TL bass for Scanspeak 25W8585-01?

[simon dart]

I have a pair of Scanspeak 25W8585-01's in 70L 25mm birch ply braced sealed enclosures. I use as the bottom end for a 3W. I can't fault the tight punch they bring from around 40Hz up, but I can't help wondering what they do in a TL. I'm lucky I have a nice big room to play with so the extra reach would be welcome.

The trouble is I've looked at the available material (I can understand!) and have come to the conclusion that my maths skills are not up to modelling it accurately and I don't really want to invest tons of money in hacking up sheets trying to get there by trail and error, its going to take a lot of wood! Anyone willing to lend me a hand?

25W/8565-01 | Scan-Speak

[pkitt]

Using Martin King's TL worksheets, I modeled a 10:1 tapered TL. The line length is 85", the woofer's center is located 17" from the beginning of the line (the "closed" end), the area at the beginning of the line is 12"W x 10"D, the area at the end of the line (the "open" end) is 12"W x 1"D, and the terminus, of course, is 1"H x 10" W, which you could also make 2"H x 5"W. I used a stuffing density of 0.75 lb/cu.ft., requiring a total of 32-33 ounces of stuffing uniformly distributed by density in the first 57" of the line. I arbitrarily modeled with 0.5 ohms in series with the woofer to simulate crossover inductor resistance. The line would need to be folded once and could be built into a cabinet with internal dimensions of 12"W x 11.75"D x 42"H, allowing the single fold and separating the two halves of the line with an angled, 3/4"-thick divider. The gross volume of this box would be ~97 liters. I'll attach a graph showing the system response from which you'll see that F3 is 30 Hz, F10 is ~21 Hz, and the rolloff below F3 is ~15 dB/octave. This box is larger than your current box; if you'd be happy with a higher F3, the box could be made smaller. I also tried an ML-TL but didn't like particularly like the results, although I didn't spend a whole lot of time on it.
Paul

Originally Posted by simon dart

I have a pair of Scanspeak 25W8585-01's in 70L 25mm birch ply braced sealed enclosures. I use as the bottom end for a 3W. I can't fault the tight punch they bring from around 40Hz up, but I can't help wondering what they do in a TL. I'm lucky I have a nice big room to play with so the extra reach would be welcome.

The trouble is I've looked at the available material (I can understand!) and have come to the conclusion that my maths skills are not up to modelling it accurately and I don't really want to invest tons of money in hacking up sheets trying to get there by trail and error, its going to take a lot of wood! Anyone willing to lend me a hand?

25W/8565-01 | Scan-Speak

[simon dart]

Thanks very much for looking at it Paul. Your simulation is interesting. Compared with my closed box the f10 isn't much different, f3 is 10hz lower though. Subjectively in room is a different story though! Only one way to find out...

[pkitt]

You're quite welcome. Obviously I used the woofer's T/S values from the spec sheet you linked and it would be nice to know how closely your woofers matched these. Do you happen to know the T/S values for your woofers? It would obviously be better to model with known T/S values. Also, for expediency I used Martin's ML-TQWT worksheet which, for a folded,tapered TL, isn't quite as accurate as with his Sections worksheet. Even so, and disregarding the accuracy of the T/S value, I would expect what I did model to be 95% accurate.
Paul

Originally Posted by simon dart

Thanks very much for looking at it Paul. Your simulation is interesting. Compared with my closed box the f10 isn't much different, f3 is 10hz lower though. Subjectively in room is a different story though! Only one way to find out...

[pkitt]

In my first post where I was listing dimensions, I incorrectly stated the terminus would be 1"H x 10"W, or 2"H x 5"W. I meant to say 1"H x 12"W, or 2"H x 6"W. The modeling and the graph were correct, however.
Paul

[simon dart]

Hi Paul, no I havent measured the actual TS parameters, I know they can vary significantly, I believe a TL is less sensitive to driver variation than a ported enclosure though?

[pkitt]

Yes, that's generally true although there will likely be some changes to the system's expected performance if the actual T/S values are significantly different from the T/S values used in the modeling. It would be very helpful if you could somehow get your drivers measured if for no other reason than to possibly make the issue moot.
Paul

Originally Posted by simon dart

Hi Paul, no I havent measured the actual TS parameters, I know they can vary significantly, I believe a TL is less sensitive to driver variation than a ported enclosure though?