Woofer Physics

[fredk]

The woofer in the svs is a monster more on the tc2000/revo level

I guess I hadn't thought about that.

Thanks for replying here Chris. From what you wrote on the other forum, the Kappa 12 just isn't going to do what I want it to, regardless of how good a driver it is. Thanks also for the suggestion of the JL 12 W7, I wish I could afford it.

So, it looks as though, if I want to reach down to 15Hz on the budget I have (200-300CDN) for a driver, I will have to compromise on the linearity/performance of the driver.

I am curious to know what all of you would recommend in my situation.

[Blaser]

Linearity and performance will be highly dependant on the enclosure design as well!
Better to have an average driver in a good/well designed enclosure than a perfect driver in an inappropriate box.
That is why modeling is very important.

The amp driving the sub will be also very important. Get the best you can and run it slowly.

At the end, your room might not be perfect as far as modes and resonance, decay .... are concerned, as you see the driver is important but SQ will be also dependent on many other factors!

[WmAx]

fredk wrote:

I guess I hadn't thought about that.

Thanks for replying here Chris. From what you wrote on the other forum, the Kappa 12 just isn't going to do what I want it to, regardless of how good a driver it is. Thanks also for the suggestion of the JL 12 W7, I wish I could afford it.

So, it looks as though, if I want to reach down to 15Hz on the budget I have (200-300CDN) for a driver, I will have to compromise on the linearity/performance of the driver.

I am curious to know what all of you would recommend in my situation.

It's so hard to figure out because of your budget.

If you could deal with a 10 to 11 net internal volume cabinet with a folded slot port, the Dayton RS 15" driver could be used with considerable output(115dB or so in room, real world) down to 13 Hz or so. Can you deal with that size cabinet? My in room SPL estimates are presuming you have an average size room(2000 cubic feet or less), that is closed(not large open corridors to rest of house) and that you will corner load the sub.

You could use the same driver in a 5 cubic foot sealed cabinet and a parametric EQ band to compensate for LF loss, but output ability would be vastly reduced in the lowest octave that you consider so important.

-Chris

[SteveCallas]

wmax wrote:

I would rather have 14mm of extremely linear travel than the non-linear 25mm of the typical SPL sub that has substantial BL change after just a few mm of excursion

You must be talking about something like a LMS Ultra then, because the Kappa does not appear to be extremely linear. Measurements show that it isn't anything more than a standard 12".

Sonnie broke it down earlier for you:

Sonnie wrote:

Figure $275 for a Behringer EP2500 and $100 for construction materials... you have $200-$300 left for a driver, which should allow you several choices.

You can get a Mach5 IXL 15/18, A FiCar Audio Q15/18, a Dayton Reference HiFi15, and so on and so forth.

[WmAx]

SteveCallas wrote:

You must be talking about something like a LMS Ultra then, because the Kappa does not appear to be extremely linear. Measurements show that it isn't anything more than a standard 12".

If you wish to continue to assume that the link at hand, which used the driver with severely under powered amplifier and grossly improper cabinet is a fair representation, then by all means, continue to believe so.... but under 200 or so watts that amp was delivering and the fact the driver is completely ill suited to sealed cabinets, makes the test completely worthless in regards to the applications that I mention. Besides my use/verification, and avaserfi's measured performance, several people have built avaserfi's cabinet and used a high power amp, and found this subwoofer to easily exceed costly high end commercial 12" designs and provide immense SPLs at 20Hz on up compared to commercial designs. The avaserfi design, for example, will easily exceed a JL Fathom F113 overall, which is considered a superb subwoofer system. The F113 may have longer stroke and an inch gain in diameter, but it's use in a sealed system vastly reduces it's effect lower octave output.


-Chris

[fredk]

Blaser:
1. With all the help I will get here, how can I end up with anything but a brilliantly designed cabinet? I also happen to be reasonably handy.

2. I am also working on my room, but being in a concrete box doesn't help.

3. I will be carefull to give whatever driver I choose the power it needs for the designed enclosure.

4. Yes, I am avoiding modeling drivers (for now). Winisd confused the out of me the last time I tried it.

[fredk]

Chris. I could probably live with something in that size range. In fact, I was looking at a couple of designs that would turn the sub into a statement, after all, its hard to hide an elephant in a small room. I do need to figure out what the maximum I can live with is though.

Steve. I did take note of Sonnie's most helpful posts at the beginning of this thread.

You can get a Mach5 IXL 15/18, A FiCar Audio Q15/18, a Dayton Reference HiFi15, and so on and so forth.

To rephrase, you are telling me that these drivers are all the same performance wise? (just confirming, the pedant in me finds it hard to believe they are all the same).

I can pick up the Mach5 ixl 18 for $265 delivered. They list their prices in Canadian and US as almost the same so that driver seems like the best deal for me at the moment. For the most part, the 49th parallel seems to be a very expensive line to cross.

As a comparison, the Kappa 12 is listed at around $220 delivered in Canada.

[Ricci]

WmAx wrote:

The avaserfi design, for example, will easily exceed a JL Fathom F113 overall, which is considered a superb subwoofer system. The F113 may have longer stroke and an inch gain in diameter, but it's use in a sealed system vastly reduces it's effect lower octave output.


-Chris

Andrew's design is nice and I'm sure it sounds great but I would like to see any kind of hard data that supports this. The driver in the F113 has twice the linear displacement of the Kappa at the least. You are comparing a vented cab to a sealed one. I would bet that the F113 has more output above 30-35hz and significantly more below the tuning of Andrew's cab as well even in the sealed cab.

Ditto on everything Steve has already said. The kappa is a nice driver but it is not anything spectacular in the linearity/ displacement department. I've never seen any testing on it that indicated that it was.There are plenty of other 12" drivers that are comparable in the high SQ, medium duty segment. Dayton RS series, Peerless XLS, JBL W12GTI, Aura NS12, etc. Klippel testing has been done on all of those and they have comparable capabilities.

[WmAx]

Ricci wrote:

Andrew's design is nice and I'm sure it sounds great but I would like to see any kind of hard data that supports this. The driver in the F113 has twice the linear displacement of the Kappa at the least. You are comparing a vented cab to a sealed one. I would bet that the F113 has more output above 30-35hz and significantly more below the tuning of Andrew's cab as well even in the sealed cab.

Yes, at about 35Hz, the Fathom starts to increase in maximum output as compared to the Kappa Perfect, with at least 4dB higher output to be expected. The JL driver has over 2x the thermal capacity of the Kappa Perfect in addition to the displacement advantage.

-Chris

[SteveCallas]

Ricci wrote:

Andrew's design is nice and I'm sure it sounds great but I would like to see any kind of hard data that supports this

You know how it is on audio forums, anybody can and will say anything - and in the process, they convince a few people along the way. Then when the real data comes to light, the backpeddling begins.

[fredk]

Klippel testing has been done on all of those and they have comparable capabilities.

For whatever reason, this is the first time I have seen this testing mentioned. A search last night brought up some tantelizing reading, but leaves me with more questions.

Is there other reading you folks would recommend or other terms I can search on that will help me understand the topic of driver quality/performance?

[robertcharles123]

From what your needs are and the 500-700 range you gave, you could easily build a quality unit like sonnie layed out previous. The amp is going to be the most expensive part of your system if you do not go with the ep2500. Based on the specs and the other experience of the other guys on this forum the mach 5 drivers will give you all you need. The CSS would do the same. All are in the mid 200's. Try elemental designs.

[fredk]

What a coincidence that you poste here. I just finished reading your amazing build thread. Your cabinets are giving me some ideas for styling/look. If I go big, the sub should make a statement.

It may well be that any of the drivers already mentioned will be fine for me, but I need to go through the learning and reading process. It is as important to me as the build and the results.

[Ricci]

fredk wrote:

For whatever reason, this is the first time I have seen this testing mentioned. A search last night brought up some tantelizing reading, but leaves me with more questions.

Is there other reading you folks would recommend or other terms I can search on that will help me understand the topic of driver quality/performance?

It is basically a technical assessment of the driver's abilities in various domains. It has problems with it too and ain't the whole picture either, but it is WAY better than just looking at manufacturer's specs.

A general rule of thumb for me is assume that the driver will do 80% of the performance the manufacturer claims unless given cause to believe otherwise. Also use general deduction and common sense logic to educatedly guess.

You basically have a budget of $500-700 and no size restriction. You want to get deep effortless bass to at-least 16hz and you also want it to sound good with music. Sound about right?

[Sonnie]

To lay the claims and requests to rest concerning Andrew's DIY sub build... I believe his sub will be among the units that will be tested during his first couple of sub testing phases. This is by no means meant to get the thread off track... just FTR and FWIW.

[fredk]

Ricci. That about sums it up. I found out that the lowest stop on large pipe organs is 17Hz, so 16 Hz would be a good target.

Size is not a major constraint, but given my current room, I think 12 cuft. would be the upper limit (more or less) and 8-10 would be ideal.

Sonnie. I'm looking forward to the measurements.

[imbeaujp]

fredk wrote:

I have been lurking here for a number of months trying to get my head around sub design. At some point I am looking to build a sub that will reach down to ~15Hz to accurately reproduce pipe organ music.

I am still not clear on what characteristics in a woofer limit or allow it to reach the lower frequencies. It seems that producers of commercial subs are able to make subs that play flat down to 15Hz, but that dosn't seem to be the case in the DIY world. I am wondering why that is.

From some discussions I have read, it also seems that there are some tradoffs between linearity, headroom, and driver size that leave me wondering whether it is better to choose a smaller driver with better linearity and drive it harder, or a larger driver with lesser linearity and retain more headroom (less risk of non-linear distortion).

Some other details. My rig is in a small room - 20x11x8 - and I do not listen at really high volumes - usually 70-80db. Though I will be using this for HT as well, my primary concern is accuracy.

Budget: as little as possible, but realistically $500-700??

So, how do I go about narrowing down my choices?

Hello Fredk,

DIY sub is a great project to do. I just finish building 2 other subs in my home theater. Now I have 4 subs, 12 inches drivers with passive radiator. I really go down to 16hz with no problems and I can appreciate that in many recordings like "I Love you" from Sara Mclaclan and some movies like Shooter or transformers.

There are many oppinions on a forum. But to me, only one sub is toomuch "directional" I can locate where the sub is in the room. I prefer using multiple sub to extend the sweet spot.

In that configuration, I chose 4 small subs instead of a big one.

Here is my building thread and mesuring with the first 2 subs. Mesuring with the 4 subs have not been post yet but the result is marvelous.

Hope this can help.

JP

[fredk]

Thanks for the link to your build JP. Thats impressive output for 2 12"ers.

OK, I finally downloaded WINisd and did some playing. Somehow, the program made more sense this time around. I guess somthing did sink in from all the reading I have been doing.

I modeled the Kappa 12, Mach5 15 IXL and MJ18 all in 10cuft boxes tuned to 15Hz. Ignoring that I have no clue what I am doing yet, I was interested in general relative performance of a 12 and 15/18" driver at given tunings and volume.

interestingly, for music, up to around 95db it looks like any of these would perform just fine barely breaking a sweat. I was expecting the Kappa 12 to work a lot harder at this volume.

Here are cone excursion and spl graphs for the Kappa 12 and MJ18M:



If I am reading this correctly, both drivers are going to produce around 95-100db. While the Kappa 12 requires more excursion to do this, it is still only at 33% of Xmax.

Add another 5 db and the Kappa tops out at Xmax while the MJ18 is only at 44% of Xmax.

So, a modestly priced 18" woofer is not going to work particularly hard in the same range where a good/pricey 12"er is going to have to be pushed to the limit.

If the above modeling is correct, as long as the MJ18 is a reasonable performer, I should get clean output from it over the same range as that of the capacity, but I have headroom to spare.

[Ricci]

A couple of pointers here Fred.

See how the Infinity is peaking around 20hz? You don't want that if you can help it. It indicates that the cab is too big for the tuning/driver used. Try decreasing the volume of the Kappa box until it looks more like the MJ-18M box, which is a nice looking response. Probably more like 5 or 6ft would be better for the Kappa.

Some other things to remember and these are generalities so don't take them as concrete. The bigger your airspace for your driver to work in the more efficient the low end and the shorter your port will get for a given tune. However it becomes easier to bottom your driver with less power the larger your enclosure gets.

See how the excursion charts have a notch in them near the enclosure tuning and how the cone excursion increases above that? With a properly implemented HPF on your sub this is the area you need to worry about, not what happens below tuning. The lower you tune the more excursion demands are placed on your sub in this area because it moves lower in frequency. See how much power it takes to get your driver close to xmax in that area and then buy an appropriately powerful amp, or choose the amp and driver first and scale the enclosure size and tuning to make the most of the available power.

Also in general the larger your driver, or when you add multiple drivers the bigger the cab is going to need to be to keep a good FR or vent tuning profile. This is why you don't need 10cu ft for the Kappa. That's a huge amount of space for a single vented 12", it's just about right for a big vented 15" and it's on the small end for a vented HT 18".

Try scaling down the Kappa enclosure to get a similar FR to the MJ18M enclosure and then apply enough power to get the cone excursion right at xmax above tuning for both. Then compare the xmax limited SPL graphs for both. It should make things a bit more clear. Perhaps try tuning them both a bit higher as well to 16 or 17hz. This will allow you to still get to 15hz with some authority, but give a shorter port length and also allow a bit more power handling above tuning.

[fredk]

Thanks for the comments Ricci. I didn't think that uptick at the bottom was good. I went back and looked at avaserfi's design and noticed the cabinet was much smaller.

I did figure out the power handling/excursion stuff and with your comments I now understand what the notch is. If I understand correctly, the filter at the notch also protects the driver from unloading at lower frequencies?? In the max power curve I noticed a point at around 10-12Hz where the power requirements for excursion drop off a cliff. I take it this is where the driver is at risk of unloading without a filter?

If I understand the max power graph properly, the notch shows you where you have enough power to exeed Xmax (and possibly Xmech)??

One thing modeling does not tell me is how cabinet size and venting affects the performance of the driver. A larger cabinet allows the driver to move more freely this use much less power for a given amount of travel. I wonder how this affects damping (hope thats the right term)? If a driver is able to start moving more easily, does that mean it will have a harder time stopping (mass vs resistance)? How might this affect the quality of sound?

I keep thinking that there must be some way to take advantage of my lower spl requirements to reduce build costs. It looked like going to the MJ18 could shave $100 off the driver cost and maybe give some additional savings on an amp, but now I need another gizmo to keep the driver from self destructing. Where is my free lunch golly gosh geewhizzikers!!

[Sonnie]

You probably want to use the Kappa VQ version using the Low Q configuration for modeling. I have it at home and will upload it later for you.

Oh yeah... and there is an ED eQ.2 in classifieds for 60 bucks... HPF variable down to 5Hz.

[SteveCallas]

fredk wrote:

A larger cabinet allows the driver to move more freely this use much less power for a given amount of travel. I wonder how this affects damping (hope thats the right term)? If a driver is able to start moving more easily, does that mean it will have a harder time stopping (mass vs resistance)? How might this affect the quality of sound?

A lower Q equates to being more damped. A larger enclosure equates to a lower Q. Therefore, a larger enclosure equates to being more damped.

In a very small enclosure, air compression inside the enclosure is on a higher magnitude than that of a larger enclosure - additionally, heat builds up more quickly. Since air doesn't compress completely linearly when heat is applied, and more power to reach a given excursion level is needed in a smaller enclosure, a smaller enclosure results in more distortion at a given excursion level as compared to a larger enclosure. This has been measured and confirmed by Ilkka.

To Ilkka, the difference was smaller than expected - to me, it was larger than expected. On only the 105db sweep, the THD at 20hz was 10% in the small enclosure vs only 3% in the large enclosure. The difference only got bigger at higher sweeps.

So to answer your question, a larger enclosure equates to better performance for a given driver.....despite what an "audiphile" may tell you.

[fredk]

Thanks for the reply and the link Steve. I don't think I quite understand the usage of damping yet. It makes sense that linear distortion would increase as the power load to move the driver a given distance increases.

I remember reading somewhere that distortion also happens when a driver dosn't stop moving when it is supposed to. Maybe I am using the wrong term(s)? Anyway, if a driver meets less air resistance to movement as box size increases I would think the tendancy to keep moving when its not supposed to would increase.

Links or search terms to help me understand the above would be greatly appreciated.

[fredk]

Thanks for the heads up Sonnie. I have been so focused on drivers, I missed that.

At this point, my understanding of subwoofer design is very shaky so I have no idea where I should spend money first (or at all).

I wonder if I could rig something up with sonotube that would allow me to adjust internal volume and port size so that I could do some real world playing in my room??

[SteveCallas]

fredk wrote:

I remember reading somewhere that distortion also happens when a driver dosn't stop moving when it is supposed to. Maybe I am using the wrong term(s)? Anyway, if a driver meets less air resistance to movement as box size increases I would think the tendancy to keep moving when its not supposed to would increase.

In a small enclsoure with greater resistance to cone movement, you have both the driver's suspension and the linearity of air compression to deal with - in a larger enclosure, just the suspension linearity. Also, in a small enclosure, as the cone moves inward, there is resistance. When the cone then moves back to neutral, that resistance now becomes a force pushing the cone forward to equalize the pressure in the enclosure. As the cone moves out and away, there is resistance again, trying to pull the cone inward. All this additional influence on cone movement has to be factored in addition to the linearity of the driver's suspension - better to get rid of it and just let the driver move freely, relying primarily on the linearity of it's suspension. The net affect is less distortion.

If you want a term to search for, check air spring distortion.