When you were at school there were relatively few marks for stating an answer with most marks assigned for stating the principles and assumptions used to work out the answer. There were good reasons for this. When there is disagreement, as there is in this thread, looking at those principles and assumptions is how to determine where the disagreement has arisen, fix it and make progress. Of course, this only works if people share the same scientific principles on which to build. It doesn't work if people draw conclusions based on magical beliefs like painting patterns of dots on cones but it might work for this discussion.
I've been following all this whilst building a new speaker. I used Visaton W200S bass and DT94 tweeter on BW3 reflex in a simple old Sony 57H x27W x22D 20mm Chipboard cabinet.
Cabinet sides and top and bottom lined with rubber and cork Regupol sound insulating sheeting glued on. Sony just add some tiny 2" long fillets at the side of the bass mounting and top and bottom on the front panel. Strengthening really. Sheet of BAF wadding thrown in.
Nothing sophisticated. It all works very well. The tap test is quite low frequency.
The third image shows a BIG MISTAKE I made with another cabinet in my more naive noobie days. I put two pine battens across the front panel thinking it would strengthen it. It merely added a very unpleasant wooden colouration. I must cut/chisel them out.
Cabinet sides and top and bottom lined with rubber and cork Regupol sound insulating sheeting glued on. Sony just add some tiny 2" long fillets at the side of the bass mounting and top and bottom on the front panel. Strengthening really. Sheet of BAF wadding thrown in.
Nothing sophisticated. It all works very well. The tap test is quite low frequency.
The third image shows a BIG MISTAKE I made with another cabinet in my more naive noobie days. I put two pine battens across the front panel thinking it would strengthen it. It merely added a very unpleasant wooden colouration. I must cut/chisel them out.
Remark:
The basic excitation mechanism of a structure capable of modal vibration can be e.g. taken from
"Introduction to distributed mode loudspeakers (DML) with first-order bahavioural modelling"
by N.J. Harris and M.O.J. Hawksford
IEEE Proc. - Circuits Devices Syst. Vol. 147, No. 3, June 2000
Figure 4 can be used to illustrate excitation of a cabinet or panel by a dynamic loudspeaker driver.
But the masses "Mc" and "Mm" may be "swapped" in the picture for our purposes:
- the mass "Mc" thought as coil's mass from a moving coil exciter attached to the panel can be seen as the mass of "Motor+Basket" of our driver attached to the cabinet
- the mass "Mm" thought as the "seismic" mass of the magnet of a moving coil exciter can be seen as the "seismic mass" of our drivers "coil+cone".
This simplification holds as long as our drivers Basket+Motor and also the moving cone can each be viewed as lumped masses, without internal vibration.
Basically unwanted radiation from a loudspeaker cabinet can be modelled as a kind of (3D) bending wave panel, driven by an exciter that has high moving mass and low seismic mass.
Roles are kind of swapped between seismic mass and moving mass compared to a bending wave exciter, as we attach magnet and basket to the cabinet's baffle instead of the coil.
The basic excitation mechanism of a structure capable of modal vibration can be e.g. taken from
"Introduction to distributed mode loudspeakers (DML) with first-order bahavioural modelling"
by N.J. Harris and M.O.J. Hawksford
IEEE Proc. - Circuits Devices Syst. Vol. 147, No. 3, June 2000
Figure 4 can be used to illustrate excitation of a cabinet or panel by a dynamic loudspeaker driver.
But the masses "Mc" and "Mm" may be "swapped" in the picture for our purposes:
- the mass "Mc" thought as coil's mass from a moving coil exciter attached to the panel can be seen as the mass of "Motor+Basket" of our driver attached to the cabinet
- the mass "Mm" thought as the "seismic" mass of the magnet of a moving coil exciter can be seen as the "seismic mass" of our drivers "coil+cone".
This simplification holds as long as our drivers Basket+Motor and also the moving cone can each be viewed as lumped masses, without internal vibration.
Basically unwanted radiation from a loudspeaker cabinet can be modelled as a kind of (3D) bending wave panel, driven by an exciter that has high moving mass and low seismic mass.
Roles are kind of swapped between seismic mass and moving mass compared to a bending wave exciter, as we attach magnet and basket to the cabinet's baffle instead of the coil.
Remark II:
Assuming in Figure 4 "Rp" and "Xp" as being constant, which at least holds for high frequencies, where the modal overlap factor is high, we see basically a 1st order lowpass filter function for the force acting upon Rp and Xp.
This is the main reason for wall velocity decreasing with frequency in vibrating loudspeaker cabinets: The exciting force decreases with frequency in tendency, as long as the drivers components do not vibrate inherently.
This may hold up to the upper bass region ... which of course is of high interest here.
>> Interpreting wall velocity decreasing with frequency as "rising dissipation in the cabinet's walls" is a severe misunderstanding, as the Qs of modes are quite independent from modal number and thus frequency.
Advocating bracing without damping (increasing stiffness) for non subwoofer enclosures is thus based on two main misunderstandings
- misinterpreting lower velocity levels (on average) of the walls at higher frequencies as being due to "dissipation", which is not the case
- misinterpreting e.g. Fryer's studies in a way that "higher Q of resonances" should be "less audible" in mechanical systems, which is not the case, as the contrary is true:
Higher Q means higher level of the resonance, which means higher audibility.
Furthermore advocating bracing without damping for "non subwoofer enclosures" is based on neglection of
- dispersion of bending waves and coincidence frequency, as "faster" (stiffer and low mass) materials and structures radiate sound more efficiently, which is counterproductive in a loudspeaker cabinet
- resonances shifted up in frequency are more audible
- Qs of resonances rise, when "reactance per resistance" rises, which is the case when just stiffness of a panel or wall is increased without increasing damping at the same time to maintain the Q.
(Stiffness goes cubic with wall's thickness while mass goes linear.)
Kind Regards
Assuming in Figure 4 "Rp" and "Xp" as being constant, which at least holds for high frequencies, where the modal overlap factor is high, we see basically a 1st order lowpass filter function for the force acting upon Rp and Xp.
This is the main reason for wall velocity decreasing with frequency in vibrating loudspeaker cabinets: The exciting force decreases with frequency in tendency, as long as the drivers components do not vibrate inherently.
This may hold up to the upper bass region ... which of course is of high interest here.
>> Interpreting wall velocity decreasing with frequency as "rising dissipation in the cabinet's walls" is a severe misunderstanding, as the Qs of modes are quite independent from modal number and thus frequency.
Advocating bracing without damping (increasing stiffness) for non subwoofer enclosures is thus based on two main misunderstandings
- misinterpreting lower velocity levels (on average) of the walls at higher frequencies as being due to "dissipation", which is not the case
- misinterpreting e.g. Fryer's studies in a way that "higher Q of resonances" should be "less audible" in mechanical systems, which is not the case, as the contrary is true:
Higher Q means higher level of the resonance, which means higher audibility.
Furthermore advocating bracing without damping for "non subwoofer enclosures" is based on neglection of
- dispersion of bending waves and coincidence frequency, as "faster" (stiffer and low mass) materials and structures radiate sound more efficiently, which is counterproductive in a loudspeaker cabinet
- resonances shifted up in frequency are more audible
- Qs of resonances rise, when "reactance per resistance" rises, which is the case when just stiffness of a panel or wall is increased without increasing damping at the same time to maintain the Q.
(Stiffness goes cubic with wall's thickness while mass goes linear.)
Kind Regards
Last edited:
LineArray, no one (well at least I am not) is advocating additional bracing without damping. By dissipated, I mean damped as that is exactly what damping is intended to do. Take the semantics however you wish, but nothing stated here goes against fundamental acoustics, loudspeaker theory or otherwise.
It is the notion that having a well braced loudspeaker is going to create more distortion that I take issue with. Any competent designer would understand the purpose of bracing the cabinet (i.e. to resist motion and excitation imparted from the drive units) and does not do so singularly. I believe that is where the misunderstanding lies. I don't believe there is an industry conspiracy or industry eye wool regarding what a handful of people understand and what everyone else does WRT cabinet stiffness.
It is the notion that having a well braced loudspeaker is going to create more distortion that I take issue with. Any competent designer would understand the purpose of bracing the cabinet (i.e. to resist motion and excitation imparted from the drive units) and does not do so singularly. I believe that is where the misunderstanding lies. I don't believe there is an industry conspiracy or industry eye wool regarding what a handful of people understand and what everyone else does WRT cabinet stiffness.
Let me put it this way:
If i was to weigh some commercial loudspeaker cabinets and ask the question:
"Is this cabinet's radiation - auditively weighted - as low as can be achieved using this mass of the cabinet ?"
Then my answer would be "Clearly not !" in almost all of the cases.
- many subwoofers are still not stiff enough, mainly because too large in many cases for using the "push modes out of band" strategy successfully.
- in most multiway designs (bass enclosure used more widebanded to upper bass or even midrange) there is excess stiffness and in relation too less "damping&mass"
Why i weigh the cabinets ?
Because i want to be fair:
If i would say "use more mass and more damping" i could even say "use a more expensive cabinet". Of course it is more easy to find solutions with "cost doesn't matter" kind of budget.
If i was to weigh some commercial loudspeaker cabinets and ask the question:
"Is this cabinet's radiation - auditively weighted - as low as can be achieved using this mass of the cabinet ?"
Then my answer would be "Clearly not !" in almost all of the cases.
- many subwoofers are still not stiff enough, mainly because too large in many cases for using the "push modes out of band" strategy successfully.
- in most multiway designs (bass enclosure used more widebanded to upper bass or even midrange) there is excess stiffness and in relation too less "damping&mass"
Why i weigh the cabinets ?
Because i want to be fair:
If i would say "use more mass and more damping" i could even say "use a more expensive cabinet". Of course it is more easy to find solutions with "cost doesn't matter" kind of budget.
Last edited:
lemans23 said:
Maybe not a "conspiracy", more often a kind of inertia or "dullness":
Let's use the same sh** as always:
> 19mm MDF with nice veneer outside
> some braces
> a 2-3 mm polymer bitumen mat inside, where we do not have to cutout complex forms
> Ready.
Such a cabinet may often be "not too bad". But it is far below, what can be achieved using matched materials and composites.
Using some creativity - e.g. making use of materials already common in several fields of architectural acoustics - even rather low cost solutions of high quality are possible in making "acoustically high insulating" loudspeaker's cabinets.
edit:
The pics at the bottom show a "nearly perfect" subwoofer cabinet, to be used at about 80Hz crossover frequency:
http://www.dipol-audio.de/projekt-schwingungen-an-lautsprechergehaeusen.html
The lowest mode is found at 220Hz.
Material: Just 19mm MDF, no bracing, no frills. Just adequate form and material.
Of course this cabinet would be inacceptable as a 2-way ...
Last edited:
I'm with you 100%. The challenge is that the easiest way to increase cabinet mass (thicker walls) will in every case increase flexural stiffness as well. The trade off is worth it, because the materials being used typically have both A. a large amount of internal damping and B. are used in conjunction with damping.
I could not agree further on subwoofers having too little mass and stiffness. We had a piece of competitor product in here the other day, large sealed box, 15mm MDF all around with no bracing added. 12" driver with a 500W amp at that. The box pulses like it's alive.
I do not know if you are active in the industry (especially now), but the challenge for loudspeaker engineers isn't their understanding of science. It's what they're allocated via budget. Even optimizing cabinet performance often comes second to ease of mass production assembly. It's then up to the engineer to optimize given that (s)he doesn't have unlimited reign. But woe is the world, no?
This is why there are two schools of thought. Heavy or light.
What do you think of the Spendor BC1, LineArray? I like it myself. Very BBC. The cabinet is 12mm birch ply. Damping panels are wood fibre impregnated with bitumen. Then some loose foam lining rather than stuffing because it is reflex.
Spendor BC1
No braces at all.
Isn't the main theory that a rigid light damped cabinet can't store much energy? And won't give you a hernia when you lift it.
What do you think of the Spendor BC1, LineArray? I like it myself. Very BBC. The cabinet is 12mm birch ply. Damping panels are wood fibre impregnated with bitumen. Then some loose foam lining rather than stuffing because it is reflex.
Spendor BC1
No braces at all.
Isn't the main theory that a rigid light damped cabinet can't store much energy? And won't give you a hernia when you lift it.
system7 said:
... i guess the cabinet is basically a way to go e.g. in a smallish 2-3 way speaker.
I would have used other materials for mass and damping on the panels.
And - i already said that - i see no reason not to use stuffing material in a BR:
One just has to keep that material out of special zones like around the inner port (hooks, nets, grids etc. are great helpers).
So i would have used a different strategy and materials for stuffing too ...
____
edit: Building such a "BC1 style" cabinet for private use, my version would have considerably more mass in the end. Applying that "how good is the actual mass invested" criterion, i feel that spendor "BC1" is a "better than usual" cabinet. If so, "why this style has not established itself?"
Possible answer: Emotions and "contrafactual intuiton" may often have stronger impact on marketing than knowledge, wisdom or even excellence in engineering ...
Last edited:
Birch ply is expensive. I think that's why you don't see it much.
I've done experiments with stuffed reflex, and it does have a nice midrange. I guess it becomes "aperiodic" or "Damped reflex" at the bass end. A bit of a halfway house between reflex and closed box that might fit room gain better. Steen Duelund liked aperiodic.
I believe there is a notion that panel damping material should ideally be around twice the thickness of the walls, and attached with glue or staples. It certainly works well.
http://www.diyaudio.com/forums/multi-way/223174-interesting-read-i-found-lossy-cabinet-designs-harbeth.html#post3234256
TBH, my own recent design is crying out for some eggcrate lining foam too. Just don't have any lying around.
Do you have any photos of the heavier braced designs?
I've done experiments with stuffed reflex, and it does have a nice midrange. I guess it becomes "aperiodic" or "Damped reflex" at the bass end. A bit of a halfway house between reflex and closed box that might fit room gain better. Steen Duelund liked aperiodic.
I believe there is a notion that panel damping material should ideally be around twice the thickness of the walls, and attached with glue or staples. It certainly works well.
http://www.diyaudio.com/forums/multi-way/223174-interesting-read-i-found-lossy-cabinet-designs-harbeth.html#post3234256
TBH, my own recent design is crying out for some eggcrate lining foam too. Just don't have any lying around.
Do you have any photos of the heavier braced designs?
Last edited:
system7 said:
Not at all, if you avoid "overdoing" it with the stuffing, and let the port "breathe" - also the stuffing behind the driver has great impact on the upper resonance - you can have a "rather high Qb" BR-enclosure, having bass response "as designed", even with considerable stuffing.
Around the inner port and behind the driver are the zones to be controlled by the mentioned helpers. That's all to it in IMHO ...
Last edited:
I don't now whether this "shearing brace" has been mentioned here already:
US 7,270,215 B2
Patent US7270215 - Loudspeaker enclosure with damping material laminated within internal ... - Google Patents
I used some similar "dampening braces" even before that patent was issued - never mind - but Mr. Styles has worked out some interesting variants IMO.
The whole thing lives from proper dimensioning e.g. of the dampening layer. Also this technique needs some vibrational analysis to be applied efficiently IMO: It is not a method to be applied "simply without thought" and there is IMO always just a subset of modes to be adressed with a special brace (i would say "damper" here instead of "brace").
This is why i do not forecast high commercial interest in this particular patent: But maybe i am wrong.
Nevertheless i like the inventor's ideas and sketches.
Furthermore: Some of those braces can easily introduce additional modes of their own (hehe ...)
Last edited:
... Regarding the above patent:
At least, to me it is aiming for some kind of "constraint layer damping" to be applied to a cabinet that is "basically conventional" in construction to save costs. But i am not sure, whether the above patent might achieve that fully.
I have found a way to do something similar in a much more simple way, needing less analysis in construction and less effort in application. Usable materials are available commercially already and also moderately priced. Preferred materials do not imply environmental riscs and are easy to apply. It is just the transfer step(s) for specific application in loudspeaker cabinets, which is missing.
Interested parties and investors are welcome, as for myself i won't be filing any wider ranging than german patent ... for sake of costs.
At least, to me it is aiming for some kind of "constraint layer damping" to be applied to a cabinet that is "basically conventional" in construction to save costs. But i am not sure, whether the above patent might achieve that fully.
I have found a way to do something similar in a much more simple way, needing less analysis in construction and less effort in application. Usable materials are available commercially already and also moderately priced. Preferred materials do not imply environmental riscs and are easy to apply. It is just the transfer step(s) for specific application in loudspeaker cabinets, which is missing.
Interested parties and investors are welcome, as for myself i won't be filing any wider ranging than german patent ... for sake of costs.
Last edited:
And now a public proposal - surely not mature - just to think about:
What about integrating rope dampers - multiple shapes, sizes and materials are possible - into enclosures ?
http://www.enidine.com/core/medialibrary/engvalves/website/CPTwiregrp.gif
http://img.directindustry.com/image...solator-anti-vibration-mount-7068-4097539.jpg
As an alternative to "bracing" e.g. ...
Kind Regards
What about integrating rope dampers - multiple shapes, sizes and materials are possible - into enclosures ?
http://www.enidine.com/core/medialibrary/engvalves/website/CPTwiregrp.gif
http://img.directindustry.com/image...solator-anti-vibration-mount-7068-4097539.jpg
As an alternative to "bracing" e.g. ...
Kind Regards
Last edited:
Hi may i ask what "bracing" has done to you ?
Is there a reason to hate bracing ?
I had a look to the cabinets of REL sub-woofers. I know that they are very very good because i listened to them in many occasions.
Simply put, they have all bracing ... designed of course rightly but still there is bracing in all.
I have come the conclusion that bracing in some cases (sub and bass enclosure) is mandatory, in other cases (limited range speakers) maybe avoidable.
But if you have a single cabinet full-range speaker you need bracing because the lower driver has the same requirement of the sub-woofer cabinet.
And i agree that with bracing, as you shift vibes freq higher, the same vibes will be much easier to cancel out even with some moderate use of butyl pads.
They will be less "energetic".
An externally hosted image should be here but it was not working when we last tested it.
Kind regards, gino
P.S. of course a mini-monitor with a let's say 4" woofer does not need extensive bracing, having no real bass to speak of.
Last edited:
ginetto61 said:
Hi gino,
yes there are reasons, as most actual bracing is in my view wrong in at least 3 respects
- wrong executed
- indicating wrong design or dimensioning elsewhere
- superflouus as effect is mostly not documented by measurement but "just believed in"
I am a "data driven person" not a "story teller" you know and i am talking about "pure bracing" without damping here.
If you manage e.g. to push the lowest modes in a subwofer cabinet from say 120Hz (unuseable cabinet even for a sub then IMO) to 220Hz (far better with e.g. XO at 80Hz and say 18dB/octave) by using bracing solely, than i would say: "Just do it, i am fine with it."
The only thing: I'd like to see measurements that show clearly, that this has really been achieved.
I love (my!) things simple:
When a subwoofer - i mean for home(!) use - is so large, that it needs bracing, it is a misconstruction to me and most probably will have even further problems (e.g. resonances of the enclosed air volume) coming with size. This is one reason why we have a rather large furnace for wood: It can convert misaligned speaker cabinets and prototypes into heat.
(BTW: Yes, i know this kind of pads)
Kind Regards
Last edited:
Very interesting discussion. It makes you think.
I just looked up the classic Rogers BBC birch ply designs at Mark Hennessy's Site. Surprisingly, most of them are reflex. The only exception is the closed box BBC LS3/5A that causes so much consternation to this day.
I do suspect that the standard Golden ratio cabinet designs don't have much need for bracing. I have already mentioned certain personal disastrous bracing attempts. When you move to odder ratios like Troels Gravesen's tall TQWT, then the tapered line cross brace serves several functions.
planet10's Onken designs have built-in bracing too.
Cabinet design is a bit like trial and error cathedrals or ships built before computer modelling. They have evolved, because if the cathedral or ship design was top heavy, they fell down or turned turtle and sank.
My own conclusion is that you should try to use the same cabinet material throughout, with small fillets just for strengthening, and keep everything as light and rigid as possible. Then apply panel and standing wave damping as appropriate.
It's one of those oddities that rolled off closed box is much easier to align with room gain. With flatter reflex, we often end up doing stuff like damping the port with straws or wadding, because it booms more than you might expect. I really used to do that stuff with straws in the port myself! These days I'd just reduce the bafflestep.
I just looked up the classic Rogers BBC birch ply designs at Mark Hennessy's Site. Surprisingly, most of them are reflex. The only exception is the closed box BBC LS3/5A that causes so much consternation to this day.
I do suspect that the standard Golden ratio cabinet designs don't have much need for bracing. I have already mentioned certain personal disastrous bracing attempts. When you move to odder ratios like Troels Gravesen's tall TQWT, then the tapered line cross brace serves several functions.
planet10's Onken designs have built-in bracing too.
Cabinet design is a bit like trial and error cathedrals or ships built before computer modelling. They have evolved, because if the cathedral or ship design was top heavy, they fell down or turned turtle and sank.
My own conclusion is that you should try to use the same cabinet material throughout, with small fillets just for strengthening, and keep everything as light and rigid as possible. Then apply panel and standing wave damping as appropriate.
It's one of those oddities that rolled off closed box is much easier to align with room gain. With flatter reflex, we often end up doing stuff like damping the port with straws or wadding, because it booms more than you might expect. I really used to do that stuff with straws in the port myself! These days I'd just reduce the bafflestep.
Hi Oliver and thank you sincerely for the very kind and valuable reply.
I think i see your point. You may mean that some speakers are unnecessarily over-braced ? that could be possible of course.
Especially in the "high-end" there is a tendency to overdesign the equipment.
Moreover I agree completely with you when you say
I trust a lot well done measurements as well.
I have seen measurements on cabinets carried out by the Stereophile magazine with accelerometers fastened to the side panels.
If these kinds of measurements can be considered valid i have seen that braced cabinet usually behaved very well
I do not think that the bracing should be monstrous by the way.
Even the subs i was mentioning seem to have quite thick panels and light bracing in the strategic place.
I guess if you go with thinner panels you have to increase bracing ? i do not know for sure.
Just a last question. Have you performed tests like the Stereophile tests on some speakers ?
that could be the decisive test.
Just as an example i am attaching the page with the test on a braced speaker
B&W Matrix 801 Series 2 loudspeaker Measurements | Stereophile.com
look at fig.2 and fig.3.
Maybe as you say very similar results can be obtained with less bracing and more damping panels ... of course.
Thanks a lot again.
Kindest regards, gino
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.