default outputs maximum flat, minimum F3 designs, with adequate box volume (which might seem low in volume, for someone). But you can input in the sims different box volumes, to see that bigger sealed box volume increase F3, but decrease F10 (for example). But, and this is VERY important, maximum SPL at 16 Hz from that bigger volume sealed box is well under maximum SPL from a good vented or back-loaded (or tapped) horn. By the way, the same came from every good or crappy simulation software I tried so far. Theory says the same, as well.
Big sealed box is not the right solution for a high output organ speaker at 16 Hz. Maximum SPL at 16 Hz with the bigger sealed box will be slightly higher than a maximum flat minimum F3 sealed design, but not enough! Sealed box with infinitely big volume always will output much lower maximum SPL at 16 Hz than a good vented box or back-loaded horn.
Last edited:
Thanks, Oliver. I appreciate your answers to my rather geeky questions.
I've stated my goal of having a system capable of producing a nominal 100dB for the low sounds (plus or minus 15dB). But I won't know if the SPL needs to be higher or lower until I put it in the mix with the other ported boxes. It appears to me that the sims are primarily focused on a single box. It's the cumulative effect of three boxes operating together in the same range that is harder to predict.
The essence of what I THINK you are saying is that having amp power closer to the 600 watt Peak rating would be a plus for the T-TQWT cabinet with the HT18. Another Crown XLS1500 in bridged mode could supply this at 4 ohms.
I've used the Crown XLS series so far in our system. The variable crossover circuitry is a plus. We won't be using it on the first three XLS1000s powering the HC12s for the sounds coming from the manuals. But we are using it for all the bass boxes - including the Allen HR100 that will take care of the upper bass.
One poster suggested trying one of the INUKE amps from Beringher because of their circuitry to do shaping of the signals. I think some of the Beringer stuff is pretty good. But getting parts is nearly impossible. Crown has a better reputation in this area.
One of the guys over at the organforum also had reservations on the SI HT18. He also felt the T/S figures were suspect.
Here's what is posted at the SI site for the 2 ohm voice coils. They don't provide the figures for the 4 ohm voice coils:
HT18 Dual 2 ohm voice coils
Re 3.5 ohms
Fs 17 Hz
Qes 0.41
Qms 5.8
Qts 0.38
Le 3.2 mH
Sd 117841 mm^2
Vas 411.9 L
Bl 19.7
Mms 416 g
Cms 208.8
Sensitivity (1w/1m) 88.7 db
Xmax 22.5 mm
Do you feel these figures are suspect?
Bach On
Last edited:
Ben & Sonce,
I think sims are good tools when used correctly. But they can't predict matters of personal taste. That's why experience can be another useful tool.
You guys can battle it out. But that's my personal take on the matter. Your mileage may vary.
BO
I think sims are good tools when used correctly. But they can't predict matters of personal taste. That's why experience can be another useful tool.
You guys can battle it out. But that's my personal take on the matter. Your mileage may vary.
BO
The original mistake was people going to their favourite sim and their sim commanding them to build an enclosure that is flat to 16 Hz.... as Sonce explains above in sort of answering my question.
What people should have done is gone to their sim to figure out how to get a real big whomp at 16 Hz or thereabouts. Nothing wrong with sim math, if you ask it the right question.
And just to defend such sims further, issue is (1) the opaque nature of sim assumptions that people rarely re-examine and (2) the inability of people to understand how to translate their personal taste, as Bach On says, to physical parameters so as to read sims beneficially.
BTW, looks to me like a T-TQWT is just a high-priced stuffy name for a transmission line enclosure*. Like the old joke about how to pronounce "vase".
Ben
*lots of people like TLs a lot; many well-absorbent TLs are, in turn, nothing but leaky sealed boxes
What people should have done is gone to their sim to figure out how to get a real big whomp at 16 Hz or thereabouts. Nothing wrong with sim math, if you ask it the right question.
And just to defend such sims further, issue is (1) the opaque nature of sim assumptions that people rarely re-examine and (2) the inability of people to understand how to translate their personal taste, as Bach On says, to physical parameters so as to read sims beneficially.
BTW, looks to me like a T-TQWT is just a high-priced stuffy name for a transmission line enclosure*. Like the old joke about how to pronounce "vase".
Ben
*lots of people like TLs a lot; many well-absorbent TLs are, in turn, nothing but leaky sealed boxes
Last edited:
Hi Bach On,
Post #83: "...SI HT18...T/S parameters...Do you feel these figures are suspect?"
When I enter T/S parameters into Hornresp I use Hornresp's ability to recalculate the parameters. The manufacturer posted T/S parameters calculate spot on. That is all I know about them. The design margin for the vented boxes is much larger than for the T-TQWT, so I think you'll be just fine. As this new one might become your winter project you have a bit of time to measure the parameters of the driver you have on hand.
Here are two good references:
Measuring Loudspeaker Driver Parameters
The Subwoofer DIY Page - Measurements
You probably recognize the 2nd one as Brian Steele's "The Subwoofer DIY Page". Brian has done a great job describing how to do this.
Post #83: "...a nominal 100dB for the low sounds (plus or minus 15dB)..."
10dB means 10x the amplifier power.
According to Hornresp this latest attempt provides 115.66db @ 16Hz in a 2xPi environment @ 1m. There'll be a bit of a peak (117.91dB @ 18Hz) if the box is used w/o any fiber fill, again in 2xPi. Add 6dB for 1xPi which is probably what you are looking @ in the chamber. I think you'll be fine.
One of the neat features of this type of enclosures is the tuning capability through simple mechanical means, primarily fiber fill, but also duct length and mouth (terminus, port, opening) size.
My gut feel is that you have adequate bass w/ the two vented boxes, and that you will not need the full output of your Crown XLS1500.
Regards,
Post #83: "...SI HT18...T/S parameters...Do you feel these figures are suspect?"
When I enter T/S parameters into Hornresp I use Hornresp's ability to recalculate the parameters. The manufacturer posted T/S parameters calculate spot on. That is all I know about them. The design margin for the vented boxes is much larger than for the T-TQWT, so I think you'll be just fine. As this new one might become your winter project you have a bit of time to measure the parameters of the driver you have on hand.
Here are two good references:
Measuring Loudspeaker Driver Parameters
The Subwoofer DIY Page - Measurements
You probably recognize the 2nd one as Brian Steele's "The Subwoofer DIY Page". Brian has done a great job describing how to do this.
Post #83: "...a nominal 100dB for the low sounds (plus or minus 15dB)..."
10dB means 10x the amplifier power.
According to Hornresp this latest attempt provides 115.66db @ 16Hz in a 2xPi environment @ 1m. There'll be a bit of a peak (117.91dB @ 18Hz) if the box is used w/o any fiber fill, again in 2xPi. Add 6dB for 1xPi which is probably what you are looking @ in the chamber. I think you'll be fine.
One of the neat features of this type of enclosures is the tuning capability through simple mechanical means, primarily fiber fill, but also duct length and mouth (terminus, port, opening) size.
My gut feel is that you have adequate bass w/ the two vented boxes, and that you will not need the full output of your Crown XLS1500.
Regards,
The dominating factors in a subs performance are not as critical as many would suspect.
10 even 15% difference from the stated specs makes little to no difference.
The current problem with the 1/4 wavelength design is that it is not providing anything different than an equivalent sized ported enclosure.
I simmed that a few posts ago and put up the comparison.
10 even 15% difference from the stated specs makes little to no difference.
The current problem with the 1/4 wavelength design is that it is not providing anything different than an equivalent sized ported enclosure.
I simmed that a few posts ago and put up the comparison.
Ben,
A couple of the guys over at the organ forum like sealed boxes - and the bigger the better. They view sims with considerable suspicion and refer to them as voodoo. Ported boxes and horns are voodoo designs to them. Experience tells them big sealed boxes are winners. Everything else is suspect. There are some people who believe sims are the answer to everything. I view them as a potentially helpful tool for estimating what we might expect.
I think our organ is going to have a specialized need. And the printouts from the sims Oliver has posted seem to indicate the design he has posted would POSSIBLY fill that need. I really don't care what it is called. I can't even keep all the various kinds of horns straight in my head. But if it will fill the need we have, I'm willing to give it consideration.
I've tried sealing the ports on my ported bass boxes for comparison. There were still some leaks. So they were essentially leaky sealed boxes. They were unable to do 16 Hz. with anything approaching authority. So that solution didn't work for me. I'm not saying it can't be done. I just didn't obtain the results we needed. So I went back to the ported boxes.
Oliver's design seems to offer a reasonably flat frequency curve from 90 to 20 Hz. That sounds like a good thing. I don't see the advantage of a box that is virtually silent until we hit 16 Hz. when it suddenly pipes up, if that's what you were suggesting.
I'm trying to understand what you are suggesting might be a better solution? But I just don't yet get it.
I'm hopeful that snarky isn't my nature. And I'm not being hostile or belligerent. I'm asking for your concrete suggestions.
BO
A couple of the guys over at the organ forum like sealed boxes - and the bigger the better. They view sims with considerable suspicion and refer to them as voodoo. Ported boxes and horns are voodoo designs to them. Experience tells them big sealed boxes are winners. Everything else is suspect. There are some people who believe sims are the answer to everything. I view them as a potentially helpful tool for estimating what we might expect.
I think our organ is going to have a specialized need. And the printouts from the sims Oliver has posted seem to indicate the design he has posted would POSSIBLY fill that need. I really don't care what it is called. I can't even keep all the various kinds of horns straight in my head. But if it will fill the need we have, I'm willing to give it consideration.
I've tried sealing the ports on my ported bass boxes for comparison. There were still some leaks. So they were essentially leaky sealed boxes. They were unable to do 16 Hz. with anything approaching authority. So that solution didn't work for me. I'm not saying it can't be done. I just didn't obtain the results we needed. So I went back to the ported boxes.
Oliver's design seems to offer a reasonably flat frequency curve from 90 to 20 Hz. That sounds like a good thing. I don't see the advantage of a box that is virtually silent until we hit 16 Hz. when it suddenly pipes up, if that's what you were suggesting.
I'm trying to understand what you are suggesting might be a better solution? But I just don't yet get it.
I'm hopeful that snarky isn't my nature. And I'm not being hostile or belligerent. I'm asking for your concrete suggestions.
BO
Last edited:
Since I'm the one that originally recommended this driver on a price/performance basis I'll comment quickly.
Yes, they are suspect. All SI product specs are suspect. Especially the xmax spec. Often, SI will pull Klippel numbers on a prototype version and publish them, then make changes and have a different production version and never change the published specs. Sometimes he will even take the Klippel numbers, add a bunch of mm of xmax based on how he thinks the production version will measure and publish THAT as his xmax spec. I have proof in links if you want it.
Also, when quoting xmax he ONLY considers the Bl graph, ignoring all the other limits that would give a far lower xmax, like Le, suspension nonlinearity, etc. I also have links that show his SI 24 is limited by Le to 20 mm xmax and yet he published a spec of 36 mm for that driver.
The SI 18 D2 and D4 versions have very noticably different specs and will give different response in the same box, although I believe only the D2 version specs are published. And your guess is as good as mine as to how accurate these are. I certainly wouldn't trust them.
Here's a little gem from a recent conversation about the accuracy of his specs. It's from here - SI HT24 sealed test results - Page 6 - AVS Forum | Home Theater Discussions And Reviews
An externally hosted image should be here but it was not working when we last tested it.
Electrodynamic is SI, and he doesn't care in the least about providing accurate specs, as this image clearly shows.
I've been speaking out about SI's aggressive marketing and bad attitude and outrageously incorrect published specs for a couple of years now, he is a blight on the industry if you are interested in manufacturers publishing accurate specs.
Furthermore, this particular driver, along with all the other drivers that suffer the "high inductance" effect ABSOLUTELY WILL NOT perform as simulated unless you account for it in the sim, which no one ever does. (I put "high inductance" in quotes because as Kravchenko pointed out, it may not necessarily be the inductance that causes this effect in these massive coil/motor designs.) I have a method to simulate these drivers accurately but no one uses it. My sims are always accurate.
So why did I recommend this driver? Very simply, it performs. Third party measurements have shown that it's the value leader.
But it absolutely must be simulated properly or you will end up with a mystery response similar to the massive RE XXX ported build on data-bass that tb46 has mentioned a few times. It's measured response is not even close to what a simple sim would suggest. But I can sim it pretty accurately.
And while I'm here I might as well address this nonsense too.
I've told you a few dozen times already, Ben, sims don't recommend anything. The better simulators (Akabak, Hornresp, MJK's worksheets) are blank by default and it's up to the user to choose his own alignment.
The simulators for beginners and lazy people like WinISD can be used to give a mathematically manufactured result, but the user still has options of which alignment they would like to choose, and the user is free to alter these results after the choice of "default" alignment is selected.
How do you not know this? Could it be that you've never actually used a simulator and have NO IDEA what you are talking about?
Box size is not the only factor in sealed boxes either, as you seem to be hinting here. Not all drivers are going to produce a good response in a generic box size. A lot of other things must be considered too, like qts and the target response that will yield a flat in room response (or whatever other goal is desired). Any driver with qts above .5 or so has no chance of giving a flat in room response in IB (big sealed box) due to room gain. .7 qts in IB will give the lowest possible F3 with flat response anechoically but will sound boomy in room, just like max flat ported designs. Qts higher than .7 will give a humped up response at fs and will sound boomy in room in IB. This issue is nowhere near as simplistic as you are trying to present it.
A few weeks ago you were saying the lowest possible q was important, now you seem to be advocating bigger sealed boxes that will produce a higher q. You are contradicting yourself again, Ben.
As Sonce has mentioned, making the sealed box bigger might give a lower F3 but a higher F10, and no matter what you do, the sealed box will never equal the output of a ported box.
Not sure why you think we don't understand sims, Ben. Your inability to answer even the simplest of technical questions, and your queries like this one that are meant to stump the "sim lovers" show that you don't understand the issues yourself.
Last edited:
Data-Bass
Independent driver test:
From Josh at data-bass.com
Manufacture Stereo Integrity
Motor Type Overhung
Driver Size 18 "
Driver Weight 30 lbs
Coil Diameter 2.5 "
Coil Material CU
Coil Length 1.772 "
Manufacture's Xmax 23.5 mm
Aprox. Price $190 USD
Year 2012
TS Parameters (Measured by Data-Bass.com)
Qts 0.296
Qes 0.33
Qms 2.88
Fs 15 Hz
Res 3.5 Ω
Le 1khz 3.86 mH
Sd 0.115 m2
Vas 465 liters
Mms 446.6 grams
Cms 252 μm/N
BL 21.14 Tm
BL2/Res 127.7 N2/W
L/R Time Constant 1.1 milliseconds
SplSens 88.62 dB
Independent driver test:
From Josh at data-bass.com
Manufacture Stereo Integrity
Motor Type Overhung
Driver Size 18 "
Driver Weight 30 lbs
Coil Diameter 2.5 "
Coil Material CU
Coil Length 1.772 "
Manufacture's Xmax 23.5 mm
Aprox. Price $190 USD
Year 2012
TS Parameters (Measured by Data-Bass.com)
Qts 0.296
Qes 0.33
Qms 2.88
Fs 15 Hz
Res 3.5 Ω
Le 1khz 3.86 mH
Sd 0.115 m2
Vas 465 liters
Mms 446.6 grams
Cms 252 μm/N
BL 21.14 Tm
BL2/Res 127.7 N2/W
L/R Time Constant 1.1 milliseconds
SplSens 88.62 dB
Now keep in mind the variables that are hard to control.
The spider and the surround dominate the suspension characteristics. If they are consistently from the same supplier you can have some faith in the specs.
The spider and the surround dominate the suspension characteristics. If they are consistently from the same supplier you can have some faith in the specs.
Hi Bach On,
Post #88: "...A couple of the guys over at the organ forum like sealed boxes - and the bigger the better. They view sims with considerable suspicion and refer to them as voodoo...."
Anybody who has build, and compared the same driver in different enclosures will confirm that there is no voodoo to the differences between enclosure types, and that a properly designed vented box will have a higher output @ the bottom of the frequency range than a properly designed sealed box.
Assuming proper construction techniques, the biggest problems w/ vented boxes are: wrong tuning frequency, insufficient duct size and lack of an appropriate low cut (or high pass) filter. For a subwoofer that is supposed to cleanly reproduce 16Hz the low cuts build into the signal chain will probably already take care of the third of these points, but it still has to be addressed.
For anyone who wants to mess around w/ this I'll attach two Hornresp simulations, one of the SI HT18 in a 99999.99L sealed box (Hornresp Vint_max), and the other w/ the SI HT18 in the Bach On triangular corner box sealed. I'm also attaching a Hornresp graph showing both of those boxes @ Xmax. In practice (?) they would still need a low cut filter, as the the excursion is still increasing beyond Xmax below 16Hz.
Regards,
Post #88: "...A couple of the guys over at the organ forum like sealed boxes - and the bigger the better. They view sims with considerable suspicion and refer to them as voodoo...."
Anybody who has build, and compared the same driver in different enclosures will confirm that there is no voodoo to the differences between enclosure types, and that a properly designed vented box will have a higher output @ the bottom of the frequency range than a properly designed sealed box.
Assuming proper construction techniques, the biggest problems w/ vented boxes are: wrong tuning frequency, insufficient duct size and lack of an appropriate low cut (or high pass) filter. For a subwoofer that is supposed to cleanly reproduce 16Hz the low cuts build into the signal chain will probably already take care of the third of these points, but it still has to be addressed.
For anyone who wants to mess around w/ this I'll attach two Hornresp simulations, one of the SI HT18 in a 99999.99L sealed box (Hornresp Vint_max), and the other w/ the SI HT18 in the Bach On triangular corner box sealed. I'm also attaching a Hornresp graph showing both of those boxes @ Xmax. In practice (?) they would still need a low cut filter, as the the excursion is still increasing beyond Xmax below 16Hz.
Regards,
Attachments
Last edited:
That's the independent third party test I was referring to, but he tested a prototype unit, not a production unit so those specs may or may not be accurate compared to a production unit. As always, SI's published xmax spec (included here) is pure fantasy. This driver did beat the Dayton 18 HO (which has only 1/2 the published xmax as the SI driver) in torture testing but only by a bit, about 1.5 db. The Dayton also had better response, no big humped up "inductance effect" at the box's impedance peak, indicating that the Dayton will perform much closer to the way a simple sim would predict than the SI. The CEA 2010 max burst testing was even closer, 1 db or less difference across the 10 - 100 hz range, but that's not a reliable indicator or performance, it speaks more to distortion profile than long term max output.
An externally hosted image should be here but it was not working when we last tested it.
The SI driver is a better value, but it won't measure as a simple sim would indicate.
15 Hz driver resonance, as per Mark's post. Bingo!
But tb46's drawing from HR doesn't match my idea of an infinite baffle response of a driver with a 15 Hz resonance. Can you please explain that to me in simple language.
Just for fun (and I mean that), here's a mic trace from sitting position of a 1960 15-inch, 21 Hz resonance driver on a large open board. Is this nice? OK... I know Just a Guy loves to learn about old equipment.
Ben
Attachments
Last edited:
Ben,
In simple language, it appears your idea of an infinite baffle response of a driver with a 15 Hz resonance differs from the Hornresp simulation because your "idea" does not use the same math that Hornresp does.
Having found Hornresp simulations match measured response quite well, it would appear to me that the math used by Hornresp is correct, while your idea (whatever it is) may not be.
Repeatedly showing the equalized, uncalibrated small room response of your old AR LF OB doesn't help us understand why your ideas differ from what a Hornresp simulation would indicate.
Art
Since I'm the one that originally recommended this driver on a price/performance basis I'll comment quickly.
Yes, they are suspect. All SI product specs are suspect. Especially the xmax spec. Often, SI will pull Klippel numbers on a prototype version and publish them, then make changes and have a different production version and never change the published specs. Sometimes he will even take the Klippel numbers, add a bunch of mm of xmax based on how he thinks the production version will measure and publish THAT as his xmax spec. I have proof in links if you want it.
Also, when quoting xmax he ONLY considers the Bl graph, ignoring all the other limits that would give a far lower xmax, like Le, suspension nonlinearity, etc. I also have links that show his SI 24 is limited by Le to 20 mm xmax and yet he published a spec of 36 mm for that driver.
The SI 18 D2 and D4 versions have very noticably different specs and will give different response in the same box, although I believe only the D2 version specs are published. And your guess is as good as mine as to how accurate these are. I certainly wouldn't trust them.
Here's a little gem from a recent conversation about the accuracy of his specs. It's from here - SI HT24 sealed test results - Page 6 - AVS Forum | Home Theater Discussions And Reviews
Electrodynamic is SI, and he doesn't care in the least about providing accurate specs, as this image clearly shows.
I've been speaking out about SI's aggressive marketing and bad attitude and outrageously incorrect published specs for a couple of years now, he is a blight on the industry if you are interested in manufacturers publishing accurate specs.
Furthermore, this particular driver, along with all the other drivers that suffer the "high inductance" effect ABSOLUTELY WILL NOT perform as simulated unless you account for it in the sim, which no one ever does. (I put "high inductance" in quotes because as Kravchenko pointed out, it may not necessarily be the inductance that causes this effect in these massive coil/motor designs.) I have a method to simulate these drivers accurately but no one uses it. My sims are always accurate.
So why did I recommend this driver? Very simply, it performs. Third party measurements have shown that it's the value leader.
But it absolutely must be simulated properly or you will end up with a mystery response similar to the massive RE XXX ported build on data-bass that tb46 has mentioned a few times. It's measured response is not even close to what a simple sim would suggest. But I can sim it pretty accurately.
JAG,
So the SI HT18 can be simulated, but only by using unpublished variables that need to be included in the calculations.
How were/are these variables determined? And what are the accurate T/S parameters for the HT18?
I guess many companies are guilty of providing specs that make their products look better. I've always avoided Peavey products, for example, because I felt their stuff seldom met the published specs. They have a low cost 21 inch subwoofer driver at Parts Express right now that sounds good on paper. But I just don't trust the specs.
BO
I'm curious as to what that "idea" is too, because Ben does not mention Qts, which will have a significant impact on the driver's response around its resonance frequency when used in an IB alignment.
In room measurements mean nothing to anyone who isn't in that room and sitting in the exact spot you placed the mic.
I've offered to sim your antique 2 mm xmax woofer in OB several times so you can see how it compares to other designs, you've never taken me up on the offer.
What I can tell you for sure is that the in room response measurements of your various speakers that you have posted is nowhere near typical. So either you are measuring in a very bassy spot in the room, or you are using eq, or you are in a concrete bunker (small apt construction), or your measurement gear isn't calibrated, or any number of a dozen other things. Your AR speaker measured about the same IIRC, and isn't it a ported design with a port tune well above 20 hz? In typical rooms with no eq neither of these speakers will measure anything like what you've shown.
Now concerning this speaker specifically, OB doesn't have room gain per se, but if I understand it correctly it's not really pure OB, as there's a cavity resonance between the large board and the wall and other details of this "subwoofer's" construction are sketchy at best. Regardless, there's resonances going on in the speaker itself, between the OB and the wall, and room resonances as well.
Art also points out that this measurement may have eq applied, which makes the measurement absolutely useless for anyone except you and even then it's only useful with the same eq settings and in the spot the measurement mic was placed.
As usual, the situation is nowhere near as simplistic as you try to make it seem.
If you would like IB response explained, with qts driver details and how this affects overall system q, I can do that. But for now, here's a quick example. This is the IB response of a driver with different qts values. No other parameter is changed, just qts. Therefore it isn't a real world example but it does show how qts alone affects IB response. This is 2 pi response, not in room response.
If the numbers are too small to see, this is shown at .25, .5, 1 and 2 qts. As you can see, large sealed boxes are not a good idea for drivers with qts higher than .5 or so if flat in room response is desired, as the room will boost the lower frequencies. When qts gets to 1 or higher, a sealed box is a bad idea with almost no exception (unless you want a humped response at fs), even with very large sealed boxes the overall system q is going to be very high.
An externally hosted image should be here but it was not working when we last tested it.
Most of the very large coil/motor drivers can not be accurately simulated with a simple sim with the exception of a very few examples that are very expensive like TC Sounds drivers. This is easily seen when simulating any drivers of this type and comparing to measurements on data-bass.com. He gives measured t/s parameters as well as details of the box he used to measure the drivers. None of the sims will match measurements with this type of very large coil/motor drivers, although TC Sounds does a lot better than most.
The sim can be made more accurate by adjusting t/s parameters (although you need accurate t/s parameters in the first place). Adding a bit of Re to the sim and adjusting power to compensate for the added Re losses gives a much more accurate indication of what the frequency response will look like (although it does make some of the other graphs less accurate due to having to input more power than will actually be used). Another guy that has played with my tweak has also noted that decreasing Bl will also do much the same thing, and in that case you don't have to adjust power level, so all the other graphs will also be accurate.
I've discussed this tweak quite a bit since last October, very few people have taken notice, but a couple of key players on avsforum have endorsed this tweak, in fact Josh Ricci (the guy that runs data-bass.com) mentioned that he did something very similar to make his sims more accurate.
I've got dozens of measurements taken by data-bass.com compared to sims in all different types of enclosures with a variety of different drivers. Sealed, ported, tapped horn, front loaded horn, this tweak works every time in every enclosure with every large motor/coil driver using nothing more than published t/s specs (assuming they are accurate).
It can be as simple as applying exactly the same amount of Re (or decreased Bl) to every large motor/coil sim or you can adjust for best match compared to Ricci's measurements to get more accuracy but this will obviously take longer.
The tweak is not 100 percent accurate but it is exponentially more accurate than a simple sim for this type of driver.
If you want to know more I can dig up links or you can ask specific questions. TB46 is aware of this tweak but I don't think he used it in your sim.
Like Art, I too get all choked up with strong emotions and post factless criticism.
I have no issues with HR or statistics.... but as Mark Twain famously said about stats not lying....
No Art, AR did not make 15 inch woofers. Were'd you ever hear of that? Emotions just got the better of ya judgment? Not previously posted, I think, but thanks for thinking of me.
I guess I need to explain my point of view to Art: a driver with a resonance at 15 Hz ought to show a clear whomp-up at 15 Hz in a near-infinite baffle, as tb46 says he sim'ed. As my posted data (with no EQ to speak of, no moveable mic position, too small for LF waves, and definitely no such thing as "cabin gain") clearly shows.
At the risk of boring everybody be repeating the obvious (that still seems to be missed), I believe Bach On would benefit from a "house curve" with a big whomp-up around 16 Hz (and starting with a 15 Hz driver seems pretty nice). Beats me why some people are persevering in modeling flat response* just because their sim marches them in that direction.
A ported box or TL might be dandy - just so long at the house curve is the end result.
Ben
*everybody should aim for a rising bass; flat sounds lousy
Last edited:
- Status
- Not open for further replies.