diyAudio (
-   Full Range (
-   -   Accidental MLTL Technique (

xrk971 15th February 2018 05:25 PM

The Qts is 0.40 - on the low side and means it has a powerful motor that is not suitable for a MLTL to pull bass tuning below fs. If you can provide me the linear-straight line representation of the MLTL with cross sectional areas in square inches or cm with associated segment length in inches or cm, I can run a model for you. Also, if you can type the TS parameters into this format (for example), it will save me some time too.


Def_Driver 'SB12NRX25-4' | comments after vertical bar here SBA 4in Norex cone mid bass, Qts 0.40, 87.5dB, 30W rms, 5mm xmax

oon_the_kid 16th February 2018 03:35 PM

Dear XRK,

Thanks for your help in the modelling.

Cross section would be 153cm^2 (l=60cm) followed by woofer and fold, followed by 176cm^2 (l=50cm) ending with a port that is 25cm long (5cm diameter).

Def_Driver 'SB16PFC25-4' | comments after vertical bar here SBA 6in cone mid bass, Qts 0.34, 89.5dB, 40W rms, 5mm xmax

Vas= 32.7mH
Le= 0.56 ohm.

Many thanks.


outfoxhyperion 7th August 2018 04:34 PM

2 Attachment(s)
Hi All,
I used the MLTL technique to build these towers. The innards are a transplant from my first pair of DIY bookshelf speakers that I wasn't happy with the performance of (too far on CTC caused lobing and sealed didn't really get good bass extension). I used an enclosure calculator on my phone to get the dimensions to something I was happy with, just told it to calculate the other dimensions needed with a 1"x 6" slot port, 6.5" wide baffle, and 36" tall. This gave me a 14" depth with a 5 1/2" long port for a net volume of 55l. Port tuning was set to 35 Hz, and box tuning to 60 Hz, which in theory should give a bit of a saddle between them. Didn't really seem to work out that way, unfortunately, and I think that's mostly down to the fact that the Qts on these drivers is just too high to do really well in MLTLs (0.92). Still, they sound better than they did in the sealed bookshelfs both in terms of bass extension and the lobing that I mentioned earlier, plus the sound quality to time investment ratio is really high. These are stuffed with polyfill somewhere between halfway and two thirds down the line, and driver is placed one third from the top. I don't remember the exact crossover frequency, but it's a second-order Butterworth somewhere in the 2.5k range. Lower than it should be, but the tweeter doesn't sound harsh and I wanted to avoid a pretty serious ~+7 dB hump the woofer has at 3k. Probably didn't work out that way, as the lower treble range is really quite a bit stronger than it should be. No measurements, unfortunately, as I haven't gotten around to buying a calibrated mic yet. I would post the drivers I used, but then you might be tempted to use them, and I wouldn't advise that in a similar build.

EDIT: the importance of bracing can't be understated. I only used 1/2" panels for the sides, and you can absolutely tell where the bracing is and isn't. I plan on gluing another layer to the outside to give it a bit better rigidity.

Here's a couple of pictures:
Attachment 695890

Attachment 695891

xrk971 7th August 2018 05:02 PM

Thanks for sharing - I once made very similar tall towers for budget 6.5in $5 woofers from PE (used in their popular sealed bookshelf speakers). The drivers indeed had too high Qts and did not work well here. So it should be stressed that Qts should be in range around 0.45 to maybe 0.65 ideally. Too low Qts is not good either as you won't get bass extension below fs of the driver from the mass loading.

GlenP 12th August 2018 09:08 PM

I'm designing some desk speakers around the Mark Audio Pluvia 7, which may get occasional other use (or I may build another set of for other locations, depending on how they come out).

I have realised that if I go to the recommended size for a reflex design (11-12 litres) and if I combine that with a cabinet height putting the drivers near ear height I can end up with something quite tall and slim - temptingly close to being a MLTL folded in half. However, to keep the cabinet depth reasonable I end up with a cross section for the constant width line being about 2.8 X SD.

Question is; is that too tight? Should I just revert to a standard BR?

Also; I don't have the wherewithal to model the design before I build it - but in very general terms I have the driver about 30% of the way down the line and the port about 20% from the other end - would that be a reasonable starting point?

Line is 85cm, driver is at 24cm, port is at 68cm.

xrk971 13th August 2018 12:08 AM

Those dinensions and proportions sound like a fine AMLTL. Bottom port maybe 10% from bottom and add stuffing from top to just where driver is. Use mesh cloth to hold in place so doesn’t fall. Probably add 1mH and 6.8R (10W) BSC to even the response out for more rich bass.
Good luck!

GlenP 13th August 2018 08:34 AM

That's great news, thanks for the tips. Earlier in this thread I got some great advice when I was trying to figure out a MLTL bass section - as is often the case daily life got in the way of me finishing that design, but these desk speakers are going to be quite simple and inexpensive to build - and hopefully they will be quite flexible. That's the idea anyway.

atmo 10th September 2018 02:32 PM

I don't quite understand how the BR volume relates to that of the MLTL, for smaller drivers at least. For example WinISD suggests a 6 litre cabinet for the TB W3-871 at 75 Hz, yet that of the TABAQ is more like 10 litres and 55 Hz. Such a small volume makes it infeasible to have a CSA that isn't narrower than the driver itself, for any kind of TL length.

xrk971 10th September 2018 04:42 PM

If you tune the BR for 55Hz (and this only works with Qts > 0.42 or so) you will find that it can tune lower than fs. Volume usually increases to make box frequency go down. Once you have volume, set the baffle width, and set the TL length, typically no shorter than 30in to 32in. Solve for depth to make volume work out.

GlenP 10th September 2018 04:44 PM

Do you tune the reflex port to the lower tune, as well as using a larger volume? In this MLTL example I mean, not in general

All times are GMT. The time now is 02:03 AM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 17.65%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio