Better Bass vs Better Room vs Better Driver

[fredk]

After comments on a sub design post, I think it is worth while for me to take a step back and evaluate my objectives and the most efective way to get there.

I think I rolled up several issues into one post without identifying or looking at any of them closely.

My Objective: better sound for pipe organ music. Well OK, I do also want a better response for movies, but music comes first.

Better Room. This is prompted by a comment by TLSguy about where most of the bass in pipe organ music resides: up in the 30Hz region. My current sub has a -3db point of 28Hz, and with significant room gain, I think I get good extension down to around 21Hz.

I am quite happy with the integration of my current sub into my HT, is is quite seemless/transparent. Frequency/room response is another issue.

The yellow is with some treatment, purple is without. There are two issues I find: boomy bass at some frequencies, notes that just loose their visceral feel and strength at the bottom end of the range.

I am confident proper treatment will take care of the boom. Will it help put some of the oomph back in??

Better Bass. In the HT world, lower is better seems to be the mantra. On the sites I currently frequent, there is little/no discussion of the possible tradeoffs of design choices for subs just: big driver, big amp, low tuning.

There are two things I wonder about:
1. how low a tuning do I really need to get the bottom notes on a pipe organ.
2. If I tune lower, am I trading something off at the top end where more of the music resides? When looking at a larger driver, the constant temptation is to tune to 15Hz and get all that lf stuff in movies while I am at it. I don't see any discussions of the tradeoffs of low tuning.

Better Driver. This one really confounds me. When I was researching may mains purchase, there is a lot of discussion of driver performance and design tradeoffs to achieve a level of quality. That discussion seems to be completely missing from sub design discussions (expept the Kappa Perfect build).

As an example, I can choose the Perfect 12 or the Mach5 MJ18M to do the same thing. One is pushed to the limit, the other plays well within spec. Outside of size and cost tradeoffs, I have no way of knowing what the relative performance of the two drivers is. Outside of one car audio site, nobody seems to look at the performance of the raw driver. Does physics change below 80Hz??

Help me sort through this to figure out where to concentrate to reach my goals.

 

[TLS Guy]

After comments on a sub design post, I think it is worth while for me to take a step back and evaluate my objectives and the most efective way to get there.

I think I rolled up several issues into one post without identifying or looking at any of them closely.

My Objective: better sound for pipe organ music. Well OK, I do also want a better response for movies, but music comes first.

Better Room. This is prompted by a comment by TLSguy about where most of the bass in pipe organ music resides: up in the 30Hz region. My current sub has a -3db point of 28Hz, and with significant room gain, I think I get good extension down to around 21Hz.

I am quite happy with the integration of my current sub into my HT, is is quite seemless/transparent. Frequency/room response is another issue.

The yellow is with some treatment, purple is without. There are two issues I find: boomy bass at some frequencies, notes that just loose their visceral feel and strength at the bottom end of the range.

I am confident proper treatment will take care of the boom. Will it help put some of the oomph back in??

Better Bass. In the HT world, lower is better seems to be the mantra. On the sites I currently frequent, there is little/no discussion of the possible tradeoffs of design choices for subs just: big driver, big amp, low tuning.

There are two things I wonder about:
1. how low a tuning do I really need to get the bottom notes on a pipe organ.
2. If I tune lower, am I trading something off at the top end where more of the music resides? When looking at a larger driver, the constant temptation is to tune to 15Hz and get all that lf stuff in movies while I am at it. I don't see any discussions of the tradeoffs of low tuning.

Better Driver. This one really confounds me. When I was researching may mains purchase, there is a lot of discussion of driver performance and design tradeoffs to achieve a level of quality. That discussion seems to be completely missing from sub design discussions (expept the Kappa Perfect build).

As an example, I can choose the Perfect 12 or the Mach5 MJ18M to do the same thing. One is pushed to the limit, the other plays well within spec. Outside of size and cost tradeoffs, I have no way of knowing what the relative performance of the two drivers is. Outside of one car audio site, nobody seems to look at the performance of the raw driver. Does physics change below 80Hz??

Help me sort through this to figure out where to concentrate to reach my goals.

They look like classic wall reflections adding and canceling depending on frequency. Try different woofer positions. Another woofer on the opposing wall might help.

I think you sub hand over is around 80 Hz, and it looks as if there could be some phase issues there.

The boom is coming from the 52 Hz peak.

You seem to have good in room response to 22Hz which should be adequate.

 

[fredk]

They look like classic wall reflections

Indeed. I live in a concrete construction apartment. The main area is 20x20 (with a kitchen centered on one wall) bounded by concrete block and poured concrete. The big peak ranges between 52 and 56Hz depending on where in the room the sub is placed and where I measure. I have tried multiple positions with no success so far.

The measurement is the sub only crossed at 80Hz.

Another woofer on the opposing wall might help

Hence the questions about frequency, drivers etc. That, and I seem to have an irrepressible desire to build something.

which should be adequate

I lust after something more than adequate. I suspect that a second sub will help with the nulls at 40 and 76Hz, but if I am going to build a second sub, I may as well build something that extends lower.

 

[highfigh]

Indeed. I live in a concrete construction apartment. The main area is 20x20 (with a kitchen centered on one wall) bounded by concrete block and poured concrete. The big peak ranges between 52 and 56Hz depending on where in the room the sub is placed and where I measure. I have tried multiple positions with no success so far.

The measurement is the sub only crossed at 80Hz.


Hence the questions about frequency, drivers etc. That, and I seem to have an irrepressible desire to build something.


I lust after something more than adequate. I suspect that a second sub will help with the nulls at 40 and 76Hz, but if I am going to build a second sub, I may as well build something that extends lower.

The 2 dimensions being the same is a big reason for this. Standing waves are bad enough in wood frame construction but in concrete, they just reflect instead of making the walls resonate (which partially dampens those frequencies). Have you tried reversing the wires on the sub?

 

[annunaki]

A powerful EQ/x-over such as the Behringer DCX2496 will aid in your room response greatly as well.

When running dual subs it is highly recommended that the same subwoofer & enclosure be used for both units.

As for tuning, most sub systems, when tuned lower typically have a more favorable overall sytem "Q". Lower tuned enclosures tend to have transient response characteristics to that of a sealed enclosure with substantially more output in the lower reaches of the operating range.

A low tuned enclosure will give up a bit of output at the higher range of it's response, but in most cases this trade off totally acceptable, especially since most x-over points occur around 80hz.

The woofer's parameters really determine how low an enclosure can be tuned. Some woofers do very well tuned extremely low, others will have large drop offs in response below certain tuning frequencies. Some woofers tuned too low will have group delay issues below a certain frequency. Audio is always a set of compromises. There are many great subs out there in which to build a powerful articulate sub sytem around.

Choices high on my list would be the following woofers:

JL Audio W7, W6v2, W3v3, 15W3v2

Audiopulse (nearly any of the drivers except the Epic)

DIYCable.com Maelstrom-x, Shiva-x, DPL series

Infinity Kappa Perfect VQ series drivers

Dayton Audio Reference series

Image Dynamics IDMAX

Selected woofers from Soundsplinter

Most any driver equipped with an XBL2 enhanced motor assembly would be acceptable.

I am sure there are others, but that is a start

 

[fredk]

A powerful EQ/x-over such as the Behringer DCX2496 will aid in your room response greatly as well

It may work well for the 52-56Hz peak because it seems pretty consistant throughout the room. Otherwise, I am more interested in a broad consistant room response than tweeking the critical listening position.

When running dual subs it is highly recommended that the same subwoofer & enclosure be used for both units.

I wish I had the cash to build two subs at once. The plan is to build one and as the funds become available, build a second.

Audio is always a set of compromises

I am still trying to get a handle on the tradeoffs of various drivers, driver parameters and enclosure sizes.
For instance, the Exodos DPL 15 is an xbl^2 motor with lots of excursion, yet it seems to be a much lower price than many other similar drivers. That leads me to wonder why and what the tradeoffs are.
My suspicion is that, because I do not have cabinet size restrictions, or extreme spl requirements, I can probably get by with something less than a top tier driver. A jl 13w7 would be nice, but that money could buy me 2 Mach5 MJ18s and an ep2500.

 

[annunaki]

It may work well for the 52-56Hz peak because it seems pretty consistant throughout the room. Otherwise, I am more interested in a broad consistant room response than tweeking the critical listening position.


I wish I had the cash to build two subs at once. The plan is to build one and as the funds become available, build a second.


I am still trying to get a handle on the tradeoffs of various drivers, driver parameters and enclosure sizes.
For instance, the Exodos DPL 15 is an xbl^2 motor with lots of excursion, yet it seems to be a much lower price than many other similar drivers. That leads me to wonder why and what the tradeoffs are.
My suspicion is that, because I do not have cabinet size restrictions, or extreme spl requirements, I can probably get by with something less than a top tier driver. A jl 13w7 would be nice, but that money could buy me 2 Mach5 MJ18s and an ep2500.

Fred,

The DPL woofers appear to use an inexpensive frame which is where they probably cut some costs. They also use a cupped spider (from pictures) which saves on a spider mounting platform for use with a flat spider. Small trade offs yes, but it does keep the cost down with a very capable/linear motor system. The woofer probably suffers a little bit in the suspension linearity department (because of the spider) but nothing that is too detrimental.

In any case it appears to be a solid value of a woofer, on a budget.

I would say that a W6v2 driver or the Infinity may be more along the lines of what you may need. Both will play quite deep and remain very linear in modest size enclosures. Obviously the 18" maelstrom will allow the extreme low tuning but at the expense of an enormous enclosure.

What is the total budget available for this build and I can give you a few options? Especially since we know the objectives now.

 

[fredk]

My budget is an etherial thing. I work in automotive and already have more time off than I would like. It really depends on wether or not I end up working short weeks going forward or not.

I was looking at around 200 for the driver, less would be better.

I have modeled the Mach5 MJ18M and it would seem to be another good budget choice. The problem with modeling is that it tells you nothing about actual driver performance and how/where non-linearities or other distortion would kick in. It also forces you to trust the manufacturers specs, and that is not always wise.


I am also not sure what amp would be best. For instance, with the MJ18 tuned to 14Hz in a 10cf cab, max power required is around 450w. An EP2500 would seem to be overkill.
FWIW, living in Canada, Mach5 and CSS would seem to be the best value unless the C$ rises against the USD considerably.

At the moment I am watching the local classifieds closely to see if something comes up second hand at a reasonable price. Unfortunately, most of what I see are cheap 10" car subs.

 

[annunaki]

Do you have the T/S parameters or a link for the Mach5? Does it use any thing special to ensure motor & suspension linearity? 3rd party measurements & review become important with individual drivers. You could check to see if the driver has ever been run on a Dumax or Klippel analyzer through a google search.

I have heard decent things about the CSS woofers but I am unsure of their linearity.

 

[highfigh]

My Objective: better sound for pipe organ music. Well OK, I do also want a better response for movies, but music comes first.

Better Room. This is prompted by a comment by TLSguy about where most of the bass in pipe organ music resides: up in the 30Hz region. My current sub has a -3db point of 28Hz, and with significant room gain, I think I get good extension down to around 21Hz.

I am quite happy with the integration of my current sub into my HT, is is quite seemless/transparent. Frequency/room response is another issue.

The yellow is with some treatment, purple is without. There are two issues I find: boomy bass at some frequencies, notes that just loose their visceral feel and strength at the bottom end of the range.

I am confident proper treatment will take care of the boom. Will it help put some of the oomph back in??

Better Bass. In the HT world, lower is better seems to be the mantra. On the sites I currently frequent, there is little/no discussion of the possible tradeoffs of design choices for subs just: big driver, big amp, low tuning.

There are two things I wonder about:
1. how low a tuning do I really need to get the bottom notes on a pipe organ.
2. If I tune lower, am I trading something off at the top end where more of the music resides? When looking at a larger driver, the constant temptation is to tune to 15Hz and get all that lf stuff in movies while I am at it. I don't see any discussions of the tradeoffs of low tuning.

As an example, I can choose the Perfect 12 or the Mach5 MJ18M to do the same thing. One is pushed to the limit, the other plays well within spec. Outside of size and cost tradeoffs, I have no way of knowing what the relative performance of the two drivers is. Outside of one car audio site, nobody seems to look at the performance of the raw driver. Does physics change below 80Hz??

Help me sort through this to figure out where to concentrate to reach my goals.

The fact that you have such a deep notch in the response near 40Hz suggests a phase cancellation, as TLSguy mentioned. I think reversing the leads to the sub is the first thing I would try. Re-measure and EQ after, to find out what changes.

Your room is square, which causes problems with standing waves, or 'modes'. Here's a link that explains quite a bit:
http://www.realtraps.com/pmodecalc.htm

If the sound has more than one path to the mic or your ears, you will have some notes that cancel and some that are augmented. If you have all hard surfaces, these will occur throughout the audible frequency range. If you have surfaces that absorb sound, they will help. Bass is different. There's a lot more energy in bass notes than the other frequencies, so it's harder to absorb or cancel. Bass traps make a big difference.

The Fs of the sub enclosure and port tuning frequency have to work together. If the woofer stops producing usable energy at 60Hz and your port is tunes to 15Hz, knowing that a port's response falls off at 18dB/octave means that at 30Hz, you have lost 18dB of output. By hte time it gets to 80 Hz, your port is doing nothing to help and if the driver isn't designed to produce below a certain frequency, the power handling can be reduced drastically. If the speaker becomes "unloaded", it flaps in the breeze with nothing to control it. Think of a port as blowing across a Coke bottle. It's tuned to one frequency, and one frequency only. The rate at which the frequencies attenuate above and below that are fixed. If there's a wide gap between the port tuning frequency and the high pass F3, you can't avoid having a gap in the response.

Your speakers and room will either work together or against each other. If you have a high Qts subwoofer, the bass will be boomy and annoying. Very unrealistic sounding. A lower Qts, like .7 is very good for a wide variety of music and has a smooth roll off. You can use the room to your advantage, but I don't think it will be possible with a tuning frequency of 15Hz. There's a phenomenon called Transfer Function, and it has to do with how the energy and space work together.

Read this link for more info on this:
http://www.bcae1.com/spboxad1.htm

The raw driver's characteristics mean everything to how it's mounted, if it needs to be ported, how much power it will handle and it's frequency response (The physics we know/understand doesn't change unless you approach the speed of light, so...). The Thiel/Small parameters and calculations are used to find the enclosure volume, tuning frequency, port length/diameter/tuning frequency/power handling, Qts, etc. Once this is done, you theoretically have a properly designed subwoofer system. Now, assuming the port is tuned correctly and the F3 of the box works with the driver (a box of low internal volume is high Qts and large box is lower Qts), the room's acoustical properties will do everything possible to make the speakers sound one way, or another.

Bad speakers in a great room sound better than great speakers in a bad room. The room is everything when you're trying to listen to speech or music.

Read the info at this link for information and look for the ModeCalc software:
http://www.realtraps.com/

and above in the AV University section at the AH home page.

 

[fredk]

Thanks for the explanations highfigh. I am pretty familiar with room modes calculators etc. From a calculator, I have axial and tangential modes at 56Hz and a ceiling to floor axial mode at 71Hz. I suspect that because the 56Hz axial mode is in two directions it gives such a high, consistant and strong peak.

response near 40Hz suggests a phase cancellation, as TLSguy mentioned

The same is probably true of the 70Hz null. With only the sub playing, it is out of phase with itself (a reflection issue) so changing the phase, which I have tried, does nothing.

I somewhat follow what you are writing on port tunings, qts etc. and have seen transfer functions mentioned but have no clue there. I will check out the link when I have time.

Annunaki. I got the MJ18 T/S params from the Mach5 page. When I entered them in winisd, it came up with different numbers from what is on the site in some cases.

I have done various searches using the term Klippel (and found very little), but have not tried Dumax. I will give that a try.

 

[highfigh]

Thanks for the explanations highfigh. I am pretty familiar with room modes calculators etc. From a calculator, I have axial and tangential modes at 56Hz and a ceiling to floor axial mode at 71Hz. I suspect that because the 56Hz axial mode is in two directions it gives such a high, consistant and strong peak.


The same is probably true of the 70Hz null. With only the sub playing, it is out of phase with itself (a reflection issue) so changing the phase, which I have tried, does nothing.

I somewhat follow what you are writing on port tunings, qts etc. and have seen transfer functions mentioned but have no clue there. I will check out the link when I have time.

Annunaki. I got the MJ18 T/S params from the Mach5 page. When I entered them in winisd, it came up with different numbers from what is on the site in some cases.

I have done various searches using the term Klippel (and found very little), but have not tried Dumax. I will give that a try.

Change the mic placement. You didn't say where the listening position is but against the back wall isn't where you want to be. If you do the math, 40Hz has a wavelength of 28.25'. If your speakers are 2' off of the back wall and you're sitting ~14' away from them, 40Hz is at a null. Walk around the room and listen for changes in the bass response. If your first reflections coincide, you'll lose an amazing amount of energy at certain frequencies.

 

[annunaki]

Modeling the Mach 5 MJ-18M it appears that the driver does not like tuning below 26hz or so. The response drops off rather abruptly. If you want deep effortless low end, other drivers may be a better option.

 

[fredk]

I played around in winisd with a few other more capable drivers and it seems you are right. Maybe there is no free lunch.

If I understand modeling correctly, I want the driver to be flat to about 30Hz and then start to roll off??

 

[fredk]

Highfigh. The seating position is about 3.5 feet from the back wall. The speakers are 4 feet from the front wall. I don't have much room to play with the position of either because I need room behind the speakers to get between the kitchen and the bedrooms.

 

[WmAx]

I played around in winisd with a few other more capable drivers and it seems you are right. Maybe there is no free lunch.

If I understand modeling correctly, I want the driver to be flat to about 30Hz and then start to roll off??

Just keep in mind, WinISD does not in any way show driver motor linearity. The power vs. SPL function in WinISD does not take into account this, nor even port compression effect(s).

Why are you entertaining cheap drivers like the Mach now? I mean, it's fine for low budget semi-spl use; great for HT where SQ is not a big deal at all. But with all of the posting you have done on this, you seem to be paranoid about getting 'perfect' performance.

At least the Dayton RS 15" subwoofer has a chance of being of good to high linearity, at least within 10mm or so, based on the 12" version's Klippel data, assuming the 15" has near or equivalent motor/suspension. Probably the cheapest/best, or at least one of the best 15" driver around for that low price range.

If you want to be sure you end up with extreme high quality sound with superb dynamic range, and have zero excuse/reason to upgrade later, I urge you to save a bit more, and at least get a SoundSplinter driver. These are middle line TC Sounds drivers(which is basically better than the flagship expensive drivers from most other brands - you have to realize the quality of TC Sounds drivers was defining of performance in the industry in many ways - their UltraLMS for example is the highest quality subwoofer to be yet measured and verified by published 3rd party data and their middle-line units measure incredibly as well). SoundSplinter still has some of the drivers left. But they won't be around forever, as TC Sounds is currently not producing anything; they went belly up earlier this year. If you save up and get the Sound Splinter driver, you will be content in knowing you have a top notch quality driver that you likely can not even begin to drive to a point where it becomes non linear. I'm serious; the incredible sheer physical range that these units remain linear combined with the superb power handling/ultra-low power compression, means that most people will not even use an amp big enough to come close to using these drivers even near their maximum capabilities. The RLP15 has a conservative thermal power handling rating of 1000 watts.

http://www.soundsplinter.com/rlp15_subwoofer_information.html

-Chris

 

[WmAx]

A jl 13w7 would be nice, but that money could buy me 2 Mach5 MJ18s and an ep2500.

I'm sorry, but this is an absurd comparison. The W7 is a text book perfect example of linearity, with incredible traits all the way around. No way I would ever consider two radically inferior drivers like the Mach units in exchange for the single super-driver(at W7 level performance - which is essentially the pinnacle of possible performance - I consider these super-woofers). Now, I would consider in exchange, using 'slightly' inferior drivers, such as the lower-mid range TC Sounds drivers, which the SoundSplinter RLP15 would fall under; you could get two of these for just about $100 more than a single JL W7 13" would cost. Motor linearity is not 'quite' at the same level as the W7 here. But it's still excellent. You will end up with superior linearity due to using two; radically reducing the excursion for any given SPL. You will also have far greater SPL ability, of course. A true win-win situation.

-Chris

 

[WmAx]

BTW, one more option, if you want a world-class driver with no compromise, is to wait it out a bit more and keep a daily watch on the AudioPulse AXIS drivers at Parts Express. In the last 5 weeks or so, they have had 1/2 price sales on both the 15" AXIS and 12" AXIS. These are incredible drivers in every possible respect, and only inferior to Audio Pulse's own UltraLMS, by a slight margin. I expect you will see another comparable sale price soon. These drivers are essentially clearance - and the sales are probably just a way to help clear out the stock. To put it in perspective, if you drive the 15" AXIS in a 5 cubed sealed box, it will have tremendous output down to 13-14Hz. Driving it with 2000 clean watts in this case, you won't even be able to drive it beyond about 16mm. To put it in perspective, the driver's BL at 14mm, deviates only about 1 percent from rest point according to TC Sounds data. The driver is essentially acting like a theoretical mechanism; textbook perfection. 2000 watts does not stress the driver. It is rated for 2500 continuous thermal, and can easily handle 5000 watts for actual music program. ParadigmDawg on this forum had this built for his use finally, look up his statements in use. He is using it in a massive room(8000+ cubic feet), and not even corner loaded(corner loading will produce substantial more SPL), and he finds it to be exactly as I described it. His highly regarded Paradigm Servo 15 v.2, which btw is a very expensive commercial subwoofer that measures very well compared to most, can not begin to match the linearity of the AXIS, and easily begins to become noticably non linear in comparison to at moderate high SPLs.

-Chris

 

[Warpdrv]

The 13W7 is a great driver, but be wary about trying to find a good Pro Amp that will do 3ohms without cr@ppin out on ya can be expensive. The TC-Sounds drivers are pretty darn nice, they don't do upper bass as well as I have heard, but without a doubt the SQ they deliver is spectacular. Very accurate. My F112 just does such a great job blending with my mains.

annunaki, Ilkka measured the Css SDX15 sealed, and it is almost a direct match to the TC2000 driver in performance... pretty much a drop in replacement.
http://www.creativesound.ca/details.php?model=SDX15 for $279.00 its a pretty darn good driver.

Just a little food for thought....

Fred, is there a budget in mind, are you looking to start up a new build..?
What is the largest size box you can manage...?

I would suggest that you look for a driver that would play well with a EP2500, too keep costs down.

 

[WmAx]

The 13W7 is a great driver, but be wary about trying to find a good Pro Amp that will do 3ohms without cr@ppin out on ya can be expensive. The TC-Sounds drivers are pretty darn nice, they don't do upper bass as well as I have heard, but without a doubt the SQ they deliver is spectacular. Very accurate. My F112 just does such a great job blending with my mains.

W7s are really 4 Ohms nominal. I don't know why they claim 3 Ohms. Any 4 Ohm amp will handle them. Note the DCR given for the W7s. I have a few of the 8" units on hand, and they claim 3 Ohms too. In reality, they are 4 Ohms.

annunaki, Ilkka measured the Css SDX15 sealed, and it is almost a direct match to the TC2000 driver in performance... pretty much a drop in replacement.
http://www.creativesound.ca/details.php?model=SDX15 for $279.00 its a pretty darn good driver.

I fail to see the match. The TC2000 driver produced about 50 % more SPL output across the board, and it had far less power available to it than the CSS. The TC2000 had lower distortion in every point. It actually appears the TC2000 is underpowered, and could produce even higher output. The CSS is no equal to the TC2000 according to the Illka charts. I can only imagine how the CSS would fare against say, a 15" AXIS(TC3000).

Despite the 2" smaller diameter, the JLW7 13" as used in the Fatham produces about 35-40% higher output as compared to the 15" CSS.

-Chris