Jump to content

(8) Sealed Incriminator Audio Judge 21" build


lukeamdman

Recommended Posts

  • Replies 458
  • Created
  • Last Reply

I've summarized the output values for each driver below:

 

f   #1 lo #2 lo #1 hi #2 hi

10  65    67    96    96

20  77    76    107.5 105

30  83    80    114   111

40  86    83    117   114

60  87    85    119   117

80  87    86    119   118.5

120 86    87    119   120.5

 

Indeed, the driver with the smaller motor has up to 3 dB more higher sensitivity from 20-100 Hz or so.  Below 20 Hz, however, is a different story, but unfortunately, there is a problem with the high power simulation for driver #2.  If you look closely at around 17 Hz or so, there is a knee where the roll-off slope changes abruptly.  My guess is that your simulated driver is running out of Xmax at that frequency for the power level you chose.  Now, if you are designing a system using that tool, it would be good to know when Xmax limits your output, but in this case we're talking only about sensitivity and efficiency.  For a fair comparison, that feature shouldn't be there.

 

Note that driver #2 wins by 2 dB in the low power sweeps but merely ties #1 in the high power sweeps.  The low power sweeps reflect what would happen if we weren't overpowering the driver.  If the volts per driver were lower in that same box, you would see driver #2 overtake #1 somewhere probably very near 17 Hz, and below 10 Hz, driver #2 would increase the lead up to 3 dB, which can be shown mathematically by virtue of the fact that you doubled (BL^2)/Re while leaving the other physical parameters unchanged.

 

Now I totally get that a lot of people would prefer a flatter passband and a lower F3.  For a system designer looking to get the most output from the amp in the 20-30 Hz range, a driver with less BL or more mass (for higher Q) may be preferred.  But this thread is focused on a build involving 8 X 21" sealed woofers with the express intent of going for broke in the ULF department.  By ULF, we mean < 20 Hz right?  How about < 10 Hz?  Will Luke have enough headroom for his > 20 Hz tastes with 8 X 21" drivers?  I assume so, and if he doesn't, his horns will fill that gap, right?  This is all about < 20 Hz performance, and for that, high BL^2/Re and Cms as low as reasonably possible seems to be the way to go.  I think Luke made a solid choice here.

 

And let me just remind everyone also (Ricci has mentioned it multiple times), driver #2 with more BL has higher efficiency.  Go back to my plots and compare the gray curve to the yellow curve at 10 Hz.  (Note that my gray curve has 4X (BL)^2/Re of the yellow curve, versus a 2X change in electrodynamic's simulations.)  The sensitivities are equal at 10 Hz, but the impedance of the high BL driver is 9 ohm, twice as much as the 4.5 ohm for the lower BL driver.   For the 108V RMS required to nominally output 3000W per driver, the high BL driver is dissipating only 1333W, whereas the low BL driver is dissipating 2666W.  In this sense, the higher BL driver may "win" even where its sensitivity is somewhat less than the lower BL driver.

Why even bother. Choose extremely high motor force drivers with light moving mass and let the forum know how hard you have worked to achieve proper LFE in a small sealed cabinet. Ignore Josh's data as well. Higher sensitivity drivers (higher motor force with Mms being equal) are better above 50 Hz but they are to be used where the box does the work. I.e. a vented/horn enclosure. Not sealed. 

 

But a 50 Hz F3 is a good thing I suppose. Higher motor force and all. 

Link to comment
Share on other sites

Why even bother. Choose extremely high motor force drivers with light moving mass and let the forum know how hard you have worked to achieve proper LFE in a small sealed cabinet. Ignore Josh's data as well. Higher sensitivity drivers (higher motor force with Mms being equal) are better above 50 Hz but they are to be used where the box does the work. I.e. a vented/horn enclosure. Not sealed. 

 

But a 50 Hz F3 is a good thing I suppose. Higher motor force and all. 

I don't see this "hard work" you feel he has done, relating to working with his driver choice that is. And can you point out a specific example in Josh's data that he has ignored? I am not seeing that either, in fact josh gave you a good response a couple pages back to a similar comment you made already, so why are you repeating yourself?

Link to comment
Share on other sites

Why even bother. Choose extremely high motor force drivers with light moving mass and let the forum know how hard you have worked to achieve proper LFE in a small sealed cabinet. Ignore Josh's data as well. Higher sensitivity drivers (higher motor force with Mms being equal) are better above 50 Hz but they are to be used where the box does the work. I.e. a vented/horn enclosure. Not sealed. 

 

But a 50 Hz F3 is a good thing I suppose. Higher motor force and all. 

I don't care about F3.  Outside the context of the sensitivity vs. frequency (measured response) and impedance vs. frequency (to derive efficiency and amp demand), F3 is totally meaningless.  Furthermore, I don't care about a sub that has ruler flat response down to some low frequency in GP measurements.  Because once that sub is inside a real room, everything changes.  I believe lower Q alignments often work better in real world rooms because their more gentle roll-off better complements the gentle gain.  You might still need EQ to flatten the response, but the amount of EQ required is less.

 

And umm, have you seen Josh's most recent data for his own Rockford Fosgate T3 19s?  His F3 appears to be a dreadful 60 Hz.  Wow!  It must struggle to play the bass in "The Hobbit".  Completely and totally pathetic!  What was that guy thinking, using a car audio sub in his builds?  Oh, wait.  Maybe he actually knows what's he's doing.

Link to comment
Share on other sites

I don't care about F3.  Outside the context of the sensitivity vs. frequency (measured response) and impedance vs. frequency (to derive efficiency and amp demand), F3 is totally meaningless.  Furthermore, I don't care about a sub that has ruler flat response down to some low frequency in GP measurements.  Because once that sub is inside a real room, everything changes.  I believe lower Q alignments often work better in real world rooms because their more gentle roll-off better complements the gentle gain.  You might still need EQ to flatten the response, but the amount of EQ required is less.

 

And umm, have you seen Josh's most recent data for his own Rockford Fosgate T3 19s?  His F3 appears to be a dreadful 60 Hz.  Wow!  It must struggle to play the bass in "The Hobbit".  Completely and totally pathetic!  What was that guy thinking, using a car audio sub in his builds?  Oh, wait.  Maybe he actually knows what's he's doing.

 

Maybe you can convince this forum, and AVS,  to purchase pro-audio drivers which have an F3 of a horrid "dreadful" 60 Hz and EQ it to death to attempt a flat response out of small sealed enclosure. Josh's drivers are high motor force drivers which he is using at his discretion. A Q of 0.30 or 0.50 is not largely sought after regarding subwoofer bass response as it reduces LFE as I supplied an example previously. If you are OK with an F3 of 60 Hz out of your subwoofer system, simply carry on. I, however, prefer a more deeply reaching subwoofer.

 

My HST drivers and HS and BHS drivers have F3's in the 30's. Couple that with room gain in the same situation and your 60 Hz F3 looks pretty thin. And the latter is why you strive to convince the forum to yield to your ideals when you have zero data other than WinISD to go off of. Maybe shift your target F3 up to 100 Hz and convince the forum to go after that. I believe the F30 will yield a better SPL number with the F3 100 Hz driver.  :rolleyes:

Link to comment
Share on other sites

Nick

I largely agree with SME"s post # 381 that was quoted.

 

Back many years ago when people ran their bass system raw with perhaps only a low pass filter, a  response with lower F3 meant a lot more. Now almost every system is using signal shaping or should be (Save for car audio but they have HPF and LPF usually). Sealed systems typically require boost to the low end, or cut to the mid band anyway. A low pass filter modifies the top end FR shape as well. Natively deep extension doesn't matter nearly as much as it used to for me. I look for other factors such as driver displacement capability, linearity, efficiency overall and EFF vs frequency and behavior with large signals. F3 does not determine which system will have more deep bass headroom or better efficiency. T3-19 has an F3 of 60Hz in the test cab. Dayton UM-18 has an F3 an octave lower at 30Hz in the same cab. Both of them have the impedance maximum near 30Hz because it is largely set by the air volume. The T3-19 offers roughly 4dB extra headroom in the deep bass. The 19 is an abnormal driver though. Low Qts/high EFF drivers resulting in highish F3 points rarely have more than 10-15mm xmax. (Pro drivers usually). The limited displacement capability, far more so than the native rising response shape, or even the Fs,  is what limits their usefulness for sealed systems.

 

Response shape and system Q should still be considered as part of the total enchilada of course but these days I look to avoid a peaking response in the 35-60Hz range first. That's because it interacts badly with my room acoustics (mid-room seating) and is likely to cause the need for bigger amounts of signal shaping.

Link to comment
Share on other sites

Re: break-in, I don't know how you are going to give them the necessary excursion * time without needing to re-drywall your house or have your neighbors move away.  What you need to do is park those things near Xmax in free-air for a few days.

 

So... have you tried wiring each cab with one driver out of phase, just for the purpose of breaking in?  You could run all four cabs at once with alternate drivers canceling, and you'd have a whole lot of air whooshing and hopefully little net sound output.  The cabs might do some dancing, and you'd want to first find the new shifted impedance max in that arrangement to avoid overcooking them (as they are coupled and not free-air), but besides that it sounds cake.  Also you'd want to shoot some video of the whole arrangement huffing at war volume and post that here.   B)

 

Link to comment
Share on other sites

Re: break-in, I don't know how you are going to give them the necessary excursion * time without needing to re-drywall your house or have your neighbors move away.  What you need to do is park those things near Xmax in free-air for a few days.

 

So... have you tried wiring each cab with one driver out of phase, just for the purpose of breaking in?  You could run all four cabs at once with alternate drivers canceling, and you'd have a whole lot of air whooshing and hopefully little net sound output.  The cabs might do some dancing, and you'd want to first find the new shifted impedance max in that arrangement to avoid overcooking them (as they are coupled and not free-air), but besides that it sounds cake.  Also you'd want to shoot some video of the whole arrangement huffing at war volume and post that here.   B)

 

My break-in plan is the second SP2-12K that gets here later this week B)

Link to comment
Share on other sites

Nick

I largely agree with SME"s post # 381 that was quoted.

 

Back many years ago when people ran their bass system raw with perhaps only a low pass filter, a  response with lower F3 meant a lot more. Now almost every system is using signal shaping or should be (Save for car audio but they have HPF and LPF usually). Sealed systems typically require boost to the low end, or cut to the mid band anyway. A low pass filter modifies the top end FR shape as well. Natively deep extension doesn't matter nearly as much as it used to for me. I look for other factors such as driver displacement capability, linearity, efficiency overall and EFF vs frequency and behavior with large signals. F3 does not determine which system will have more deep bass headroom or better efficiency. T3-19 has an F3 of 60Hz in the test cab. Dayton UM-18 has an F3 an octave lower at 30Hz in the same cab. Both of them have the impedance maximum near 30Hz because it is largely set by the air volume. The T3-19 offers roughly 4dB extra headroom in the deep bass. The 19 is an abnormal driver though. Low Qts/high EFF drivers resulting in highish F3 points rarely have more than 10-15mm xmax. (Pro drivers usually). The limited displacement capability, far more so than the native rising response shape, or even the Fs,  is what limits their usefulness for sealed systems.

 

Response shape and system Q should still be considered as part of the total enchilada of course but these days I look to avoid a peaking response in the 35-60Hz range first. That's because it interacts badly with my room acoustics (mid-room seating) and is likely to cause the need for bigger amounts of signal shaping.

 

Just for discussion-sake how do the two bold parts make sense? Instead of letting the driver do the work for you, you are saying you are OK with adding additional home runs and adding amplifiers to satisfy substantial voltage requirements needed to boost the low end by 6+ more dB before approaching 20 Hz? 

 

SME said it himself: "Indeed, the driver with the smaller motor has up to 3 dB more higher sensitivity from 20-100 Hz or so.  Below 20 Hz, however, is a different story," 

 

I really don't care if you guys have to spend extra money on electronics [additional home runs, more amps, etc] because your subwoofers need the extra boost, but not everyone is in the same boat. Most people are not. Factor in room gain and an F3 of 60 Hz is not desireable. In "small" sealed alignments an F3 of 30 Hz adds much more output to the overall sub-bass spectrum as the woofer is doing the work for you without the need for double or quadruple the amplifier power to achieve a flat response. Aside from your RF's most drivers will run into serious issues when you start adding that much boost to the signal down low. 

 

*To people who don't post on here but only read* Josh's RF's are special drivers. Not specially made, but special in that they have protection built in to limit them from damaging themselves. I won't get into why or how but they are drivers truly worth every penny from a technology standpoint. :D

Link to comment
Share on other sites

Just for discussion-sake how do the two bold parts make sense? Instead of letting the driver do the work for you, you are saying you are OK with adding additional home runs and adding amplifiers to satisfy substantial voltage requirements needed to boost the low end by 6+ more dB before approaching 20 Hz?

 

I'm one of those people that doesn't really post here much, but if I can learn something by attempting to answer the question, so much the better, right?

 

Let's do a quick comparison between two DIY systems that have been measured:

http://www.data-bass.com/data?page=system&id=110

http://www.data-bass.com/data?page=system&id=112

 

Now looking at the sensitivities of both on the low end, you get:

 

BMS / HST

 

20.142Hz: 82.244dB / 80.773dB

16.113Hz: 77.937dB / 77.073dB

12.451Hz: 73.66dB / 72.45dB

10.254Hz: 70.517dB / 68.948dB

 

Obviously since the BMS driver was tested with 2.83V, you'll have to knock off 3dB from it's numbers to equal the 2V test of the HST driver. Impedance is close enough between the two at the frequencies involved that it's not all that important (the BMS is higher except at 20.142Hz in any case). However, what I see from this comparison is that the low end sensitivity is relatively close, not 6dB+ off. This is in spite of the fact that the BMS driver has higher BL, less moving mass, and an Fs of 29.1Hz vs 20.1Hz.

 

What I also see is that the top end of the BMS driver appears to be much more sensitive. As such, while it appears that you need to boost the bottom end considerably to achieve a flat response, the other way of looking at it is that you're cutting back the overachieving top end.

Link to comment
Share on other sites

Just for discussion-sake how do the two bold parts make sense? Instead of letting the driver do the work for you, you are saying you are OK with adding additional home runs and adding amplifiers to satisfy substantial voltage requirements needed to boost the low end by 6+ more dB before approaching 20 Hz? 

 

SME said it himself: "Indeed, the driver with the smaller motor has up to 3 dB more higher sensitivity from 20-100 Hz or so.  Below 20 Hz, however, is a different story," 

 

I really don't care if you guys have to spend extra money on electronics [additional home runs, more amps, etc] because your subwoofers need the extra boost, but not everyone is in the same boat. Most people are not. Factor in room gain and an F3 of 60 Hz is not desireable. In "small" sealed alignments an F3 of 30 Hz adds much more output to the overall sub-bass spectrum as the woofer is doing the work for you without the need for double or quadruple the amplifier power to achieve a flat response. Aside from your RF's most drivers will run into serious issues when you start adding that much boost to the signal down low. 

 

*To people who don't post on here but only read* Josh's RF's are special drivers. Not specially made, but special in that they have protection built in to limit them from damaging themselves. I won't get into why or how but they are drivers truly worth every penny from a technology standpoint. :D

 

 

Hey Nick,

 

Yes the RF's are a bit unusual. There are very few drivers with that type of response shape and low qts coupled with a legit 35mm xmax. They don't start hitting the brakes until 45ish mm. It does not sound pretty when it happens.  ;)

 

I don't see the two bolded lines as contradictory. The higher qts (weaker motor system) driver can have a lower natural F3 with a flatter top end. Typically it would have extra sensitivity over some part of the bass bandwidth. 20-60Hz perhaps. That is true. A driver with higher motor strength in the same cab will have a much higher F3 and will have less sensitivity over the mid part of the bass bandwidth. This is obvious on a FR comparison. If you scale back and look at a wider frequency range like 5-150Hz what becomes apparent is that the high motor strength system has a smoother wide bandwidth response and will regain the advantage in sensitivity in the deepest frequencies and also at the top end. The weaker motor'd system is actually more peaky and has more of its energy concentrated into a narrower bandwidth. The knee is much sharper. If the motor is strong enough there won't even be a response knee in the bass bandwidth. None of this means that there is less bass or more boost required at 20Hz as in more energy put into the driver or required from the amp. The response shape isn't the right tool to determine that. That would be the voltage sensitivity and the impedance curve which together can give you the constant power response or the power applied to the speaker. The strong motor system might have the same 20Hz sensitivity or slightly less but a 6dB difference with the same size driver and the same amount of normalized voltage would be out of the ordinary. There are a lot of generalizations and assumptions made here of course.

 

I think I'm going to start a random discussion thread for this since this topic seems to have been discussed in a few different threads now.

Link to comment
Share on other sites

Hey Nick,

 

Yes the RF's are a bit unusual. There are very few drivers with that type of response shape and low qts coupled with a legit 35mm xmax. They don't start hitting the brakes until 45ish mm. It does not sound pretty when it happens.  ;)

 

I don't see the two bolded lines as contradictory. The higher qts (weaker motor system) driver can have a lower natural F3 with a flatter top end. Typically it would have extra sensitivity over some part of the bass bandwidth. 20-60Hz perhaps. That is true. A driver with higher motor strength in the same cab will have a much higher F3 and will have less sensitivity over the mid part of the bass bandwidth. This is obvious on a FR comparison. If you scale back and look at a wider frequency range like 5-150Hz what becomes apparent is that the high motor strength system has a smoother wide bandwidth response and will regain the advantage in sensitivity in the deepest frequencies and also at the top end. The weaker motor'd system is actually more peaky and has more of its energy concentrated into a narrower bandwidth. The knee is much sharper. If the motor is strong enough there won't even be a response knee in the bass bandwidth. None of this means that there is less bass or more boost required at 20Hz as in more energy put into the driver or required from the amp. The response shape isn't the right tool to determine that. That would be the voltage sensitivity and the impedance curve which together can give you the constant power response or the power applied to the speaker. The strong motor system might have the same 20Hz sensitivity or slightly less but a 6dB difference with the same size driver and the same amount of normalized voltage would be out of the ordinary. There are a lot of generalizations and assumptions made here of course.

 

I think I'm going to start a random discussion thread for this since this topic seems to have been discussed in a few different threads now.

 

I agree - start a new thread. Even Steve81 used 21 Hz and below for his example. What about 20 Hz to 60 Hz? Why not have the driver do the work for you rather than add 2 dB, 3 dB, plus, to flatten the response curve in a "small" sealed box? Every driver has its place. PA midbass drivers have no business being in 0.5 ft^3 sealed for subwoofer duty.  :P

Link to comment
Share on other sites

Even Steve81 used 21 Hz and below for his example. What about 20 Hz to 60 Hz? Why not have the driver do the work for you rather than add 2 dB, 3 dB, plus, to flatten the response curve in a "small" sealed box? Every driver has its place. PA midbass drivers have no business being in 0.5 ft^3 sealed for subwoofer duty.  :P

 

Ask and ye shall receive.

 

BMS 2.83V / HST 2V

 

25.269Hz: 85.709dB / 84.146dB

31.86Hz: 89.399dB / 88.206dB

40.283Hz: 92.886dB / 91.03dB

50.171Hz: 95.6dB / 91.115dB

63.354Hz: 97.047dB / 89.231dB

 

Again, subtracting 3dB from the BMS figures, you're looking at a sensitivity difference between 1-2dB from 25Hz-40Hz; 50Hz on up is in favor of the BMS (significantly so by 63Hz). In terms of impedance, the BMS never sinks below 8 ohms in this range, while the HST drops to a hair under 6 ohms around 55Hz.

 

Keep in mind, both drivers are being tested in the same 4.2 cu ft cabinet, vs handicapping the BMS with a 0.5 cu ft enclosure.

Link to comment
Share on other sites

Shocking.

 

Not for us but if I had just them running and put any umm... "normal" person in there they'd think I had a sweet sub system. I get single digit bass with them but at louder volumes you can start hearing a lot of 3rd order distortion come in. Kind of sounds raw and growl-y which is cool but it's not as clean as just letting the HT18's and RLp's do their thing instead. A bit more muddy with them trying to do all the heavy lifting with some material. Could be my amp system though. I don't think I get enough voltage out doing 8ohm stereo from the EP4k so it could just be that. It totally works though but I wouldn't recommend that everybody start doing it that way.

 

I've tried playing them alone with stuff several times and had I not experienced what it's like with everything else on I might be able to live with them that way.

 

 

I'm sure Luke's mega mains can do some serious "subwoofer-less" bass, no problem.

Link to comment
Share on other sites

I'm sure Luke's mega mains can do some serious "subwoofer-less" bass, no problem.

That's an understatement :)

 

Seriously, for just about anybody else, they'd be all they would need. Hell, I'd happily take them. They destroy my THTs in output, and they go just as low. Honestly, every time Luke plays them full range, they put a huge smile on our face. They're so incredibly good.

Link to comment
Share on other sites

Shocking.

 

Not for us but if I had just them running and put any umm... "normal" person in there they'd think I had a sweet sub system. I get single digit bass with them but at louder volumes you can start hearing a lot of 3rd order distortion come in. Kind of sounds raw and growl-y which is cool but it's not as clean as just letting the HT18's and RLp's do their thing instead. A bit more muddy with them trying to do all the heavy lifting with some material. Could be my amp system though. I don't think I get enough voltage out doing 8ohm stereo from the EP4k so it could just be that. It totally works though but I wouldn't recommend that everybody start doing it that way.

 

I've tried playing them alone with stuff several times and had I not experienced what it's like with everything else on I might be able to live with them that way.

 

 

I'm sure Luke's mega mains can do some serious "subwoofer-less" bass, no problem.

 

 

That's an understatement :)

 

Seriously, for just about anybody else, they'd be all they would need. Hell, I'd happily take them. They destroy my THTs in output, and they go just as low. Honestly, every time Luke plays them full range, they put a huge smile on our face. They're so incredibly good.

 

Thanks, Mike!  Got to really hand it to BMS and AE as they make some awesome drivers.

 

I think your THT's extend a little deeper.  Here's the full range response of my mains with a 17hz BW3 HPF:

 

mains_zpswcljcvpx.jpg

Link to comment
Share on other sites

Thanks, Mike!  Got to really hand it to BMS and AE as they make some awesome drivers.

 

+1 for the AE drivers.  I'm only a month into using mine (1 X 12" TD12M in each left and right speaker) and probably haven't even properly broken them in yet, but so far they have proven to be excellent.  I sincerely believe these sound much better than any woofers I've heard, but I don't know exactly why that would be the case.  It may have to do with low inductance, but I don't know.  The sound seems to be superior to other woofers I've heard, even at low playback levels.  The sound seems faster and more cohesive.

 

 

On the subject of pro-style woofers for sub use.  Here's some measurements of my own speakers, from a close-mic position (in black) and from the main listening position (in green).  They are ported, but I stuck foam plugs in them because I wanted some idea of what kind of room gain I have to work with when I build a sealed sub system:

 

post-1549-0-64034100-1459141212_thumb.png

 

Some caveats deserve attention.  The close-mic was done somewhat off-axis of the woofer and baffle, and this is the primary reason for the close-mic measurement to have a lower high-frequency roll-off than the MLP measurement.  The second caveat is that I made a "best-guess" effort to vertically align the curves, but I'm pretty uncertain as to whether this is the correct alignment.  I've written strong criticism for previous posts about representing data like this, so this is a heads up that either curve could be shifted up or down and the data would be just as "valid".  I have no good reference for how to line these curves up.

 

With that out of the way look at the response in the sub-bass area.  Ignoring the nasty room resonances and SBIR problems (the speakers were not optimally placed when these measurements were done), note the overall trend in-room is to be flat all the way down to 20 Hz.  What sort of woofers are these I'm using?  Why, they are the pro-style AE TD12M.  IIRC (from impedance measurements), I'm getting a sealed resonance of ~95 Hz with Q=0.6 or so.  In anechoic conditions, that's terrible for getting good bass in the 20-80 Hz region, but in-room, the response extends to 20 Hz!  Indeed, it'd be almost flat with baffle-step compensation applied.

 

So tell me again, why I want a sub that goes to 20 Hz (or lower) with a flat Q=0.707 response and a very low F3?  I just don't see the point here.  Now, I'm not saying everyone should use TD12Ms for subs.  They are mid-range drivers that only have 6 mm excursion.  The point is (edited for clarity) that the combination of big motor and low mass are favorable for many of us in our *in-room* applications.  Our rooms make up for the loss of sensitivity in the middle of the range.  The trouble is that woofers with these traits very rarely have sufficient excursion to be usable in the lowest frequencies.

 

Of course, unlike the TD12Ms, Luke's drivers have plenty of excursion, and the ports reduce the excursion (above the tuning frequency) further.  Say Luke, how often do you play those things loud enough to even see the drivers move?

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.


×
×
  • Create New...