DIY Dayton Titanic MKIII 15"

[DrWho]

Oh I forgot to mention, how about since DrWho is adamant about 12 inch

drivers, to use acoustic elegence's av12 that is now available for 199?

Well for a more apples to apples comparison...try modelling two 12"

titanics versus a single 15" titanic. You get about the same SPL and

lower extension in the same sized cabinet with about the same power

requirements too.

Two AES AV12's in a 5 cubic foot enclosure would outperform the Titanic

15". Same F3 and 1dB more output. (But you're spending twice as much on

the drivers). One AV12 would go in a 2.5 cubic foot enclosure and would

have 6dB less output. But with such a small cabinet you would be forced

to go with PR's.

Btw, a port doesn't need to reside entirely in the subwoofer

enclosure....you can have "smoke stacks" that extend out and beyond.

Depending on your room you might be able to hide such a thing - in

which case there is nothing wrong with such an approach.

[Jay481985]

obviously we both have too much time on our hands as were posting post

after post. oh well. How is Michael Hurd doing? He said he had to take

time off from the forum.

your temperature effect gives me an idea. Did anyone ever actively

tried to cool a subwoofer? I know like many fans, the motor actually

moves air through the coil to keep it somewhat cool. What if someone

did a watercooled approach like cpu's? and drag the lines outside the

box where there can be an active cooling system or so forth. I remember

bose once claiming liquid cooled tweeters. I know different size and

bad speaker company but, no one ever tried did they?

Also I have my master subwoofer using the same two tumult 15d2 in a

different enclosure. DrWho, try modeling two tumult 15d2 in a 16 cubic

feet enclosure, I was thinking 24x24x60 using the height theory and

Kyle Richardson's Everest design but bigger and six stryke 18 inch 2500

gram passive radiators instead of 4. I believe from what I can remember

a f3 of 17 hertz, pretty much flat +- 1 db from 17-200 hertz. Able to

reach 123 db anechoic in that range too.

[Jay481985]

Btw, a port doesn't need to reside entirely in the subwoofer

enclosure....you can have "smoke stacks" that extend out and beyond.

Depending on your room you might be able to hide such a thing - in

which case there is nothing wrong with such an approach.

I know I know and been told that many times but my brain and eyes would

not like to see a mini industrial factory in my own room, I live in New

Jersey which is notorious for smokestack buildings along the nj

turnpike or garden state parkway..... smokestack does not equal

acceptable in my eyes and practically every subwoofer that is mass

marketed. Heck SVS tubes get enough heat for being a mini waterheater

[DrWho]

You only need 2 18" PR's per driver. So two tumults in a 16 cubic foot

cabinet requires 4 PR's and you can get 123dB with an F3 of 18.5Hz or

120dB with an F3 of 14Hz after EQ.

Mike and I were just talking about cooling subwoofers tonite...apparently some ppl have tried fluid cooling too.

[DrWho]

I know I know and been told that many times but my brain and eyes would

not like to see a mini industrial factory in my own room, I live in New

Jersey which is notorious for smokestack buildings along the nj

turnpike or garden state parkway..... smokestack does not equal

acceptable in my eyes and practically every subwoofer that is mass

marketed. Heck SVS tubes get enough heat for being a mini waterheater

But if you have curtains or the subwoofer near a couch you can send the

"smokestack" behind where it's no lonver visible. It's just nice to

know that option is there - I've only seen it used in one situation and

it was out in the open and very ugly looking, lol.

[Jay481985]

Which too me I believe is very impressive with just two speakers and a not soo insanely huge box.

Edit: I forgot to add that I forgot exactly what but the added two pr

did lower the response or such. I have to model both of them tommarrow

and my desktop with the winisd is off but my roommate is sleeping. 19

inch crt is very bright at night.... I am on my laptop with the monitor

on lowest setting.

As with cooling I am not sure exactly where to actively cool? The magnets? the motor itself? the voicecoil?

[Jay481985]

I know I know and been told that many times but my brain and eyes would

not like to see a mini industrial factory in my own room, I live in New

Jersey which is notorious for smokestack buildings along the nj

turnpike or garden state parkway..... smokestack does not equal

acceptable in my eyes and practically every subwoofer that is mass

marketed. Heck SVS tubes get enough heat for being a mini waterheater

But if you have curtains or the subwoofer near a couch you can send the

"smokestack" behind where it's no lonver visible. It's just nice to

know that option is there - I've only seen it used in one situation and

it was out in the open and very ugly looking, lol.

The only semi smokestack subwoofer that I liked was that ultra low

hitting subwoofer by that rather famous engineer. I believe the

subwoofer could hit 130 db using a 12 inch driver. That was the system

that took up about one square foot but was 12 feet high. Though its

semi hideable....

[DrWho]

I think you're talking about the Danley DTS or something tapped horn

subwoofer....it uses the rear wave of the driver to correct for what

would otherwise be a very jagged response from having a LF horn whose

mouth was too small. Basically two horns where their responses added up

to an end flat response - way cool design.

[Jay481985]

yes its the danley dts. Can you elaborate more on the principle?

[Jay481985]

I think I get it after looking at the picture. BTW sorry zealot for

turning your thread into a drwho/me discussion of subwoofers but I hope

you are jotting notes of different ideas. Basically what I see from the

picture is two ports that are horn shaped internally? and that one port

is for the front and one port is for the back?. A more complex 6th

order bandpass system in very basic concept?

[Remotia]

expect great things from this driver.

http://www.tcsounds.com/Images/LM_news.pdf<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" />

Its not offically out, but Eclipse is already adopting the technology and Im damn sure SVS will change at least their Ultra woofers very soon.

With the new voice coil the new driver is able to increase the SPL by about 6+ dB while maintaining the same THD - yes, very impresssive.

[Jay481985]

You only need 2 18" PR's per driver. So two tumults in a 16 cubic foot

cabinet requires 4 PR's and you can get 123dB with an F3 of 18.5Hz or

120dB with an F3 of 14Hz after EQ.

Mike and I were just talking about cooling subwoofers tonite...apparently some ppl have tried fluid cooling too.

I tend to like the linear graph as shown below with 6 pr instead of 4.

It does not hit as low but the f3 is now pushed 3 hertz down and is

flatter. Although it would cost 300 more dollars for two more passive

radiators.

[DrWho]

6 PR's cost a lot more than 4....and you can achieve any tuning point

you want as long as you use the right passives. The only reason to use

more than one PR is because it reduces the amount of cone excursion

needed for the same SPL (due to the increase in surface area). Sure,

multiple PR's affect the tuning point too, but there are better ways to

achieve that. You don't need any more than dual 18's per 15" active

because they will have no problems displacing enough air (even with a

very low 14Hz tuning).

Btw, the QMS for the 3.36 Fs 18" PR is 84.53, not 92

http://yellow.mynethost.com/~bv126368/shop/catalog/product_info.php?cPath=24&products_id=39&osCsid=34fbf0b9f3d5374f3003edbc95b8cf27

And here is an alternative 18" PR depending on your intended tuning point:

http://yellow.mynethost.com/~bv126368/shop/catalog/product_info.php?cPath=24&products_id=38&osCsid=34fbf0b9f3d5374f3003edbc95b8cf27

With EQ (2nd order peaking highpass filteer) I'm looking at an F3 and

Fb of 14Hz with the 4PR design (2PR per driver). Voice coil temp rise

doesn't hurt the response much at all and you can get a clean 119dB in

the process.

Without EQ I seem to be getting the same plot as your purple one with only 2PR's per driver (so 4 total).

[Jay481985]

oh I got the QMS at kyle's website here

I prefer to have as flat a subwoofer trying to not actively using eq. I

think as a diyer you are able to manipulate the enclosure to your specs

and eq should be the last thing considered and the enclosure should try

to be as flat and low on it's own

[DrWho]

I prefer to have as flat a subwoofer trying to not actively using eq. I

think as a diyer you are able to manipulate the enclosure to your specs

and eq should be the last thing considered and the enclosure should try

to be as flat and low on it's own

DIY'ers like "dB," Danley, and Tom Nousainne (one of these days I'll

learn how to spell their names, lol) would entirely disagree with that

statement. When chatting online they say one of their pet

peaves with the DIY world is their lack of understanding of the system

as a whole. Sadly I'm also included in that group of the DIY world that

isn't aware of the entire system, but at least I understand where I am

and know where I need to learn...now to find someone willing to reveal

all the secrets [] Anyways, the use of filters does a lot of

things....it lowers distortion, it controls cone excursion, it can

compensate for large signal nonlinearities, it allows for deeper

extension in the same cabinet without sacrificing SPL, etc etc....the

model we see in winISD is merely the small signal behavior of the

driver. There is a lot more at play here - mechanical compressions,

thermal compressions, the temperature of the voice coil changes the

electric properties of the motor, etc etc...

For example (and keep in mind this is just the tip of the iceberg of

what I've learned so far)....the peaking 2nd order high pass filter was

designed to take advantage of the fact that the cone excursion of the

active driver is much lower at the tuning point of the system. And then

it also behaves as a high pass filter which prevents lower frequencies

which would totally throw the driver out of the box. Look on your

models and observe the cone excursion behavior below the tuning

point....it goes through the roof and lands the driver on the moon. It

is extremely dangerous to operate a subwoofer near it's peak at

frequencies above the tuning point because all it'd take is one low

note to destroy your system. Usually we're not listening at levels to

cause instant failure, but it's still a bad design flaw.

To see for yourself, model a subwoofer to be as flat as possible (using

the technique I described with the dayton). Now instead of adding EQ to

model the amplifier and all that, add a 2nd order high pass at the

tuning frequency with 3dB of gain (so around 24Hz?) Now lower the

tuning point until you get back to a flat response. You have just

extended the F3 point down about half an octave AND you will notice

better control on the cone excursion. Readjust your power levels to get

the excursion back to normal and then notice that you've sacrificed

maybe 2dB of output. Different drivers aren't as excursion limited and

in their case you don't sacrifice anything.

To take it a step further....notice how the voice coil temperature rise

affects mostly the excursion peak above the tuning point...and results

in a boost at those frequencies. If we implemented a filter of some

kind that brought those frequencies down as the power levels increased,

it would both increase power handling and help to maintain a flat

response. It will also increase the low frequency extension at the same

time...provided your enclosure was tuned properly to anticipate this.

How often do you get to hear that compression increases the volume? lol

Well I have to go to class now, but I would love to talk more about

this subject...I'm definetly not an expert but at least I've been

pointed in the right direction by the true experts. Btw, are you on any

messenging programs? It'd probably be a lot more productive to talk

outside the forum - unless of course everyone else is finding this

interesting....we're totally way off topic by now though []

Anyways, the end conclusion is that there are absolutely no downsides

to using EQ - it can improve the response of every system...even if you

start off from the beginning with no sacrifices on the cabinet design.

You might need a slightly more powerful amplifier, but that's not a huge

deal.

[Jay481985]

heh zealot125 must be like god I wanted help and these people are psychos.....

MSN or AIM

Jay481985@hotmail.com

Jay481985 AIM

Also for the eq and the cabinet design etc, it's like a racecar, some

start from the engine (speaker) and make the chasis (enclosure) or some

start with the chasis and make the engine as a derivative of what the

chasis can deliver.... no one principle idea is better just different.

Just like how subwoofers are seal, vented, passive, 4th order, 6th

order, horned, etc etc....

DrWho how is you cad: autocad or solidworks? I used both and have some

experience with cam. We should buy a CNC router [6] imagine what kinda

speakers we could design! I heard aluminum is great!

Just a tibit, honda spent over 200 million dollars on R&D on oval

pistons for F1 or Indy or Cart I forgot but it really got nowhere

(hence no oval pistons) but I bet they learned alot.

[Zealot125]

Guys, this thread has turned out to be a fantastic wealth of knowledge - and some of the most fun reading that I have done in a week! Thanks a million for your dedication to accurate subwoofer construction. Only one little problem. I know we modeled the FR with that parametric EQ at 20 hz with a Q of 2 +3 dB, but the parametric eq on the sub only allows a Q between .1 and 1. So, I changed the Q to 1 and the FR seemed to improve even more. Is there something else I need to change or is this right?

EDIT with a gain of 4 dB the FR improves even further, and with a 20 x 1.5 inch 25 inch long port, port air velocity is still about 28 m/s.

[DrWho]

My autocad or any 3D rendering ability sucks hardcore...I just haven't

gotten through that steep learning curve yet (and I'm trying to use

free programs). So I just use MS word and scale things down and

manually construct different views using a little bit of math.

As far as EQ goes...just keep an eye out on the cone excursion. The

wider the Q, the less honed in you get on the tuning point (where you

have extra cone excursion available). Sounds like I wasn't doing my

homework completely...though there is some ambiguity with their

description of the EQ section (because they have their description of Q

backwards in their PDF). Also, as you increase the EQ keep an eye out

on the apparent amplifier load. You have to make sure the apparent load

doesn't exceed the capabilities of the amp or the thermal limits of the

driver.

[Jay481985]

Also remember as most amps start to distort/clip bad as you near max amplification.

What wattage were you at Zealot125 with the 28.8 m/s reading? was that

near or at max thermal wattage of either the amp or speaker?

DrWho, autocad is the defacto standard and you should really really learn it. I believe I have some pdf books on it.

QTC of 2!!! wow ummm I doubt its .2 as that means that it would not

drop off until like 1 hertz.... 2.0 QTC must have some crazy drop

off rates.

BTW thanks zealot for your kind words. im me if you need any help or want me to make some autocad drawings for you.

[Zealot125]

will do buddy, but i gotta install trillian first, but its late and I have some differential equations to solve for homework, so that is a project for another day.