DIY subwoofer advice requested

[Daddy Dee]

The tuning depends mostly on how the driver behaves. I believe most of the Klipsch designs use an extended bass shelf alignment and then EQ it up inside the amplifier. For a prosound application I would favor tuning a bit higher, so the two combined would imply a tuning well above the quoted F6. I'd probably aim at around 40Hz actually and then use the subsonic filter built into the amp (highpass at 30Hz?) If you're creative, you could build in a variable tuning port and then just slide it around until it measures how you want.

Mike,

Thanks. makes sense.

[michael hurd]

For a 6 cu ft ( * net, after driver and port displacement! * ) box tuned to 40 hz requires two 6" ports 11.42 inches long.

To use triple 4" flared port kits from Parts Express ( aeroport kits ), you need to make them each 7.23" long.

If you went with triple 6" ports.... that would require them to be 19.34 inches long each.

You need to ensure that the ports are at least one port diameter away from any internal wall. You have to figure out what scheme you are going to use, then calculate the volume displaced by the ports in the enclosure, and then add that volume to the net volume to get the gross internal volume.

There is a freeware box calculation program that also generates a cut-sheet and can figure out volume internally and also the internal resonances of the box. ( as determined by the dimentions, as well as the placement of the driver in said enclosure ) It is called "boxnotes", and a google search should be able to find it easily.

Ideally you want the dimentions adjusted so that obviously everything fits.... and the resonances don't all occur at one frequency.

You should also apply a bit of foam or other material such as fiberglass ( inside a cloth covering ) to a few walls inside the box. This will reduce the tendency of higher frequency sounds emitted from the rear structure of the driver and spider from being emitted through the port.

[Daddy Dee]

Dee... you don't per chance have the T/S parameters of that driver? In my opinion, I would build a 6 cubic foot internal box, that has both drivers in it, packed tightly. ( one above the other, with frames as close as you can get ) As far as a tuning frequency, I would stick to highish 30's or 40 hz as Dr.Who suggested, as they are only 12" and most likely short throw, tight suspension drivers.

I would use either 2 6" ports if you can or 3, 4" Aeroport kits. This will give you enough port area to be pretty much compression free, and keep the velocity down.

Michael.

Wish I did have the TS parameters. Haven't been able to locate them yet.

When you say, "packed tightly" do you mean with some sort of fill for the cabs?

Also, "frames close" do you mean having the woofers mounted close to each other?

The aeroport kits... haven't identified a seller for those.

[michael hurd]

By "packed tightly" and "frames close" I mean that the drivers should be as close as physically possible to minimize the lobing that will occur at higher frequencies.

Parts Express sells the Precision Port 4" aeroport kits.... I forgot to mention that the ports also need to be tightly packaged ( extremely close together ) If you use three of them, place them in a triangular fashion, two on the bottom, one on top of those, or the opposite.

You could also use 6" PVC drain pipe.... or even sonotube for a port.

[Daddy Dee]

gotcha. Thanks Michael

[DrWho]

Yea, I'm with Hurd on this one [Y]

However, what is the HF corner for the system? If you're crossing over at like 100 Hz then you might consider spreading the drivers out a bit - like maybe 3 to 6 feet. That would put you right around a 1/4 or 1/2 wavelength spread which could gain you some directivity (since at the lower wavelengths the directivity is going to be too wide). You could also put one driver about 3 to 6 feet behind the other driver (I would need to do some calculations first) which would give you a cardoid coverage pattern. A little time delay on one driver could accomplish the same thing. Here's an article that goes over it:
http://www.prosoundweb.com/install/synaudcon/tt26_3/tt26_3_2.php

I bring this up because PWK put directivity over on-axis frequency response in order of priority - and in a prosound application it's even more vital. It's also rather inexpensive to do in a DIY application and so far my experiences have been that it is always an advantage.

[Daddy Dee]

Mike,

Thanks. 100hz crossover is what we have in mind.

[Daddy Dee]

OK. I've been thinking. (so this can be dangerous)

it is the volume of the ports relative to the interior volume of the cab which determine the tuning frequency of the cab, right?

So, for a cab with a volume of 3 cu. ft. the volume of the port would be X. right?

Am I correct in understanding, then, that it would not matter if the volume of the port is distributed between two larger ports or three smaller ports?

It would be necessary though, for the sum of the volume of the two ports was equal to the sum of the volume of the thee ports, right?

Now, would this also hold true for the shelf type port?

One more question: Is the volume of the port included in the volume of the box. For example, if the 3 cu ft cab needed to have a port volume of 1 cu.ft, would the box dimensions need to be 4 cu ft to compensate for the 1 cu ft port?

Thanks in advance for any help.

[CECAA850]

The port volume is not included in the total box volume. Your example is correct.

[DrWho]

The volume of the port is NOT what determines the tuning frequency.

The major component to the tuning of the port is the relationship between the cross-sectional area and the length of the port - and then how that relationship couples to the volume of the enclosure. Saying it this way makes it sound uber complicated, but it's really not.

Here's an example:
Take a coffee straw and a normal drinking straw. Ever notice how it is way harder to suck or blow air through the coffee straw? Now connect a bunch of coffee straws together into a long line so that you have the same volume as a single drinking straw (I'd wager about 10 in a row). It's going to be a lot harder to suck air through the string of coffee straws.

How hard it is to blow air through the port plus the compliance of the air in the volume of the enclosure determines the tuning frequency.

So... the shape of the port doesn't matter in terms of tuning frequency as long as the cross-sectional area and the length of the port are the same.

However, the shape of the mouth of the port can have an effect on port chuffing - a circle is definitely the best shape, anything else is a compromise. A slotted port is the worst kind actually.

But ya, if you want a 3 cubic foot rear volume and your port takes up 1 cubic foot, then your external enclosure is going to be 4 cubic feet.

[JohnA]

I'd build 2 and spread them out.

I've always had better results with center packing the subs since it gets rid of the power alley - dramatically improving the coverage in the room.

http://www.prosoundweb.com/live/articles/jbrusi/pa.php

Mike, nice article. I agree you could expect that effect outdoors and I think I've heard it. Probably the same in a very large venue like an arena.

Indoors in a smaller venue, you'd have a reverberant field and standing waves become a problem. Spacing the subs may well cause one to fill the nulls caused by the other. I suppose you could model the room to see, assuming you have access to the software.

[DrWho]

Indoors in a smaller venue, you'd have a reverberant field and standing waves become a problem. Spacing the subs may well cause one to fill the nulls caused by the other. I suppose you could model the room to see, assuming you have access to the software.

I totally understand what you're saying, but I'm not sure if the polar lobing could ever line up with the modal response of the room...

The polar lobing acts like a bunch of fingers, but the modal response is more like a 3D polka dot distribution through the room. The fingers from the polar lobing are going to be as tall as the room and extend back like giant rays from a sun.

I'm sure some of the rays will land on the polka dots, but inbetween the polka dots you're still going to have the ray...so basically now you're left with two sources of discontinuity instead of just one. Sure, there will be some influence from early reflections in the room, but they diminish quickly the further the boundaries get from the speakers, and those reflections will be coming from the rays or act like other point sources (and more point sources just makes it worse).

I dunno, even in the smaller venues I'll usually end up with center packing (like all things audio there are always exceptions) - though to be honest, it's rare that one gets to experiment enough with any speaker/sub placement. Usually it's a mad rush to have enough time for a sound check...

[Daddy Dee]

Gentlemen,

This is most interesting and helpful.

By building two cabs, we can try experimenting with placement and then observe the impact on sound.

[DrWho]

Well that's just silly talk! []

I'll be looking forward to your results.

[michael hurd]

Dee... you have to figure if you need a 3 cu ft box, and your port or ports... displace say 1.5 cu ft, and you can estimate the volume of the woofer ( magnet, basket, cone... ) to be about 0.1 cu ft, then your actual INTERNAL volume would be in this example 4.6 cu ft. External... I am guessing about 5.25-5.5 cu ft.

The external dimentions would be larger than a passive radiator tuned box of the same frequency.

[Daddy Dee]

Michael.

Gotcha. Hadn't thought about the displacement due to the woofer. Thanks.

Oh. yeh. Let me ask about this to see if these numbers sound right. I found a port calculator on the web:

http://www.the12volt.com/caraudio/boxcalcs.asp#por

given a 6" port in a 3cu ft box tuned to 40hz it calculates the port length to be: 11.48557"

using two 6" ports, the port length of each would be half of 11.48"

Does that jive pretty close with what your software suggests?

[michael hurd]

Dee, I came up with 11.42 inches... it depends on how you terminate the port.... as the tuning frequency would change if you have a flange or a flare on the inside as well as the outside. IF you decide on TWO 6" ports per single 12" driver box, then they would need to be 27.23 inches long each.

If you are going with a two-box approach, then I would build them so that when stacked both the port and driver are close to the other respective port and driver. IMO, simpler with a single box approach, less cutting, so on. More versatile with a two-box approach yes. Remember that the drivers are only 12" and they are not going to give you ground-shaking performance. They will fill in the low-end nicely to a point.

You should have a high pass filter in the amplifier OR crossover device for the sub or subs. If you do decide to pick up a Behringer DCX 2496, I would place a 2nd order butterworth high pass at 40 hz, or just a shade under it. ( likely tuning will fall within a few hz, have to check the tuning with test tones )

This will ensure that the system is trouble-free and that it can't be damaged easily by simple mistakes or playing back some material with sub-sonic content.

[Daddy Dee]

Michael,

Thanks. I'm still kind of dense of these port length and volume issues.

In an earler post, were you thinking of three 4" ports in this enclosure?

Would there be any advantage to having a single six or pair of six inch ports? The length of each port at 27" or so is amazing for two six inch ports. That wouldn't seem to be too attractive a choice.

What would the length of three four inch ports be?

Any advantage to using three four inch ports?

[michael hurd]

You have to make a choice as to what you are going to build first. One box or two boxes... both have their pro's and con's. If I were to build it myself... I would probably chose to build a large box with both drivers in it, and throw in ( 2 ) 6" diameter ports, 11.5 inches long, router the opening on the baffle board with a roundover bit and call it a day.

I already figured out the length that ( 3 ) 4" diameter port kits would need to be in a common box, with a net internal volume was. I will go back and re-post the length again.

Any advantage to the triple port in a common box ( both drivers in one box ) ? Sure there is, the main thing is that they are cosmetically attractive, flared on both ends, easy to tune. ( 4 screws and the port comes out in one piece ) Downside is that they take up more volume in the box than 2 6" ports... and also the fact that they cost more. If you used ( 3 ) 4" ports, the box will be bigger on the outside than a box with ( 2 ) 6" ports.

Also, if you use any bracing within the box, the volume of it needs to be added to the NET volume ( still at 3 cu ft per driver, 6 cu ft total ) to get the actual construction volume.

EXAMPLE: 6 cu ft net + 0.2 cu ft ( drivers ) + 0.44 cu ft ( volume taken up by 3, 4" diameter ports 18" long each ) + 0.1 cu ft bracing... = 6.47 cu ft GROSS ( actual internal construction volume ).

Dual 6" ports would only take up 0.22 cu ft internally in the box... so you could make it slightly smaller.

[Daddy Dee]

Thanks Michael. We';ve got some good ideas comining together.

Also, I want to express my gratitude to Mr. Cannon and Mr. Phillips in Klipsch Engineering.

They have provided TS parameters on the K-26-K driver:

Vas 200 liter

Fs(Fo) 27 Hz

Qes 0.3

Qms 5.2

Qt 0.28

8 ohm driver

150 watts nominal

How cool is that!