DIYMobileAudio.com Car Stereo Forum banner

In search of Bass in a car with not much Space

18K views 110 replies 23 participants last post by  captainscarlett  
#1 ·
This is a writeup in the footsteps of lycan, Patrick Bateman, and so many others here at diyma, which intends to answer a specific question regarding sub-bass reproduction down to 20 Hz. How do you do it in a subcompact car, without taking up the entire cargo area? Everyone's seen that walled-off CRX stuffed with MTX jackhammers or MMATS juggernauts. Or that civic with some kind of tapped horn / t-line enclosure that hits like a pissed off Irishman. But what if you actually want to carry passengers, or maybe a bag of groceries? And what if you want to reproduce that low cleanly? What compromises can be made?

Patrick Bateman started an excellent thread last year entitled, "BASS! How Low Can You Go?" (this thread is kind of trying to pick up where he left off), where he states:
To sum it up, if you're speakers only go to 30hz, you're probably not missing a whole lot. Almost all the "bass" energy is in the octave from forty to eighty hertz. But there *are* a handful of tracks with synthetic bass lines that can only be heard properly if your speakers are flat to 20, or even 18hz. And yes, there's even bass down to 3hz.
Later, he posted:
I went to the Skrillex / Knife Party show in Vancouver last week, and that had to be, without a doubt, the most bass I've ever heard in my life. ... Anyways, Skrillex was doing thes same trick, but to the whole damn venue. I have no freaken idea how much power and subwoofers it takes to do that, but it was EPIC. Even if you hate Skrillex, hate Dubstep, and think that dance music is for douchebags, it was a neat trick.

When I was walking up to the venue you could see the walls flexing. That is an absurd amount of power. This is the kind of sound system that could probably damage a venue. I wouldn't be shocked to hear that it could actually crack a wall.

Also, I want to stress that it wasn't that the show was loud. It was that there was a level of subharmonic bass that I've never seen attempted. My home subs do 15hz, and I have a train that goes by my house twice a day. So I know the difference between a 40hz bassline and a 10hz roar. You hear one, and FEEL the other.
I saw Skrillex in December and can confirm everything Patrick says above. My pants legs were flapping from all the sub-bass, and could literally feel the floor shake. I'm pretty sure a few pair of britches got crapped in that night :surprised: If you haven't experienced such an event, get out of your mom's basement and do so, right now.. this post will be here when you get back.

Love it or hate it, Dubstep and other fledgling electronic genres of music including Complextro, Moombahcore, and Glitchhop will make your system work harder than just about any other type of music out there. And IMO there is definitely some overlap between the Bass-Head and the SQ junkie.. They are not mutually exclusive. I will wholeheartedly agree that classic rock sounds like **** when the bass is boosted. This is because the information was never there in the first place, and the tonality goes to hell when most "bass boost" circuits increase the bands from 40 to 100 Hz. The genres mentioned above, however, were mastered to include sub- and infra-sonic bass down to DC, and don't need to be boosted at all. So the tonality is still there. Combined with a great three-way setup, and a rear-fill with Logic-7 style phase steering, these tracks sound positively amazing. A quick note about the rear fill.. There's a lot of out-of-phase content mixed into electronic tracks these days -- If you have Logic 7, listen to Rusko - Hold On (feat. Amber Coffman) (Sub Focus Remix), I can't imagine listening to it with only a front stage.

I am still somewhat of a newb, but here are the facts as I understand them (I am over-simplifying some of them for sake of brevity):
  • Hoffman's Iron Law states that the efficiency of a woofer system is directly proportional to its cabinet volume and the cube of its cutoff frequency (the lowest frequency it can usefully reproduce). The obvious implication is that to reduce the cutoff frequency by a factor of two, e.g. from 40 Hz to 20 Hz, while still retaining the same system efficiency, you need to increase the enclosure volume by 23=8 times! In other words, to reproduce ever lower frequencies at the same output level you need an extremely large box! -- quoted from a PB thread
  • More often than not, larger subwoofers reproduce lower frequencies more easily than smaller ones, while being more efficient. Consequently, larger drivers require a larger enclosures.
  • Ported enclosures sacrifice low end extension for efficiency higher in the passband. Most other exotic enclosure types work in the same way, and therefore aren't well suited to infra-sonic duties.
  • Making an enclosure smaller will reduce low-end frequency response.
  • Multiple subwoofers may offer better performance than a single subwoofer due to power compression. Everything else being equal, the multiple driver setup can play 3 dB louder with the same xmax.
  • Playing a a sub near or past its xmax will usually start to develop non-linearities that results in harmonic distortion. Some speaker models use shorting rings and other "tricks" to reduce non-linearities
  • Isobarik designs attempt to reduce non-linearities by using a set of matched drivers in push-pull configuration. I've heard mixed results on how well this actually works.
  • Sub-bass frequencies take considerable power (wattage) to reproduce. For a small car, class D amps are the way to go.
  • Cabin gain = free bass, and you get a lot of it in a subcompact. My R56 gets around 27 dB on the low end.
The short story is... Bass needs space. If you got the space, sub-bass is a piece of cake. If not, you will have to work around a serious set of compromises. Cabin gain is the only silver lining in an otherwise bleak and hazy future.


And to quickly dispel one of the most common myths. For anyone that says there's not a lot of content down at 20 Hz, I give you this, a random dubstep mix out of my iTunes library. It's a spectral plot of Excision Shambhala 2010 compilation, at around 1:19:20.
Image



Here's another track entitled Scatta by Skrillex and Foreign Beggars. I consider this to be a excellent sub-bass reference track. Definitely worth a listen if your system is up to it.
Image


For comparison, here's Hank Williams Sr. singing Your Cheatin' Heart, recorded in 1952. There's literally nothing past 40 hz.
Image


So in the case of the first two tracks (and many others), the information is there waiting to be played. For my next post, I will explain the current setup I'm using and the measurements that I came up with.
 
#2 ·
A Mini Cooper S has 5.7 cubic feet of cargo space in the hatch area (compare that to 21 cubic feet in a 2012 Ford Explorer). Before I had a sub box, there was barely enough cargo room to get a load of groceries home without folding down the rear seat. Adding even a small box is a huge compromise with an area this small.

So, with 5.7 cubic feet to work with, that's:

  • Room for 1x of your average 12 in a false-floor configuration, with room for your groceries..
  • Barely enough room for two ported JL 12W7's, with no usable cargo room.
  • Room for multiple drivers in a false-floor configuration, as long as they're designed to work in a small enclosure space.
  • No way in hell of getting any amount of 15's back there.
  • The last option is to keep the cargo area and put a set of 15's where the rear seats are. The vehicle is now a two-seater with the stock cargo area.. No way to fold the seats down or get anything big in the vehicle. Also, the middle of the car isn't a great spot for them (no corner loading)

After enlisting XtremeRevolution and Pete from PWK, I was presented with a false floor sealed enclosure, designed to fit four 8" drivers with 0.3 cubic feet per driver. I went with the Alpine SWR-843D's, as they were the forum boner at the time.

Image


We knew compromises had to be made... So it was decided that the Alpines could hit greater xmax cleanly (thanks to the shorting rings and other trickery), and by having four of them, the surface area to xmax ratio would be favorable. The main advantage is that the box is extremely small and flat, and hides in the floor of the hatch area. The main disadvantage, however, was the F3 rolloff at 53 Hz. The first thing I did after the box was built and wired was to take some RTA measurements to see how bad the harmonic distortion would be below the F3. Here are my results:

Image

This was my test setup. Three of the four subwoofers wired, as I'm currently missing an amp. TrueRTA is running on the laptop, with the M-Audio MobilePre's Microphone #1 input connected to a Behringer ECM8000, and Line Output #1 connected directly to the amplifier subwoofer channels. I'm doing this outside the car, as I'm currently only interested in measuring harmonic distortion, and the rattles in my car would make that quite hard.

Image

100 Hz sinewave. Very clean.

Image

60 Hz sinewave (just above the F3 cutoff). Still very clean.

Image

30 Hz sinewave. The harmonic distortion is very obvious on the RTA display, and is somewhat audible. The 2nd harmonic is 14 dB down.

No picture at 20 Hz. At low volumes, it's masked by the ambient noise, and a higher volumes it sounded like utter crap. Lots of flappy noises from the driver suspension. No need for an RTA plot when it's that obvious.

When placed in the car, the cabin gain helped tremendously. I didn't really come up with any decent RTA measurements, however. The subs seem capable of shaking the car to pieces, but will do so with rich harmonics that sound offensive to the ear.

The jury is still out on this one. While this design meets three of my requirements (small, low, and loud), it doesn't meet my third (to do so cleanly). I'm tempted to temporarily mount the subs in Isobarik configuration to see if the non-linearities go away.
 
#3 ·
30 Hz sinewave. The harmonic distortion is very obvious on the RTA display, and is somewhat audible. The 2nd harmonic is 14 dB down.

No picture at 20 Hz. At low volumes, it's masked by the ambient noise, and a higher volumes it sounded like utter crap. Lots of flappy noises from the driver suspension. No need for an RTA plot when it's that obvious.
I'm curious - did you take the measurements at the same drive level from the amp, or did you take them at the same SPL level (which would mean that you increased the amp's output for the lower frequency measurements).

Note: A sine sweep measurement with HolmImpulse will give you a quick idea of the THD across the range of the subwoofer's response.


I'm tempted to temporarily mount the subs in Isobarik configuration to see if the non-linearities go away.
The non-linearities in your measurements appear to be mostly odd-order. Isobaric mounting is likely not going to reduce this by much.
 
#4 ·
Interesting thread. I wanted to get low, so I went simple. I bought a JBL P1224, and built a big low tuned ported box (2.5 cubes @ 25 hz) wherein the port takes up a third of the box (port makes a 90 degree turn and then a 180 degree turn). The result is zero in car rolloff all the way down to 20 hz and below. If anything, it gains steam as it gets subsonic vs rolling off. Never been happier, and did it with a single 12.

Edit: I'm thinking it's possible to sell off 2 of the Alpines, and use the rest of the space for box and port, and tune stupid low and go as big as can fit, and thus low bass in minimal space.
 
#7 ·
Edit: I'm thinking it's possible to sell off 2 of the Alpines, and use the rest of the space for box and port, and tune stupid low and go as big as can fit, and thus low bass in minimal space.
I'm with Dan here. Before you give up on these subs you might as well spend another $50 & get a ported design from Pete. Maybe he can give you some input on if there will be much of a benefit going with 2 ported over the 4 sealed.
 
#6 ·
I like the idea of this thread and the Patrick Bateman thread you linked, which I've read. Hopefully these discussions can help people understand a) inaudible bass is out there in professional tracks but only if you're searching for it b) the compromises we all make become much more severe, if that word is appropriate, if we choose to emphasize the less audible bass frequencies.
 
#22 ·
A good couple rules of thumb are:

1). use double the cone area of your active sub for passive radiators, or at least a size larger (1 x 12 inch sub + 2 x 12 inch PRs, or 1 x 15 inch PR)

2). Provide the same net volume as is optimal with traditional ported (but you save all the space used by big long ports to tune so low)

3). I've read it lends good results to mount the PRs on the sides of the box perpendicular to the side your actual sub is mounted on.

I've always wanted to try a PR setup myself, but it costs more than just adding extra walls to a MDF box. I will one day for sure though.
 
#23 ·
3 ways to get low. This is from the Home theater crowd but applicable)

Sealed: You get a high displacement linear sub (super woofer) place it in a sealed enclosure and power/EQ the bejeebus out of it. Think TC Sounds LMS 5400.

IB: Displacement Displacement Displacement. running 4 to infinity 18"+ woofers to keep output up and excursion low. Requires some EQ.



Large low tuned (LLT): Name is self explanatory. Obviously in your application, you have issues with the large. Add in the requirements for adequate port volume and that further compounds the issues.

or use a Passive radiator as mentioned above.

Here are some links for an idea.

AE Speakers --- Superb Quality, Unforgettable Performance, Definitely.

Got my PSI Passive Radiator -- What For ?! - SSA Car Audio Forum
 
#34 ·
3 ways to get low. This is from the Home theater crowd but applicable)

Sealed: You get a high displacement linear sub (super woofer) place it in a sealed enclosure and power/EQ the bejeebus out of it. Think TC Sounds LMS 5400.

IB: Displacement Displacement Displacement. running 4 to infinity 18"+ woofers to keep output up and excursion low. Requires some EQ.

Large low tuned (LLT): Name is self explanatory. Obviously in your application, you have issues with the large. Add in the requirements for adequate port volume and that further compounds the issues.

or use a Passive radiator as mentioned above.

Here are some links for an idea.

AE Speakers --- Superb Quality, Unforgettable Performance, Definitely.

Got my PSI Passive Radiator -- What For ?! - SSA Car Audio Forum
So since all of those home-theater methods would be insane and/or impossible in a subcompact (I'm not walling of my brand new car with 18's, buddy!), how would these look when scaled down to MINI proportions:

  1. Sealed single high displacement linear sub: A single 12" or a JL 13.5" sealed would be the biggest that I could fit. I'm thinking something like a JL 13W7AE-D1.5 ?? Tons of power and eq. I could certainly model this up and add in the measured cabin gain... They recommend a 1.875 cu ft enclosure, which is about 150% larger than my current sealed enclosure.
  2. IB Displacement Displacement Displacement: For a MINI Cooper? Inconceivable! The closes would be something like the array I have now, which is obviously not IB... but it is an array, and does have more displacement than a pair of 12's fwiw. Probably the least desirable option.
  3. Large low tuned (LLT): Again, I could fit as large as a single 13.5 if I give up most of my trunk space.. or something smaller if they model well. This is where my money's at right now. Either with conventional ports or (possibly) passive radiators if they'll fit and the geometry works out.
 
#24 ·
Also remember tht a lot of equipment will not play low. Not the subs but the amplifiers. Non defeatable high pass filters etc... Again something I have noticed in the HT circuit. Just something to be aware of.
 
#27 ·
While the Epic 8 requires a very small sealed enclosure, that does not answer any of the questions he is wanting answered or achieve any of the goals he is trying to achieve. So going out and spending more money on a different 8" subwoofer is bad advice. The performance will be within a few DB of each other. plus, I am not sure if the Alpine is designed to have lower Le induced distortion over the Epic but I believe it does. Xmax is close to one another I believe. And fitting the Epic 8's in a box half the size would require fiberglassing a new enclosure from the looks of his original for the Alpines.
 
#30 ·
If you've been reading Patrick's post you'll have probably noticed I bang on about my linkwitz transform enclosure I did for my 2003 Cooper S-works really well for me in my 1st "SQ" focused system. The cabin gain has added loads of bottom end and still works at motorway speeds. I do listen to dub-step, but haven't geared the system around it as it's not a major genre for me-however I have to say I'm pleasantly surprised with my results-LT was a total gamble for me, read about it, tried it - it works:)

On another forum I stumbled across this thread:Pros/Cons of large voice coils? - diyAudio

The Audiobahn subs this guy has picked up gave a nice bump in sub bass levels when modeled in BBpro using my car's cabin gain (as found in Andy W's HK graph) and compared to the Aliante-plus it handles around 4x the power. 6db louder at 350wrms 12dB up at 1200wrms 0.329cf box.

Below is the modeled output of the Audiobahn (orange trace) Vs the Aliante, a pic of my box in footwell and the double SWR box I did for a customer (budget restraints meant basic box!)
 

Attachments

#35 ·
I noticed your results with harmonic distortion. What level were you playing these? 1w/1m? some random voltage? Just curious.

Are you asking to much from this design? You know, them stupid laws of physics... And that idiot Hoffman...:D

4 subs (love the multiple motors) in a 1.3 cubic foot sealed enclosure is just not going to have ultra low output.

Next question is, does the actual performance mirror within reason what you see in WinISD or some other box modeling software?

Reason I ask is if so, could you model your subs in a bigger enclosure? See what the modeled results are. We know that going bigger sealed, reduces power handling but lowers Fs. So is there a happy medium that allows you to have the enhanced low end while keeping the enclosure reasonable in size? Or are you just going to have to bight the bullet and go ported/PR?
 
#36 ·
It was a somewhat unscientific test (I explained a bit on a previous post). However, I plan to spend at least few days truly characterizing them (wasn't able to do that Sunday evening, as it was the day before spring classes). I'm well aware that I'm asking a lot out of this box (or any other box for that matter). The idea is to squeeze everything out of it that I can, since I can't just factor in a bigger box and call it a day. That's where the measurement and planning will come in. Before it gets put into daily use and abuse, it will be fully measured and the subsonic filter and gains adjusted to make sure I don't let the magic blue smoke out.
 
#38 ·
If you went for that 13W7 sealed you'll likely hit much lower than the 4 8s.

FWIW, I modeled the Alpine 8 in WinISD Beta last night because I was curious. It recommended ~ .8 @ 30 hz each sub, so 1.6 for a pair. It models flat to 30 like that actually. I personally wouldn't tune to 30 hz with a big sub, but flat to 30 hz from a little 8 incher is pretty sick.

That said, 30 hz isn't that low, IMO. The W7 will perform far lower than that, even sealed. It has lots of excursion, which is what you want in a sealed sub.
 
#39 ·
As mentioned in my other write-up thread, the box WILL hit down to 20 Hz. The displacement is there (more surface area than a pair of 12's), it's just a matter of getting there cleanly, and I'm pretty sure I'm being more picky than most people would be. I will be rocking the 4 8's for a few months, and heck, I might even fall in love with them. They do play AWFULLY clean above 45 Hz :)
Wondering if the harmonic distortion you're hearing is actually the box flexing. I ask coz with the power you have available (1200 watts), you're only reaching up to 12mm out of 14mm Xmax. 14mm is hit with more or less 1650 watts according to WinISD...
Do you know about the Fathom from JL? It uses a box too small for the sub to perform optimally but uses EQ to boost the low end to where it should be. Seen measurements and they don't suffer from too much harmonic distortion even in the lowest octave...

How is your enclosure braced? I ask coz you have a lot of cone area and not a lot of air inside the enclosure to move freely.
Any wool or similar used inside your enclosure?

Kelvin
 
#40 ·
Wondering if the harmonic distortion you're hearing is actually the box flexing. I ask coz with the power you have available (1200 watts), you're only reaching up to 12mm out of 14mm Xmax. 14mm is hit with more or less 1650 watts according to WinISD...
Do you know about the Fathom from JL? It uses a box too small for the sub to perform optimally but uses EQ to boost the low end to where it should be. Seen measurements and they don't suffer from too much harmonic distortion even in the lowest octave...

How is your enclosure braced? I ask coz you have a lot of cone area and not a lot of air inside the enclosure to move freely.
Any wool or similar used inside your enclosure?
I doubt it was the box flexing, but that would give me an excuse to break out the accelerometer and firmly attach it to the baffle (I'm not sure my hands are calibrated to know exactly how much is too much). It's built entirely from 3/4" MDF (including three airtight inner walls between drivers), glued and stapled I believe (no screws).

They're stuffed, the first three chambers are filled with Stovetop Stuffing, and the fourth with creamy nougat :D



And as far as my initial measurements go -- I think the peer review process has succeed here and invalidated my initial measurements... I will be retesting with a more controlled setup, and posting my results back here.
 
#46 ·
To the OP, I did not mean for the suggestion of the TCs to get this thread off target.

My car,C6 Corvette, though not as small as yours is very space limited also. Especially if I want to still be able to store the targa top.

I used 2- TC Epic 10s because
1)good reviews I had heard on them
2) relatively shallow mounting depth in relation to power handling
3) very small box needed

I am running 1,500 watts at them and they take it in stride (2 - HD 750/1s)

I suggested the 8 to you because in my opinion, you could power the 8s with the same power as I do my 10s and use 2- 8s instead of 4 of them.

This is only a hobby for me so there are far smarter people to listen to on here but maybe model them with your car and see?

Oh, and in my car the 10s get very low with alot of impact, I am thrilled with the outcome.

Thanks,
LionelC
 
#47 ·
And I apologize if I came off sounding snarky as well. If you notice in my sig, I use a TC sub. Love them. It's just that in this case, swapping from 1 8" to an ultra small box 8" will not yield much if anything more in output. This is really a perfect example of how the enclosure dictates performance. It also really shows how in system design there are often compromises that have to be made. Or a hell of a lot of planning and trial and error.
 
#50 ·
parameters that show how good it can work in small sealed enclosures... but you need power for it to really WANK.
Efficiency is terrible? I'm sure you're talking about sensitivity coz the efficiency in the bottom octaves is much better than most 8"s I've used. Do you know about "Hoffman's Iron Law" -

IMO...the point i was trying to make is that these subs are inefficient at converting electrical energy into sound energy compared to other 8' woofers i like better

...and you seem to feel the same in a round about way
 
#51 · (Edited)
I would bet that these two subs (Alpine and TC) are not that different efficiency-wise/sensitivity in comparable enclosures. most same sized woofs are within a db or two of each other. (obviously pro-audio drivers are not to be included. But they come with trade offs of there own.)

Edit: I looked, the Alpine is rated as 83.5 and the TC is 81.2 as far as their sensitivity spec goes.

XMAX is a little different with the TC having 22 followed by the 14 for the Alpine.
 
#64 ·
Im not trying to "sell" you on the TCs. Its just an option. You could just put a new face on the box you already have for 2 10s instead of 4 8s.

Too bad your so far away, this kina stuff is best decided in the garage listening to the cars while drinking a few beers....
Truth!!!:)
 
#68 ·
I saw Skrillex in December and can confirm everything Patrick says above. My pants legs were flapping from all the sub-bass, and could literally feel the floor shake. I'm pretty sure a few pair of britches got crapped in that night :surprised: If you haven't experienced such an event, get out of your mom's basement and do so, right now.. this post will be here when you get back.

Love it or hate it, Dubstep and other fledgling electronic genres of music including Complextro, Moombahcore, and Glitchhop will make your system work harder than just about any other type of music out there. And IMO there is definitely some overlap between the Bass-Head and the SQ junkie.. They are not mutually exclusive. I will wholeheartedly agree that classic rock sounds like **** when the bass is boosted. This is because the information was never there in the first place, and the tonality goes to hell when most "bass boost" circuits increase the bands from 40 to 100 Hz. The genres mentioned above, however, were mastered to include sub- and infra-sonic bass down to DC, and don't need to be boosted at all. So the tonality is still there. Combined with a great three-way setup, and a rear-fill with Logic-7 style phase steering, these tracks sound positively amazing. A quick note about the rear fill.. There's a lot of out-of-phase content mixed into electronic tracks these days -- If you have Logic 7, listen to Rusko - Hold On (feat. Amber Coffman) (Sub Focus Remix), I can't imagine listening to it with only a front stage.

I'd highly recommend looking into the big bass pro subs and note how low they really go... You maybe shocked, it' not that low. and the ones that do go low you may note that you need a metric ****load of them. There's not much of a transfer function is a 15,000 seat shed.