New MJK Baffle Article

[jameshillj]

Ron,
Are these simulation problems or actual response non-linearities?
If "hands-on" and you're happy mucking about with PVC pipe, see if you can find some 10" or 12" storm water pipe (thin sidewall), cut it down lengthwise then screw one to each edge of bafflebrd thus adding a 10/12" curve to each side of the baffle - it'll add to the baffle dimensions and hence change low freq performance, but may prove useful.
I once built a full length 4ft curved 1" thick laminated baffle brd with parabolic sides that provided dramatically improved image - lot of trouble, lot of expense - could do very simple one in fibreglass using CLD (constrained layer design).
Could also incorporate driver time alignment (all voice coils equidistant to listening position) with mid/top driver mounted on a recessed part of the baffle - not difficult, really.

Hope this maybe of use.

 

[TerryO]

Ron,
Are these simulation problems or actual response non-linearities?
If "hands-on" and you're happy mucking about with PVC pipe, see if you can find some 10" or 12" storm water pipe (thin sidewall), cut it down lengthwise then screw one to each edge of bafflebrd thus adding a 10/12" curve to each side of the baffle - it'll add to the baffle dimensions and hence change low freq performance, but may prove useful.
I once built a full length 4ft curved 1" thick laminated baffle brd with parabolic sides that provided dramatically improved image - lot of trouble, lot of expense - could do very simple one in fibreglass using CLD (constrained layer design).
Could also incorporate driver time alignment (all voice coils equidistant to listening position) with mid/top driver mounted on a recessed part of the baffle - not difficult, really.

Hope this maybe of use.

IIRC, there was a lot of research carried out on edge diffraction a few years back on the old Bass List (Dave, GM?) that indicated that rounding over the edges of a baffle really was scarcely better than cosmetic until the edges had a radius of 3 inches or greater. The use of a parabolic curve probably is a good one.

As for time alignment, lining up voice coils really doesn't cut it. It may be better than nothing... maybe. The problem is that the acoustic center of a driver changes according to frequency (actually the phase) and the centers move forward or backwards. The best that you can do is to time align the drivers at the crossover frequency, which you can do it to a certain extent on the crossover itself. BTW: some very good designers don't worry very much about it at all.
😉

Best Regards,
TerryO

 

[GM]

I spouted some horn termination theory and a few real world examples from Altec's various published measurements/conclusions IIRC, but it was John Whittaker's and Rudi Blondia's extensive OB ribbon testing of the effects of different baffle shapes and later on after Linkwitz published his stuff on-line it was Don Maurer that did this baffle shape testing that advanced the subject at least WRT to simple shapes: http://home.pacbell.net/donwm/PrototypeTestResults.htm

Unfortunately, for reasons I don't recall, John removed his and Rudi's research off the net in a childish fit of anger and don't know if a complete copy has reemerged on-line, but some of it is on Rudi's excellent site: http://www.audioxstream.com/

GM

 

[ronc]

Are these simulation problems or actual response non-linearities?

\
Sims. I am old and my hearing isint what it use to be. And i kinda sorta have given up on actual response. However when i make an improvement based on sims i can usually hear a difference.
A simulation dosent give the total answer, but it does lead you in the correct direction. There are just too many variables to include in every simulation.

ron

 

[TerryO]

I spouted some horn termination theory and a few real world examples from Altec's various published measurements/conclusions IIRC, but it was John Whittaker's and Rudi Blondia's extensive OB ribbon testing of the effects of different baffle shapes and later on after Linkwitz published his stuff on-line it was Don Maurer that did this baffle shape testing that advanced the subject at least WRT to simple shapes: http://home.pacbell.net/donwm/PrototypeTestResults.htm

Unfortunately, for reasons I don't recall, John removed his and Rudi's research off the net in a childish fit of anger and don't know if a complete copy has reemerged on-line, but some of it is on Rudi's excellent site: http://www.audioxstream.com/

GM

GM,

It's been a while since AOL's new and improved spam filter knocked me off the Bass List, but I think Bob Stout had included John and Rudi's research data on his site or at least some way to access it.

I had completely forgotten about Don Maurer, this old age stuff is for the birds

Greg, do you still subscribe to the "List?" I sure miss some of those guys, what a bunch of crazy people.

Best Regards,
TerryO

 

[GM]

I had completely forgotten about Don Maurer, this old age stuff is for the birds

Greg, do you still subscribe to the "List?"

Greets!

I 'feel your pain'! 🙁 Anyway, Bob retired the basslist/DIYspeakers list on 11.15.07 and it's been replaced by this on-line forum, but apparently there's not much interest : http://forum.diyspeakers.net/viewforum.php?f=1&sid=33efe31ddc6dc2b898d2f40558e7fa21

Archives are supposed to be here, but it times out every time I've tried to access it: http://snippets.org/mailman/listinfo/diyspeakers

Ditto the LDSG, so couldn't browse it for John/Rudi's research.

GM

 

[ecir38]

I always use wayback machine to search old websites, see if this will work for you.

http://web.archive.org/web/*/http://www.diyspeakers.net/

 

[PigletsDad]

Suggestion for bass extension

As I understand it, deeper bass, and higher output capability in the midbass both require a wider baffle. This is not good for WAF.

The increased baffle width is needed for the 15", so why not put a couple of slanted wings that go back behind the baffle - they can also act to give support. If they are too deep you will get nasty resonances, but if they are say 8" deep, the effective baffle width goes up from 20" to 36", which should give about half an octave of extra extension.

Alternatively you could use a slightly narrower baffle (I suppose 17 to 18" is about as narrow as you can put a 15" driver in), and still not lose bass performance.

 

[MJK]

Re: Suggestion for bass extension

As I understand it, deeper bass, and higher output capability in the midbass both require a wider baffle. This is not good for WAF.

The bass response of an OB is set by the woofer Qts, the woofer SPL/w/m, and the baffle dimensions. It is a system so you cannot focus only on the baffle width. If balanced correctly you should be able to get 40 or 50 Hz bass from a 20" wide baffle.

The increased baffle width is needed for the 15", so why not put a couple of slanted wings that go back behind the baffle - they can also act to give support. If they are too deep you will get nasty resonances, but if they are say 8" deep, the effective baffle width goes up from 20" to 36", which should give about half an octave of extra extension.

Adding wings at a 90 degree angle to the front baffle will not automatically lead to a resonance problem. If the top is left open you can add very deep wings without exciting resonances that degrade the performance. The key is the acoustic impedance damping provided by the very large open area at the back of the winged baffle.

Alternatively you could use a slightly narrower baffle (I suppose 17 to 18" is about as narrow as you can put a 15" driver in), and still not lose bass performance.

Does this contradict the rest of your post? I am not sure I understand this last statement.

 

[PigletsDad]

Re: Re: Suggestion for bass extension

quote:
Alternatively you could use a slightly narrower baffle (I suppose 17 to 18" is about as narrow as you can put a 15" driver in), and still not lose bass performance.

Does this contradict the rest of your post? I am not sure I understand this last statement.

Sorry - rereading that, I have been pretty obscure.

What I meant was that if you make the front width of the baffle slightly narrower, the use of modest back wings can preserve the original effective width, and so keep the bass performance at the original point. Indeed one can probably even have one's cake and eat a little bit of it, by using a slightly smaller baffle, and reasonable size wings to get a slightly bigger effective baffle size and slight improvement in WAF at the same time.

I fully accept your point about the linkage between driver parameters, baffle width and crossover in this class of design, and that adding a wider baffle to an otherwise optimized design makes it worse, not better.

 

[chakija]

Would someone plz confirm if that it's no good using low Qtc FR driver ?

This one i would like to use...Qtc=0.23

If i use 2 of Alpha-s per side on 18x53 " baffle ,and it has 17" deep wings ,how does sensitivity response changes comparing only one Alpha in same baffle ?Would i be able to use more sensitive driver ,of 94 db ?

 

[MJK]

Re: Re: Re: Suggestion for bass extension

What I meant was that if you make the front width of the baffle slightly narrower, the use of modest back wings can preserve the original effective width, and so keep the bass performance at the original point. Indeed one can probably even have one's cake and eat a little bit of it, by using a slightly smaller baffle, and reasonable size wings to get a slightly bigger effective baffle size and slight improvement in WAF at the same time.

We agree. When analyzing this situation I run two simulations to bracket reality. First I run a simulation using just the front baffle, ignoring the wings, which provides the worst case solution. Then I run a simulation with the wings rotated to provide a big front baffle which provides the best case solution. Reality would lie between these two extremes.

 

[skorpion]

With 90 degree wings results will perhaps even be a bit better than Martins 'best' case. At least for degrees ranging from 45 to 90 for the wings in relation to the main baffle the result will be much closer to the best case. My own 'Volks-OB' is a very good example how good bass can be obtained with small means, and passive: http://www.audiocircle.com/circles/index.php?topic=46951.0

/Erling

 

[skorpion]

chakija,

I suppose you have some idea of how this speaker will sound. It is a bit built like the Lowther but not really. I think that the Sonido sensitivity is a bit below Alpha-15 but that can only be assesed by trying. And if so adjusted in the design. Otherwise I don't see any harm trying. If it doesn't work out you will have an excellent element for a Quarter Wave Pipe or Horn.

/Erling

 

[sreten]

Would someone plz confirm if that it's no good using low Qtc FR driver ?

This one i would like to use...Qtc=0.23

If i use 2 of Alpha-s per side on 18x53 " baffle ,and it has 17" deep wings ,how does sensitivity response changes comparing only one Alpha in same baffle ?Would i be able to use more sensitive driver ,of 94 db ?

Hi,

There is nothing wrong with using a low Qtc full range driver, in fact
any highish efficiency full range driver will have a low Qtc by default.

Two Alpha's on the same baffle are 6dB more voltage sensitive.
3dB more efficient and halve the impedance , another 3dB.

🙂/sreten.

 

[chakija]

THANKS !

I will definitely try them ,but i do not know which one i should chose.Smaller one (the one from a link above) is 92 db ,and i think that i would need to attenuate it to get proper response with 1 alpha ??
Larger one ,is 178mm in diameter and has 94 db.Now ,i can't afford 2 more alpha's right now ,but i think when i buy them i wont need any attenuation on this driver.Until i buy 2 more alphas ,i would listen them with attenuation ofc.
However ,if this smaller 92db driver doesn't need attenuation with one alpha ,then it would be mu choice because i don't have extra expense.
What should i do ? 🙄

 

[peterbrorsson]

Hi chakija!
I have SFR178 and tried them on an ob. Of course no bass at all, but rest of register sounded really good. No shouting and a nice and even response.+ a really nice treble.
It's difficult to make a judgement when bottom is missing,😀 however most drivers I listened to, without bass, sounds quite bad, but not these.
My BIB's are soon finished, then I'll see if my judgement is correct.

Maybe SFR178 together with this....http://www.traumboxen.de/kics/535rfpj.htm
or this....http://www.traumboxen.de/kics/pw-1001.htm

Dreaming........

Good luck!
Peter

 

[seanzozo]

Alternate woofer

I was thinking about doing a MJK-OBesque build this summer.

This cheap woofer seems ideal http://www.partsexpress.com/pe/showdetl.cfm?&Partnumber=290-386

Doing a rough calc in xbaffle and assuming PE's TS parameters are correct:
Adding 4 inches to baffle width (to accommodate the extra width of the 18 inch driver) results in a near identical curve to the Alpha 15A down to 50hz but with ~7hz greater extension.

Sean

 

[Norris Wilson]

Due to recent discussion here in an effort to obtain a higher efficiency OB. All based on Matin Kings fine design guide by using a fullrange with 92db+ efficiency ratings.

I was wondering if anyone has considered the Supravox 135LB fullrange driver?

http://www.supravox.fr/anglais/haut_parleurs/135_LB.htm

Possibly the use of two 12" Eminence Delta 12-B woofers would have a high enough efficiency to mate well with the 135Lb.

http://www.partsexpress.com/pdf/290-415s.pdf

The use of a single 15" Eminence Delta 15-B could work equality well, or better with a 16 ohm impedance when the use of tube SET amplifiers are taken into mind.

http://www.partsexpress.com/pdf/290-419s.pdf

In the name of efficiency and low distortion, I would propose using this type of OB as a monitor augmented by an active U, or W baffle subwoofer.
I would limit the bottom frequency of the OB to around 80Hz, where the low power of a SET could keep the SPL's up.

An in room response of around 25Hz to 18kHz or higher + -3db would be nice.

While I am thinking out loud, the thought of using a Fostex F120A mated to a pair of 12" vintage Alnico woofers per side with a mid Qts could sound nice as well.

Any thoughts, pros-cons, or experiences working along these lines?

Thanks

NW

 

[ronc]

was wondering if anyone has considered the Supravox 135LB fullrange driver?


Its not the efficency here, its the Qts.

ron