Help with 12" sub design for donation!

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi,
I'm looking for some help to see what would be the best sub design that I could achieve using 4 x 12" bass drivers with the following known T/S parameters, with or without 4 passive radiators based on the KEF B139

Resistance: 3.4 ohms
Watts: 160
Sensitivity: 95 db/w
FS: 27
VAS: 268 litres
QMS: 2.34
QES: 0.26
QTS: 0.24
Sd: 514 cm/2
EBP: 103.8

I have tried WINisd with the best looking graphs giving a Vb at about 40 litres with a port tuning of about 47 Hz. But without the necessary experience of translating graphs into an actual design I am hesitant to go any further. Also, the passive radiators that I have just acquired add another dimension to the possibilities.
I need to keep the box volume to a minimum so a certain amount of compromise will be inevitable. I'm thinking along the lines of Isobarik loaded active drivers with 2 passives on opposing side faces, but I'm probably wrong about this!
I'm not expecting to achieve a flat response to 20 Hz, as I know that this would not be possible with these drivers.

I would be prepared to make a donation to the forum if a good design could be created, if that is possible with these drivers; or are these drivers not worth the effort?

I could do with some help here.

Cheers and Happy Christmas
 
Specify "best sub design". Do you mean by best
- lowest flat response?
- optimised step response?
- optimised group delay?
- smallest enclosure? With 4 x 12" plus passive radiators probably not.
- high efficiency?

Be aware that some of those criteria are mutually exclusive. Be also aware of the speaker's interaction with the room, so the best design for one room and position therein may not be the best design for another room or for a different position in the same room.

Due to the low crossover frequency on a subwoofer, it is often good to tune the vent or passive radiators slightly lower than in a standard woofer design. That way you get a lower flat response than with a tuning that would be considered perfect in normal circumstances. The slightly mistuned curve adds up with the crossover's roll-off favourably at the cost of efficiency. That means of course it will be difficult to devise a design based on theory only. You will need to resort to the proven method of trial & error, i.e. build, measure, listen, change, measure, listen until you are satisfied.

Four drivers in Isobarik have little to no advantage over using only two drivers in a normal fashion.

If you can afford to put drivers on opposing sides, you should mount active drivers on opposing sides, not actives on one and passives on the other. That helps to cancel out their effects on the enclosure.

With 4 x 12" it should be possible to achieve 20 Hz or even lower with a bit of equalising or a Linkwitz transform.
 
Hi pacificblue,
thanks for your reply.
I understand and take note of what you say, but I do not want to get too technical with these drivers as they were cheap "old stock clearance" drivers, and therefore do not warrant
a large amount of time on. I can model a basic vented set up, but as I said, I lack the experience of translating graphs into actual designs. When modelling, a number of graphs/frequency responses are given according to changes in box and vent parameters that look like they would be viable options, but which one is something that I need help with.
Maybe it would be better if I simply asked
"what would you do (anyone) with 4 bass drivers with the T/S parameters given"

I tried a "Ripole" set up, but was not convinced really, especially with the overall efficiency losses. The drivers have an EBP of 103.8 which suggests a vented box, so I thought that a Linkwitz transform would not be an option as this only applies to sealed boxes.
 
Hi,

Any driver that has realtively high efficiency isn't go to make a great
subwoofer and really should be used a a bass unit in the sort of
speaker it was designed for, here PA / disco types I'd guess.

40L per driver tuned to 40Hz is doable. Isobaric is possible,
and two B139 PR's might suit one driver or two clamshelled.

Hard to say without knowing the drivers excursion capability.

As your after small and isobaric clobbers sensitivity, two isobaric
boxes with two PR's each is probably the way to go, 20L to 30L.

A sub plate amplifier with a bass boost option is probably needed.
Your simply not going to get deep extended bass from these drivers
due to the low Qts requiring boxes much smaller than the driver Vas.

rgds, sreten.
 
Last edited:
Thanks sreten,
this is what I had kind of come up with myself, so it is nice to have someone with more knowledge to come to the same conclusion. The next step would be for me to ask about the passive radiator loading; going by what sreten has said, what tuning to go for?
 
2x pairs of isobaric pairs (ie, 4 drivers) would require the volume of 1 driver, but have twice the displacement capability, and 4x the power handling.

They require more volume than one driver, because there are additional chambers between the drivers of each isobaric pair. For each 12" woofer they would at least be 10 l, probably more. The resulting enclosure will not be much smaller than a non-isobaric speaker with two drivers which will also have twice the displacement capability of a single driver. Advantage isobaric, but not as big as expected.

The isobaric version with four drivers has twice the power handling of a non-isobaric speaker with two drivers, but also only half the efficiency. Advantage non-isobaric due to lower power requirements.

A non-isobaric enclosure is easier to design and build and therefore less expensive. Advantage non-isobaric.
 
Hi,

For use as a 4x12 bass guitar cabinet its a no brainer,
but please don't try and tell me about isobaric loading.

I know the territory and the consequences, clamshell
needs extra volume to clear one driver magnet, and
it cancels even harmonic distortion, so called doubling.

More to the point the 4 B139 PR's are a lot more likely to
suit isobaric pairs (2 each) than any other arrangement.

Minimising box volume and using the PR's points to isobaric.

rgds, sreten.

I'd expect the PR's tuning to moreorless fall into place,
though if they don't you can always add some mass.
 
Last edited:
They require more volume than one driver, because there are additional chambers between the drivers of each isobaric pair. For each 12" woofer they would at least be 10 l, probably more. The resulting enclosure will not be much smaller than a non-isobaric speaker with two drivers which will also have twice the displacement capability of a single driver. Advantage isobaric, but not as big as expected.

The isobaric version with four drivers has twice the power handling of a non-isobaric speaker with two drivers, but also only half the efficiency. Advantage non-isobaric due to lower power requirements.

A non-isobaric enclosure is easier to design and build and therefore less expensive. Advantage non-isobaric.

Sure, yes, some (tiny compared to cabinet) extra volume will be needed for the driver motor structure etc.

Comparing 2x pair of isobaric vs 2x normal drivers isn't a fair comparison: the former will require half the box size, so of course will be less efficient.

Designing an isobaric cabinet is as easy as a normal one, as is the construction. No advantage either way.
 
I know the territory and the consequences, clamshell
needs extra volume to clear one driver magnet, and
it cancels even harmonic distortion, so called doubling.
If you mean back-to-back mounting of the magnets, yes that cancels harmonics. For two 12" woofers you can assume 25-30 l of extra volume to just squeeze them in back-to-back. Seems to overcompensate somewhat for the 20 l you save on the rear volume.

I'd expect the PR's tuning to moreorless fall into place,

Good luck with that.

Comparing 2x pair of isobaric vs 2x normal drivers isn't a fair comparison: the former will require half the box size, so of course will be less efficient.

It is a fairer comparison than comparing two pairs of isobaric to a single driver in post #4. The efficiency reduction due to the box volume is not the issue. Isobaric speakers need twice as much power, because the inner driver needs the same amount of power as the outer driver without increasing the SPL.

Designing an isobaric cabinet is as easy as a normal one, as is the construction.

It is not, because you have to design and build two chambers instead of only one. The difficult part is to mount the inner driver without affecting aesthetics, e.g. by fixing baffles with visible screws.
 
What I need is some help with the modelling, which I have only just dipped my toe into (so to speak) As I have said, it's a question of translating what modelled responses one can get using these available drivers ( active and passive) using experience into a useable design.
What I have so far using WINisd and Unibox, with Isobaric loaded active drivers in a ported and Passive radiator design is:

Three examples.

1: Vb = 40.0 litres; Port tuning = 46 Hz; Vent diameter = 0.102m; Vent length = 0.210m
Giving -3db at 47Hz. (vented design)
2: (using Unibox) Vb = 60 litres; Passive Mms = 509.00 grams; Fsp = 14.46 Hz; giving a
-6db at 30 Hz
3: (using Unibox) Vb = 135 litres; Passive Mms = 719.00 grams; Fsp = 12 Hz; giving a
-8db at 20 Hz
( all approx. )

There are many other results that obviously can be achieved with variations of Vb and Fsp and Passive Mms, many of which to my untutored eye look viable. The last example I realise is pushing the boundaries a bit, but is included to illustrate the post that I am making.
 
Well it looks like that I accidentally discovered EBS alignments, as you might know if you read my last post. I've subsequently read as much as I can about the subject and this has given me a new avenue of possibilities to explore; all very interesting!
So at the moment I am thinking along the lines of two sub's both with Isobarik loading and using passive radiators instead of vents with an EBS alignment, the degree of which is yet to be finalised. Wish me luck!

Happy New Year

And finally, please raise a glass to a friend of mine who died after a fall on New Years Eve just two days after after I saw him.
 
My sympathies for the loss of your friend.

Maybe you have already read Audio Innovation - by Dan Marx www.danmarx.org. If not, it may help you in your decision for or against EBS.

What are your size limitations? 40, 60 or 135 l are quite different conditions to base your design upon.

What are the lowest notes in the music you usually listen to? For pop music a -3 dB between 40 and 50 Hz is usually sufficient. If you listen to orchestra or grand piano you may want something around 30 Hz. Lower than that is only desirable for great church organs or synthesizers. If you want the subs for home theatre you can follow the THX standard with -6 dB at 20 Hz.
 
Hi pacificblue and thank you for your sympathies.
yes, I have read the Dan Marks website and did find it very interesting. With the drivers that I want to use I am not going to get too ambitious so I am not going to try and force them to produce high volumes at 20 Hz. as I realise that high excursion and power become more and more important. ( see my first post for driver specs) For a rough idea a -6db point at 30Hz would be nice. I don't listen to anything very load.
As for box volume, as small as I can get away with according to what degree of alignment I finally go for.
Cheers
 
Hi pacificblue,
I had to cut my last post short because I was rudely interrupted.
In answer to your question; I actually listen to a bit of everything and occasionally DVD's but music is the prime motivation for some decently deep bass.
Using Unibox I am getting good responses with box sizes around 80-90 litres with -3db points around 30-35 Hz which is looking good. With some careful room placement two boxes of this size should give me all the bass that I really need. Some degree of room lift and perhaps a little equalisation should remove the "bass shelf" almost completely.
I will continue to re-work the design on WINisd and UNIBOX before I finally settle on exact dimensions and tuning frequencies.
I have now got to work out a cunning way of persuading my wife that these to boxes are going to be "absolutely critical" to our musical enjoyment!

I'm actually quite surprised that there is relatively so little in the way of discussion regarding EBS alignments, especially since most of what I have read about them is on the whole very positive.
 
Most folks won't tolerate the large cabs required and if they will, then nowadays it's usually in the form of a tower/column design [aka MLTL, MLTQWT], which are sometimes folded to get a low tuning.

So, how big/tall can you 'afford'?

GM

edit: For when corners are available, you'll be hard pressed to get more bang-buck/square ft. than with a 'BIB' pipe horn.
 
Last edited:
Hi GM,
I'm thinking about making the boxes long and low and hence visually unobtrusive if anything. At 27 litres per cubic foot they would only need to be around 48" long
(give or take) and 14" square to give the alignment that I might need or rather want.
Can you give me a linked example of a BIB pipe horn, out of interest.

Cheers
 
Greets!

Sounds like a plan! Many of the large pipe horns I did were mains speakers, equipment rack/TV platforms.

Hmm, 1.0 ft^3 = 28.31684659 L.

BIB basic layout attached.

BIB doc: http://www.quarter-wave.com/General/Fostex_FE-167E_BIB_Design.pdf

Main thread: http://www.diyaudio.com/forums/full...ose-fostex-craft-handbooks.html?highlight=bib

Plus you can search ‘BIB’ on the Loudspeaker forums link for some extra individual builds: Loudspeakers - diyAudio

Unfortunately, Godzilla’s consolidated BIB website is gone for the time being.

WRT the EBS, I’m assuming you still plan to make it isobaric and vent it with a PR. If so, making it a MLTL will acoustically lower its natural ‘shelf’ BW, ergo Fb. Using MJK’s MathCad software to accurately enough sim ¼ WL alignments, it’s down near 20 Hz.

This of course will increase the PR’s suspended mass and depending on the performance desired may cause it to easily bottom out due to the B139 theoretically not having enough usable excursion and piston area [Sd] even in pairs.

For my records: what is the driver’s make/model, Le, Xmax? Your available power and do you plan to wire them in parallel or in series?

GM
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.