Design check of a simple <100€ sub

[hurtz]

Hi folks!

I finally decided to dive into speaker building after finishing some amplifiers.
Since this is my first speaker I wanted to keep it simple. A single 30cm/12" chassis, sealed enclosure.

The chassis I want to use is a Mivoc AW 3000 ~45€
As an amplifier a symasym ~45€

I simulated with winisd with a box volume of 45l internal volume, which with boxnotes gives something around 40cm*40cm*44cm.

As I understand for sealed enclosuers, keeping specific dimensions is not really as important as for vented subwoofers. So the Shape shouldn't really matter until it gets extreme right? (as for volume winisp suggested 25l and the datasheet 40 to 55l)

A 4th order Linkwitz-Rily:
Lowpass at 175Hz,
Highpass at 19Hz to protect the cone.

A comparison with and without active filters:


Cone excursion:


SPL:


TFM:


So, this configuration is more or less the standart configuration for this driver, I know it's not exactly hifi.. but for a first this should work.

For tweaking I wanted to use dampening material in order to "simulate" a bigger enclosure.

As the graphs suggest the response is not exactly linear, is there a good way to create a more linear response (by means of active filtering etc...? Or is it simply not that important/worth the effort?

I'd be glad to hear your thoughts about this :-)

cheers!

[bjorno]

Originally Posted by hurtz

..For tweaking I wanted to use dampening material in order to "simulate" a bigger enclosure...

Hi, Here is a suggestion that uses damping material(not shown) for the first half of the enclosure starting at S1 in Hornresp program:

b

[hurtz]

Damn that looks nice...

But waaaay to big/hard for my first speaker ;-)

Thanks for the sim though, really interesting to see the differences in performance!

By chance I found the "peaking 2nd order high-pass" which looks like it can be used fairly easily to set a gain for specific frequencies. That would make 3 aktive filters in series.. what is the consens about loss of signal quality with that many filters?

cheers!
hurtz

[hurtz]

Ok, after playing around with the filters and some reading the current data is:

- Lowpass L-R 4th, 175Hz
- Highpass L-R, 4th, 33Hz
- Peaking 2nd order Highpass 3db @ 50Hz

The lower cutoff is at 50Hz, Higher at 150Hz.

But still some questions:
- what do you think about using damping for this sealed enclosure.
- Do there have to be reinforcements in the enclosure or is the 40x40x44cm design stable enough without crossbars(?)
- For anyone who has used products from "Teufel" How do you think will this system compare with the E200/E300 Subwoofer? I don't have any useful comparison between subwoofers of different brands/makings since I only used above system for some time (until I built my own amp).
- Is the additional 10ms group-delay from the peaking filter worth the additional 10Hz in cutoff?

Groud Delay


SPL


tfm


cheers!
hurtz

[bjorno]

[QUOTE=hurtz;2810449]

Ok, after playing around with the filters and some reading the current data is:

- Lowpass L-R 4th, 175Hz...hurtz

I never XO a sub above ~145 Hz using a LR4 LPF due to the localization issue at ~250 Hz and if used for the mains as stands: the possible Phantom hang down effect.

But still some questions:
- what do you think about using damping for this sealed enclosure.

Using a small amount of damping material doesn't hurt...

- Do there have to be reinforcements in the enclosure or is the 40x40x44cm design stable enough without crossbars(?)

I always use cross-bracing if the internal CSA is larger than ~10" x10" and if (x10" = volume) to make the enclosure as rigid as possible without overdo...

- For anyone who has used products from "Teufel" How do you think will this system compare with the E200/E300 Subwoofer? I don't have any useful comparison between subwoofers of different brands/makings since I only used above system for some time (until I built my own amp).

To be frank...I would never buy drivers from those speaker brands..

- Is the additional 10ms group-delay from the peaking filter worth the additional 10Hz in cutoff?

IMO using as peaking filter for a driver with low x-max excursion is not to design for SQ and I never allow Gldy higher than ~30 Hz to be more than the corresponding 1/f response, i.e at 50 Hz Gldy max= 1/50 = 20 mS.

b


PS: Submitting some Mivoc calculations I forgot to post and BTW the footprint of the suggested T-TQWT is a few square inch larger than the box you are simulating but the possible SQ performance would IMO be second to none.

[hurtz]

Ok, ok you have me convinced :-p

Now you have to deal with my horn-questions ;-)

I tried to play around with hornresp but was lost after a short while... but maybe some general questions first. I can see that the horn has a very steep cutoff at 100Hz, is this a horn-specific thing?

And maybe more importantly, is there a good hornresp T-TQWT dseign/build- tutorial out there?

100l seem ok as long as its not too high to be used as a footrest below the desk :-)
No seriously, I have a very small room at the moment so size does matter. So 40x40x120 HxDxW is more or less the maximum.

It would be great if you could help me get started on this :-)

cheers!
hurtz

[bjorno]

Originally Posted by hurtz

Ok, ok you have me convinced :-p

Not my intention at all. I just found a use for your driver..

Now you have to deal with my horn-questions ;-)

Horn questions? A T-TQWT is not related to a horn at first glance but more to a quarter-wave type of enclosure that is made with BP(band-pass) characteristics:

Like a TL.TQWT or a QWP that it's possible to model in Hornresp or AKAbak or partly if using MJK:'s programs.

I tried to play around with hornresp but was lost after a short while... but maybe some general questions first. I can see that the horn has a very steep cutoff at 100Hz, is this a horn-specific thing?

The T-TQWT can use an offset driver location in order to avoid that dip.

And maybe more importantly, is there a good hornresp T-TQWT dseign/build- tutorial out there?

You have to Google or do a search of this forum..I've posted a few examples...

100l seem ok as long as its not too high to be used as a footrest below the desk :-)

A very good place to place a T-TQWT!

No seriously, I have a very small room at the moment so size does matter. So 40x40x120 HxDxW is more or less the maximum.

You can estimate the build size height by dividing the length L of the airways by two and add the thickness of the wood times two, the width is the driver frame diameter + two times wood thickness and the depth would in HR be about (S1 +S4)/(driver frame width) +3 x wood thickness(internal divider included).

It would be great if you could help me get started on this :-)

I'm not in a hurry

b

[hurtz]

Originally Posted by bjorno

You have to Google or do a search of this forum..I've posted a few examples...

Sure, but I hoped there was THE tutorial, known by all and therefore never mentioned ;-)

cheers!
hurtz

[bjorno]

There are many tutorials that easily can be found at the Net: On Hornresp (Search: Tutorial Hornresp) ..or trying to circumvent the deeper knowledge(patience) required for a successful speaker design.

An example:

Easy Horn Spreadsheet for Dummies - AVS Forum

PS:I will not reply to any question concerning this spreadsheet.

b

[hurtz]

Great! Thanks!

This makes it a lot easier for me to understand. Now let's see what we can make of this :-)


cheers!
hurtz