DIY Dayton Titanic MKIII 15"

[Zealot125]

Ok, with my tax refund this year I have decided to build myself a sub. So far, I have decided this:

-MDF cabinet

-Dayton Titanic MKIII 15" driver

-1000 watt Dayton plate amp

-ported design

Questions questions. Here goes.

Is HDF better to use than MDF?

Front or rear ported?

Flared or regular port ends?

Where can I get the materials to construct the port?

Do I need the 1000 watt amp or will 500 watts suffice?

Do I need absorption inside the cabinet?

Do I need to account for the volume of the absorption when calculating cabinet volume?

How do I reinforce the cabinet?

What is the best way to build a cabinet (wood screws, wood glue, epoxy, silicone, etc.)?

Should I seal the seams with caulk or something similar?

Do I need to use wood glue or epoxy at the joints?

When modeling in WinISD, how do I know if port noise will be an issue at high volumes?

If I can come up with any more I will post later, but it would be great if anybody could answer just a couple questions. Also, I tried entering the TS parameters into WinISD but I am not confident that I did so correctly. Can someone email me a correctly configured .wdr file for the Dayton Titanic 15" driver so I can model it (Jetsch@insightbb.com)? Thanks in advance guys.

[Jay481985]

HDF is cost prohibited, around 180 dollars for a 4x4 if you can even

find it most likely need a special order. Lowes sells 3/4 inch mdf for

22 dollars a sheet (4X8). Just make sure its the light brown one not

the darker one which I personally think that they mixed the name and

that is really just fiberboard. Besides you can just double up the mdf.

MDF weighs 52-55 pounds per square foot. HDF weights 62 pounds so per

price per pound it is not really worth it.

Front ported since it gives you better options. A rear ported means

that you cannot place it against a wall. Front port allows you too if

the amp is not in the back.

Stick with flared ports since they reduce turbulance and they allow the

port to be about 5-10% smaller. plus price is negligable.

Parts express sells ports or ends. You can just use a plastic plumbers

pipe to enlarge it. Or I was thinking of using thick cardboard tubes

that they use for mailing.

I would match amp with speaker, even more so I would match a higher

rated amp with a speaker than a amp with less power than the speaker

can take. Which, can equal less clipping. But the Bash amp that is 500

watts does have one benefit, no low pass filter.

You should stuff you subwoofer regardless. Stuffing with dacron

(polyfill or that pillow stuffing) makes the enclosure seems bigger

than it is. It lowers the total Qtc of the box also. Usually you would

go 1 pound per cubic foot but it all depends on port/sealed/pr.

normally no unless you use closed cell foam? but don't forget to take

into account of loss of space due to the physicall driver, and the

port. Everything inside the port does not count as internal volume

By adding braces? can be intricate like mine here or this or just a circle cut out wihin a flat board.

Silicone is meant to make the seals airtight after you use glue. Epoxy

is prohibitively expensive (the real kind that you need to mix).

Gorilla glue (polyurethane glue) is expensive and has alot of mess. I

use wood glue, Titebond type III which is cheaper than gorilla glue and

supposedly stronger. You don't need a gallon, try the quart size. Wood

glue technically is stronger than the wood. If you cut a piece and

glued it and hit it with a hammer the glue part would not fail first.

Wood screws are a problem with mdf. They cause mdf to split very easy

unless properly tapped and countersunk. Another idea is to use void

free (furniture grade no void) baltic birch though it is double the

price of mdf. Expect it to be 45 dollars per 4x4 sheet. though it is as

strong.

Seal the seams, I am not sure you mean the subwoofer or the seams of

the enclosure. If it is the enclosure yes, but make sure the silicon is

urea free, urea is that acid that make it smell horrible (is piss) and

it actually eats away at the subwoofer over time if it is not proper

set and vented before the subwoofer is in place. If it is between the

subwoofer and enclosure I would say no. Use insulation sealing tape or

they sell speaker insulation tape on partsexpress.com

Wood glue is fine, epoxy expensive.

If it is past 5% of the speed of sound or 17 m/s but pro speakers

usually hit 20-25. At insane db your not going to hear it as bad.

Another post I believe DrWho said the Klipsch kw-120 hit 40m/s

The dayton 15 inch driver should already be in the file list.

[Zealot125]

I looked in the list but still did not see the entry for the dayton subwoofer. I am using the alpha version, should I be using the older one? Also, what is the best way to secure the double layer of MDF together?

Also, when speaking of the diameter of the flared port, are we talking about the narrow diameter in the middle, or the flared diameter of the ends?

[DrWho]

I don't think the dayton drivers were included with my version...I

emailed you all three though and the IB, and I can send a few others

too if you're interested.

What is your budget on this sub and are you looking to build just a single enclosure? (ie, no stereo setup?)

The reason I ask is because you would get far better performance going

with dual PR's instead of a port - but they also cost more money too

One of the cool things about DIY is you can custom build the enclosure

to your listening environment - so odd shapes and whatever can be

beneficial.

For a single driver you're looking at spending

Dayton Tiantic mkIII 15" $175

AES 2 x 15" PR $90

Dayton 1000W AMP $400

TOTAL PARTS $665

(not including shipping or tax)

+ the cost of the enclosure which is ~ $200 if you hire out the work and get it finished real nice.

So you're looking at around $1000 for a very nice performing unit.

Why don't you try an enclosure like the new RT-12d? Just a touch bigger to support the three 15" drivers.

[Zealot125]

That all sounds fantastic, but this is my first endeavor into sub building and I want to keep it simple. One of my goals is to learn some woodworking skills as well. Unfortunately, my budget maxes out at about $550-$600, which fits this particular project well. I am pretty set on using this driver with a bas reflex enclosure, but I am also up for suggestions if there is a way to get more bang for my buck.

[rayok275]

Hi,

I have the Dayton Titanic MKIII kit from Parts Express. The

sealed box really performs well for both music and home theater.

I would suggest that you post the same question on the PE tech

board. Those guys build subs all the time and can really steer

you in the right direction.

Ray

[Zealot125]

I am messing with some filters on WinISD. Is room gain 3 or 6 dB gain?

[Jay481985]

usually 6-12 db.

flared port is measured in the smallest part. A 6in port has 9.5 inch ends 5.75 inch internal diameter.

btw you have a pm

[Jay481985]

I looked in the list but still did not see the

entry for the dayton subwoofer. I am using the alpha version,

should I be using the older one? Also, what is the best way to

secure the double layer of MDF together?

[Jay481985]

or do not mind my trebuchet (11th grade state finalist for principles

of technology demostration of a lever. It can toss a mouseball 300 feet

with 20 pounds of counterweight. BTW that is all solid oak [] or lead

or metal rods.)

In that picture you can see about 44 pounds a brick times 8 bricks so

352 lb of weight. Its ontop of cardboard to prevent damaging the wood

and hopefully sorta evening the weight around. also you can see a line

around the cardboard (yellow) that is a tie down with a rachet. On the

edges there are L pieces of wood (scrap) that prevent the pieces of

wood to shift. With that, and the tie down forcing the two pieces to

merge (same size boards) and the weight produces pretty dead on 1.5

inch wood. Also I used Titebond III. It has good inital tack and sets

in 7 minutes if I remmeber correctly so you have some fudge factor.

[Jay481985]

this is my baffle, 2.25 inches thick and it is pretty spot on using my

method. BTW the ruler is bend cause one hand ruler another on camera

[DrWho]

I am messing with some filters on WinISD. Is room gain 3 or 6 dB gain?

Don't worry about room gain - both from corner loading or pressurizing the room ("cabin gain").

First of all, corner loading emphasizes all the frequencies - maybe a

little boost in the low end, but nothing to worry yourself about. And

the cabin gain of the room is a natural thing that you associate with

sounds happening in the room....so if their was a kick drum in your

room, it would benefit from the same cabin gain as your subwoofer - if

you make the sub flat with room gain, then you end up with a system

lacking a realistic reproduction.

Anyways, here is my ideal cabinet for the Dayton using that same amp - and since you're learning I'll describe how I got there.

So starting with the dayton subwoofer - just click through the menus making sure you clicked the ported version.

Ok, now make the cabinet size way too big, say 12 cubic feet for now.

And now set the tuning point way too low, like 15Hz. Now click in the

tuning point box (so that the cursor is flashing in there) and then

press the up arrow on your keyboard. Move it up until that dip in the

middle flattens out with the higher frequencies. Now reduce the cabinet

size. Sometimes when you reduce the cabinet size you'll see it set your

tuning point to zero - this is because you end up with a negative

length for your ports (silly error with the algorithm). go back to your

ports and pick 8 for the number - we'll fix this later. Ok, so reduce

the cabinet size until you get rid of the peak in the low end. You

should end up with a 7 cubic foot cabinet tuned to 24Hz. This will be

the tuning that has the best transient response.

Ok, now we need to set your power levels for the sub. So go to the

maximum power graph and note the lowest of the highest power level. You

will notice a flat line of 800Hz until about 23Hz where it starts to

drop off (because you are now below the tuning point of the system).

For the sake of example, change your cabinet size to 17 cubic feet and

look at the graph...notice a huge dip centered around 31Hz...depending

on the cabinet/tuning you end up will want to use the value at the

bottom of this dip. So enter 800 watts for the power (signal tab on the

driver window) and then go to the cone excursion graph (without fixing

the cabinet size). You will notice that the driver is going 2mm over

the maximum excursion and you will see how the tuning point lowers the

cone excursion for a bit too. Now change the cabinet size back to 7

cubic feet and notice the change in cone excursion. Now if we go to the

SPL graph we will be looking at the maxSPL.

Ok, now it's time to model your subwoofer amp. The first thing we must

take into account is that it has a slight rolloff. To save you a bit of

work just go to the EQ/filter tab and add a highpass butterworth filter

with an Order of 3 and a cutoff of 18Hz (you can ignore the "Q" on this

kind of filter). If you want to see the effect of your EQ, you can go

to the "transfer function magnitude (EQ/Filter)" all the way at the

bottom of the menu (under "filter/equalizer"). You will notice that the

graph looks very much like the graph that you'll find on the PE website.

Ok, so go back to the driver response curve and notice how the EQ

totally knocks out the low end. And then go back to the cone excursion

graph and notice how the EQ controls the excursion below the tuning

point (this is a good thing).

One cool feature with that amplifier is that it has a parametric EQ on

it...which we can use to further extend the LF response. One thing

you'll notice is that if you add EQ at the tuning frequency, that you

have a lot of cone excursion available. But since our response is

already "flat" - let's move the tuning point down so that when we add

3dB of EQ at that frequency we end up with a flat response. Since I've

already done the work for you, go ahead and change the tuning down to

20Hz now. And then add a parametric EQ at 20Hz with 3dB of gain and a Q

of 2. Now go back to your cone excursion chart and what do you notice?

The maximum excursion above and below the tuning point are equal. Now

all we need to do is lower the power level so that we're not running

into over-excursion (500 watts is where you should end up).

Ok, we're not done yet...because we added some EQ we need to go back to

the "apparent amplifier load power" graph. Our driver is only capable

of handling 800 watts and our amplifier can only put out 1000watts into

a 4ohm load (our driver is a 4 ohm driver). A quick check and we notice

that we're looking at about 520 watts - so not a big deal with this

driver.

Now we can go back to our SPL graphs and note the maximum SPL (117dB @

70Hz) and then our system has an F3 of 20Hz. When on the SPL graph, you

can double click at any point and then move the cursor to get a

relative read out. So if you double click at 70Hz, you will see the

window change to "0.000 rel dB" and "0.00 rel Hz" - now just move your

cursor to the left until it reads -3dB and then double click again to

see what frequency you're at. I think this is one of the way cool

features of the program. If you want, you can find the -6dB point

(18Hz) to get an idea of how your sub compares to the measurements

klipsch posts on their website (+-3dB is a total of 6dB).

So, now the fun part of finding the optimum port size and number.

You're already looking at a 7 cubic foot cabinet so we need to get an

idea of the shape your subwoofer is going to be in. And then we can go

about constructing the port around that. I personally like to go with

slot ports because they're easy to build and easy to calculate their

volume. It is important to note that you can't include the volume of

the port in the volume of your cabinet...so if your port takes a cubic

foot, then your entire enclosure will be 8 cubic feet (or a bit more

considering you have an amp and speaker to put in there too). So to

change to a slot type port, just click on the shape button (it looks

like a circle and will change into a square). Let's assume the front

face of your cabinet will be 18" wide...so let's go with an 18" wide 2"

tall port. Now change the number of ports back to 1. And change the

"end correction" to "two free ends." If you were using a flared port

kit, then you would enter how many flared ends the port has....and

you'll notice that the length of the port changes. If we go with an 18"

wide cabinet then it will need to be about 24" tall and 36" deep (which

yields 9 cubic feet, i know). In such a scenario our port would just

barely fit and we'd be at a port velocity around 24m/s. Not too

bad....but since our cabinet is going to be a bit smaller why don't we

go with a 1.5" tall port. This gives us a port velocity of 32m/s and a

port length of 22" which is quite a bit more doable. Just for kicks,

reduce the power to the driver down to 250 watts and notice how the

port velocity drops down to 22m/s, which if we go back to the graph

corresponds to 114dB of output.

So there you go. When building your cabinet just be sure to double up

on the front baffle that the driver is mounted on (or triple up, lol).

Also be sure to implement a lot of cross bracing...narrow strips of

wood going between the center faces of panel sides should be enough (or

put a few strips evenly spread out). The idea is to minimize the size

of a free radiating surface - which will then push the resonate

frequency up above the range of the woofer. And lastly, keep the

cabinet dimensions different from each other (and don't do multiples

either). This helps to avoid harmonic amplification, which will reduce

any muddiness.

Well that's about it on the design stages...now you have to go build it (an area that I'm not very good at). Hope this helps.

So in summary, my ideal version for the Dayton Titanic mkIII 15" with the Dayton 1000W amplifier is as follows:

7 cubic foot cabinet tuned to 20Hz with a 3dB boost (Q = 2) at 20Hz

(yielding an F3 of 20Hz and a system that it +-3dB from 18Hz up to

around 200Hz). To model the frequency response of the amplifier we are

using a butterworth filter with an 18Hz cuttoff and an order of 3. And

then we are expecting to draw 500W from the amplifier for a maxSPL of

117dB.

There, that wasn't so complicated now was it? []

[Jay481985]

DrWho you know your jealous of my triple layered baffle []

But again DrWho its easy to have the optimal subwoofer on computers but as you said

"now you have to go build it (an area that I'm not very good at)"

7 cubic box is neither room friendly, moving friendly, building

friendly, etc etc 7 cubic feet is roughly 24x24x24 in smallest

displacement. not including the 1 cubic foot port size. How long was

the port and is it going to be bent in order to fit?

[DrWho]

A rough port size came out to be 1.5" tall nad 18" wide and 22" deep

(that's what, 0.34 cubic feet...let's just call the port, driver and

amp all .5 cubic feet). If you made the enclosure 18" wide and then 24"

tall (including the port), then you're looking at a 30" deep cabinet -

which leaves over 6" of space from the inside mouth of the port to the

nearest wall (which is more than enough).

I know it's a "big" cabinet but it's the ideal for the dayton. And

really, anything under 8 cubic feet is considered small in my books

[] If you really wanted to go even smaller, then you're going to have

to move to PR's since you won't be able to fit any decent sized port in

the cabinet. And then you're looking at spending more on the driver

itself (as it will have to have far more power handling ability). It's

just a downward spiral of more cost for the same performance...but if

you can't fit the larger cabinet, then that's the route you have to

take.

For this same 7 cubic feet, you could totally get away with a triangle

shape like the RT-12d and tuck it into the corner (which saves a ton on

useable floor space). Though this will also require some modification

of the port dimensions. I even wrote myself a nifty excel program to

calculate the dimensions of a triangular box...both an equilateral and

isoceles right triangle.

oh i almost forgot....when doing multiple layers for the baffle, be

sure to use lots of wood glue inbetween (and even rough up the surfaces

a bit with heavy grit sand paper). Wood glue creates a chemical bond

with the wood when it dries and this will double the effectiveness of

doubling the baffle (making it seem twice as thick as it really is). I

might be exagerating the benefits of glue a bit, but all I know is that

the stuff is insanely cool. And be sure to use lots of it so it can

soak into the wood (and apply lots of pressure because pressure creates

a tighter bond too).

[Jay481985]

The glue thing I think your not looking at the other perspective. If it

floats you have a problem, its hard without my method or similiar to

get two piece of wood to rest ontop of each other without it slipping

out. Glue tends to be rather dense and does cause the two pieces of

wood to slide very easily until the glue tacks up. As with veneerers,

you wait pretty much until the last moments of tackiness and place on.

Also if your drenching the glue like DrWho says, expect a full week of

curing since most glue manufactures say give atleast 24 hours dry time

but prefereable 3 days. Hell if you ever watched the New Yankee

Workshop with Norm Abrem (sp) he sometimes lets his glue set for a

week. Also it becomes a mess when you add pressure with overflow of

glue. It spills out the edges and goes all over.... Also the very glue

benefits that DrWho says will dissapate as most of the glue is now run

off. I would say there is there is a point when too much pressure or

too much glue is worse off. I brought the one gallon titebond III at

lowes for 27 dollars and still have not used 1/2 the bottle, but

anyways I spilled what I thought was enough glue on both pieces of mdf.

Then I had a 11 paint roller with a special glue roller. It looks like

carpet and causes a texture on the glue to get a better and less

spillage coverage. Also keep a bucket of warm water, even though

titebond type III is waterproof, it is not until it cures. So if you

don't mind spending 6 dollars for that glue roller, you will buy alot

since I had to do each side one by one since I didn't have that many

cinder blocks nor space etc etc.

Also I think DrWho thinks too exotic... Thinking more New Yorker, if

you have limited space, build up. The Everest subwoofer by Kyle

Richardson takes no more space than an RF7 floor space but is 10 cubic

feet. It also looks like a floorstanding speaker.

I remember Wilson Audio talks of using different glue for the enclosure

do to the elasticity properties of glues and they actually can enhancce

or detract from the quality. But that most likely is not the case but

they make 100000 dollar speakers....

Also another nifty trick with glue that helps with reflections and

such; Get a gallon size of that cheapo elmers white glue in preschool

(around 6 dollars) and add lots of sand and small gravel to make a

slurry paste. Then spread that paste all over the inside of the speaker

creating a rough sidewall that reduces reflection and actually

increases the rigidiness of the wood as per DrWho said with the

benefits of glue.

[DrWho]

lol - how'd you know I get glue all over the place? [][] Seriously,

building is not my strong point...I can think of how I would go about

doing it, but I never have the right tools and after a few hours of

using a jigsaw you start to get sloppy (yes, I use a jigsaw to cut all

my pieces). I also never wait long enough for the glue to set and so I

end up having to reglue or even recut a piece because the glue dried

and preventing me from using more glue. But I'm still a young guy so

maybe after a few more projects I'll have gained more experience (hah,

right).

Also I think DrWho thinks too exotic... Thinking more New Yorker, if

you have limited space, build up. The Everest subwoofer by Kyle

Richardson takes no more space than an RF7 floor space but is 10 cubic

feet. It also looks like a floorstanding speaker.

I remember Wilson Audio talks of using different glue for the enclosure

do to the elasticity properties of glues and they actually can enhancce

or detract from the quality. But that most likely is not the case but

they make 100000 dollar speakers....

And there's nothing wrong with exotic thinking [] I have to agree

with the building up idea...in fact, that was what I was thinking with

the triangle corner sub: make the front face just large enough to fit

the driver - put the amp on the very top and then a port along one side.

As far as wilson audio goes....well.....I think no comment is the best

response, lol. Value is in the eye of the beholder - not the cost of

the product.

[Jay481985]

Also where have all the DIY gone! where are you FirePinch, Formica, Michael Hurd, etc etc?

BTW I feel that as much knowledge that I have, I know Michael Hurd will school me as to how to build an enclosure, paint, etc.

DrWho seems to be our in forum Guru on subwoofers though.

TheEar has the most though...

[Jay481985]

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?

[DrWho]

DrWho seems to be our in forum Guru on subwoofers though.

Bah, I just have too much time on my hands. There are plenty of guys on

the forum that understand this stuff...they just have better things to

do, lol. The funny thing is that everyone would probably pump out a

different design...even for the same driver. Me and Michael Hurd chat

on msn all the time and model drivers - we're always on the hunt to

find that hidden jewel of a driver that outperforms all the rest.

Rarely do we end up with the same designs.

Speaking of driver nonlinearities - winisd will show the effect of

temperature rise in the voice coil. Under the "driver" tab you will see

a box labelled: "voice coil temp rise." Try inserting values of 50 and

100 Kelvin...and notice the change in the frequency response and cone

excursion. It just so happens that this particular design handles the

temp rise rather well - actually improving the response in some ways

(and hurting it in others). So with no VC temp rise you are looking at

the small signal behavior of the driver and then with temp rise you're

looking at the larger signal behavior. So the more power you pump in,

the hotter the VC gets.

[Jay481985]

lol - how'd you know I get glue all over the place? [][] Seriously,

building is not my strong point...I can think of how I would go about

doing it, but I never have the right tools and after a few hours of

using a jigsaw you start to get sloppy (yes, I use a jigsaw to cut all

my pieces). I also never wait long enough for the glue to set and so I

end up having to reglue or even recut a piece because the glue dried

and preventing me from using more glue. But I'm still a young guy so

maybe after a few more projects I'll have gained more experience (hah,

right).

You kinda remind me of a SNL episode where you need more cow bell...

Also I became patient with my glue waiting atleast a full 1 day before handling it.

And there's nothing wrong with exotic thinking [] I have to agree

with the building up idea...in fact, that was what I was thinking with

the triangle corner sub: make the front face just large enough to fit

the driver - put the amp on the very top and then a port along one side.

As far as wilson audio goes....well.....I think no comment is the best

response, lol. Value is in the eye of the beholder - not the cost of

the product.

you need better tools. I suggest to you DrWho and everyone else in the

DIY to get a very good router. As with a router, a drill, and if you

cannot have a tablesaw a very good circular saw. As I was new into wood

working I brought the Jasper Jig from partsexpress and could not be

more happier though it did cost 30 dollars for some arcylic and holes,

it saved me numerous minutes and frustration. How else could I have

made that intricate internal bracing? Also buy atleast 4 36" clamps if

you can afford to buy small and large then go for it. But If you cannot

buy 4 big ones since you can always step down. Also when your gluing

the final box together I will use the L wooden pieces again and use the

yellow tie down with the rachet to make sure the box is in place. The

nylon web and rachet is good for up to 2000lb of force so I will use a

couple around the box.

I agree with the exotic thinking but some things are easier to make in

autocad than in real life. Also cutting wood at an angle without a good

table saw or miter saw is hard.