Enclosure design help

[pjanda1]

I have a driver that I'm having a hard time simulating and enclosure for with WinISP. It is something I'm beta testing for a manufacturer, and I know it is being used in bass reflex enclosures a little larger than a cubic foot. WinISP is adamant that it needs an enclosure of about 7.75 cubic feet. Is there something I'm doing wrong? Here are the parameters I've been given.

Revc= 6.500 ohm
Fo= 41.203 hz
Sd= 22.167m M3
Krm= 124.110m Ohm
Erm= 0.306
Kxm= 1.035 H
Exm= 0.000
Vas= 105.116m M3
Cms= 1.506m M/N
Mmd= 8.006m Kg
Mms= 9.904 g
BL= 5.225 T-M
Qms= 2.880
Oes= 0.610
Qts= 0.504
No= 1.165%
SPLo+ 92.681 db

thanks,

Paul

[GM]

Greets!

Nope, I cross calc'd the various specs to confirm Vas, Qts, and using my reference box formulas I get ~219.139L (~7.738821 ft^3)/33.41 Hz for a T/S max flat. Even a prosound alignment is ~52.556L (1.856 ft^3)/ 30.9 Hz, so maybe they prefer the 'boombox' sound.

GM

[pjanda1]

Thanks for the reply.

I've been doing as much reading/searching as I can. It seems like this might be borderline for an open baffle, but would need some equalization. I've been reading about transmission lines, but designing is probably out of my league. I imagine some sort of aperiodic solution may exist as well. In any case it seems that any enclosure would either be difficult to design, domestically unacceptable, or some combination of the two.

If I'm wrong about this, feel free to chime in, otherwise I'll look for another project. Too bad creativesound is out of Jordans, my new gainclone is tiring of its present company.


Paul

[GM]

Greets!

You're welcome! With the possible exception of a horn, all the cabs it's suitable for are simple designs, even TLs, so I'm not sure what you're referring to.

Anyway, with a 0.5 Qts, it's suitable for any box design, so it boils down to what F3 and box Q you'll settle for if the design is Vb specified. Want a low F3, the box must be large, and vice versa.

GM

[pjanda1]

By "simple" I was refering to my ease of design. My TL reading has been mainly restricted to MK's website, and most of the discussion is beyond my current understanding. I'm also mostly restricted to a Mac, so I can't run MathCad.

Domestic acceptability is limited to a maximum of 60 or so liters, and I can't find any tuning that seems to work for a Vb in that area. Any sealed box in that area is an awfully high Q too.

I'll be the first to admit that my knowledge is of loudspeaker design is subpar. I'm hoping to extend it with this project, but for that I need to find a suitable design. I've seen pics of this very driver playing to acclaim in vented boxes smaller than 60 liters. I expected to be able to plug the specs in to winISD and start cutting wood. Either those using this driver are modifying it, or the enclosure solution is elluding me.

I appreciate all your help; I just wish I didn't have to ask such silly questions.

Paul

[pjanda1]

When I model a smaller Vb tuned higher, I get a huge peak. If I model a 33L box tuned to 32, I get a 2db peak at 110 or 120 hz, that tapers up to 300hz. Would this be noticable without BSC? I know ears are sensitive to peaks, but whats the big deal about 2db? Is something other than frequency response affected with a tuning like that?

paul

[GM]

Greets!

Sorry for the tardy reply. Did you build something that sounds decent, or still 'fence sitting'?

GM

[VadimB]

pjanda1,

Before you give up on this project, please, doublecheck the T/S parameters you've enterd in WinISD. I'm affraid they don't match, or you have mixed the units somewhere. For example, if Sd is in cm sqared, you driver is 2.5" diameter, and then no way it has 105 liters Vas.
Check if you are running the latest version of WinISD. If not, download WinISD pro Alpha. This version checks T/S for consistency when you enter the driver parameters and will not let you save the driver file, until everything matches.

Another issue is the driver Qts. If you want to have a classic reflex box alignment, the one where you use the air VOLUME resonance, Qts of about .44 is the upper limit. Above that you will have to use some sort of Chebyshev alignment (bet that's waht WinISD suggested?) with some ripple. These require huge boxes with Vb >> Vas. I've built quite a bit of those in the beginning of this hobby. Looking back - they did not sound good. Lots of boom - not a lot of control.

However, a MLTL is a different story. It uses a quarter-wave resonance, not the volume resonance. Although it looks like a ported box, the physiscs behind it is different. This box has to be tall, to fit that quarter of the wavelength. Unfortunately, the only way I know of to model this is MKJ worksheets. It may be worth to find a PC (work, library??), download MathCAD explorer and worksheets and model it using "Ported box" sheet. The good news is - I've used this with drivers having Qts as high as .9 with decent results.

In any event, make sure the numbers and units are correct.

Vadim

[pjanda1]

Wow, somebody did some digging to bring this up. Heres the deal. This was an early version of the Hemp Acoustics driver. I couldn't come out and say it at the time. I wondered if someone would read my later posts and put it together.

I'm not sure why Sd was quoted the way it was. It is an 8" driver, and 221cm (squared) is a better spec (that's how I entered them). I tried them both in a 50L enclosure, both sealed and tuned to 40ish hertz. I know it wasn't ideal, but I had the box. It worked either way, obviously with more bass in the later. The response of this driver closely mirrored that of the Fostex FE206E, and was ultimately too hot for my tastes in a BR or sealed enclosure. The current Hemp Acoustics drivers are a couple of generations beyond this.

I've still got the drivers. I've toyed with the idea of building a compensation network to tame them, and ultimately a more suitable enclosure. I think a TL of some sort woudl be ideal, but with that compliance, it would be big. I've been too happy with the newer Hemp Acoustics 8's to give it much thought.

thanks for checking up,

Paul

[VadimB]

This was an early version of the Hemp Acoustics driver. I couldn't come out and say it at the time. I wondered if someone would read my later posts and put it together.

Aha! I thought the spec looked familiar. The thing is, I am working on a disign for the same driver. It will, probably, be the newer version, since I am still waiting for Hemp's shipment to a distributor for several weeks now.
I, actually, want to try it in a back-loaded horn - believe it or not. This weekend I'm gonna try to get the new version of MJK's worksheets ($25 is a bargain for what they can do) and model my horns. The hope is to skip the baffle step compensation.

So, what is a difference between older and newer versions of the Hemp FR8?

Vadim