SB15NAC30-4 , thinking outloud

Nicholas_23

I have a pair of tritrix cabs I built quite a few years ago but never completed ( no drivers ) . I got to thinking , I think I could fit the aluminum SB acoustic 5 inch drivers in them but Im not sure how they would model. Do any of you have way to model two of the 4 ohm drivers in the tritix TL cab? Is it a viable option ?

Thanks guys 
Nick

jhollander

The tritrix had multiple cabinet sizes what did you build?  I'd model with unibox two drivers in series, but any box program would work.

Nicholas_23

John , I built Curts MTM TL per his specs, on his speaker design page. Image is from Curts site. Thanks for the reply

jhollander

Haha, now I see why you asked.  That can be modeled in Horn Response.  Not sure if I would have time this weekend to look at.

D1PP1N

Worst case you can bore some holes in the back and add another panel to the back for a traditional cabinet.

JasonP

Assuming I can find the time tonight I'll throw the specs intro Martin King's mathcad sheets and see what they predict

Nicholas_23

Thanks for the input guys , I would most definitely take up the offer if anyone wants to attempt to model these cabs with these drivers. Thats most generous and I would appreciate it greatly ! When time permits of course 

JasonP

I modeled a pair in series with a supposed 1 ohm air core coil in front of them to get this result:

F3 about 60 hz, looked ever worse before the 1.0 series resistance, since that raised the system Q a bit. The junk up top is effect of the TL open rear. If you wanted to install a port at the end, things might look better. Maybe I'll model that for fun in the MLTL sheet.

Nicholas_23

Jason , first off , thank you for taking the time to model this . I owe you , if I remember correctly, youre a whiskey man right?
So the 300-700 hz ? How noticable would that wonkiness be in that range ? Would it be audible? It looks to be about a 5db +- swing

PWRRYD

What did you use for stuffing in your model Jason?

JasonP

PWRRYD said:

What did you use for stuffing in your model Jason?

0.5 lb / cubic foot

JasonP

Nicholas_23 said:

Jason , first off , thank you for taking the time to model this . I owe you , if I remember correctly, youre a whiskey man right?
So the 300-700 hz ? How noticable would that wonkiness be in that range ? Would it be audible? It looks to be about a 5db +- swing

I am indeed a whiskey man!

So that is the thing right? The model doesn't look too good above 300 hz. Without a change to listen to the Tritrix, I can't verify the predictions. That said, let me explain why the predictions above 300 hz might be wrong from these sheets. In King's model, he only allows for a single driver. That means in the case of two drivers, we adjust the T/S params for them (in series or parallel, series in your case since two four ohm drivers) and put the placement in as the center of both drivers. Now in the deep bass region, where the wavelengths are large in comparison to the driver and box, I trust the predictions of the sheets. Once you get above 300 hz, the wavelength drops under four feet. Now the interaction between two drivers with a wide spacing, like the one we have in the Tritrix should hold more influence on the actual performance.

tl;dr version: Model might be broke because of driver spacing.

PWRRYD

I agree Jason.  And floor bounce and room interaction....  models are great but...

Nicholas_23

So , with that being said , would it still be worth pursuing this build ? It would be easy to cut another baffle for something else if it falls short .

JasonP

Nicholas_23 said:

So , with that being said , would it still be worth pursuing this build ? It would be easy to cut another baffle for something else if it falls short .

I will bet money that if you'd do the work to put a port at the opening of the cabinet, it'll model much better. I'll work up that sim tomorrow.

JasonP

Finally got a chance to model this. As I suspected, it looks damn nice. Since it was so good, I'll share more data.

First, the model is the same as for the open end TL, but: You place a 3" diameter, 2.5" long port at the terminus. Increase the stuffing to 3/4 lb per cubic foot, and you get:

F3 of ~40 now, much better. Less port interaction up top too. Individual responses for driver and port:

Still more output than I'd like to see about 300 hz, but to fix that we'd need to redesign the whole box.

Excursion at 1W. Don't forget (as I often have) that this is RMS excursion so to get limit, multiply by the root of 2. At a mere watt, the design seems to take even 10 hz in stride. Higher power? Let's try 20W:

5.6mm limit excursion at 30 hz, not bad. Getting out of control under that though, which is expected since we are under tuning frequency. Model showed 103db output at this wattage, FWIW.

Nice group delay for a ported system, let alone an MLTL, which tend to fair worse. You'll have no worries about "late bass".

In a nutshell: I'd 3build this.

Nicholas_23

Thanks Jason for messing with this . So, confused .... there is the line that already runs from the top of the cabinet and angles down towards the bottom . Are you saying , keep the line in place , then block the back of the speaker off with some mdf and drop a port towards the bottom? 

JasonP

To be clear: Around the place in the diagram where it shows the 4" distance between the rear and the tapered panel, block that off and put in a 3" diameter port, about 2.5" long.

JasonP

Nicholas_23

Perfect , thanks for clearing that up Jason. Beats sealing up the whole back end of the cabinet.