Omnipole, monopole, dipole and...nopole??

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
It probably won't be long until I find out if this question is a good one or a stupid one....the gales of laughter which may ensue will tell the story!!

Been following Lynn's thread beyond the ariels, and recently it has taken off, exhausting to keep up, and what's more it has been spawning offshoots!!

One thing that has piqued my interest is the one of power response, and whether or not the FR should be flat on or off axis etc. I do like the sound of that one and will check it out soon, will see what new information comes up on it before that.

Obviously, given the nature of Lynns thread the subject of dipoles and the difference with conventional boxes has come up. When reading about dipoles and how they work it all makes perfect sense, but my problem is that in practice the only two dipoles I heard did not impress me at all (one of them was the Orions, so that definitely suprised me).

That experience kinda left me a bit iffy on dipoles, but it was obviously not a very stringent test and probably insufficient to make sweeping general statements about the different alignments.

On the other hand, one thing that became obvious to me (others pointed this out, I'm not smart enough to work it out for myself) is that the back wave of a sealed driver must, to some degree, make it back out of the box through the driver cone itself. That is if we assume absorption internally is incomplete. This must degrade the sound, and possibly quite considerably.

Before I (finally!!) come to my question, whilst I can see the advantage of the dipole 'taking the side walls' out of the equation due to side cancellation, by the very nature of the beast are we not bringing (forcefully) into play the wall behind the speakers?? After all, we do in effect have a speaker deliberately aimed at it.

Lynn mentioned in passing the possibility of a lossy mesh etc, and even Sigfried mentions something like that on his website. It got me wondering if it could be used on 'normal boxes' too.

So, what theoretically to me would be the best of both worlds is a means of having no re-radiated sound coming thru the driver cone, and no sound radiated directly at the back wall.

What if my midrange chamber was built out of something like, I don't know, compressed fibreglass?? Forget for now the bass in a three way etc, ie the pressure not affecting the mid. The 'lossy box' would mean only a slight sealed box effect, yet allow the rear wave to escape and not be coming back through the cone. Any sound that makes it out rearwards thru the cone is hopefully attenuated severely and so does not reflect strongly off the wall.

So there is the question, not a dipole , not a monopole so maybe it is a nopole?

let the laughter begin
 
Terry J,

I have been EnABLing a pair of Lowther PM6A's over the last two weeks and have had them mounted on no baffle. Just sitting on a platform stand with a sheet of packing material under to help with the one close wall of the stand top plate. The magnet structure is supported by a plastic winding bobbin (from my business) so the cones are perpendicular to the room. I did not expect much in the way of satisfaction. I have been surprised.

There is not a "null" zone to the side and if you step behind the stands it sounds just as it would if you were standing behind the performers. Not muffled or lacking in frequency, but definately not the front side of the energy. As I have progressed and listened I have become quite enamored of this "complete" sound. The Lowthers are about two feet from a brick wall and I notice no back wave problems, just an amazingly full sound down to 80 Hz or so.

So I vote for a no baffle dipole, at least for the range above 120 Hz or so.

Bud
 
terry j said:
... the back wave of a sealed driver must, to some degree, make it back out of the box through the driver cone itself. That is if we assume absorption internally is incomplete. This must degrade the sound, and possibly quite considerably.


I totally agree! People often seem to only focus on making the box solid, while forgetting the most obvious exit point: the speaker cone itself. I've been thinking about ways to make sealed boxes more absorbent than the average "hollow box with bracing". So far, the best I've been able to come with is... (see the pic) and written up some comments about it on my site here.
 

Attachments

  • wedges.gif
    wedges.gif
    2.5 KB · Views: 493
thanks bud, have been watching your thread, not sure I understand it totally, probably one of those 'just do it' type of things.:)

Lech, yeah thats kinda what i was thinking. Might have to do the maths another day when I'm a bit more awake.

That was one way of trying to remove the rear wave, but I wonder how complete the absorption would be? On another point, but still related I suppose, can the T-S parameters of a midrange be used in box size calculations etc just as the bass drivers are?? I would suppose so, and then pushing this a little further, just as there are recommended qts values etc for a bass driver used in IB, or indeed OB, are there corresponding q values for the midrange as well which make them suitable for any given alignment??

Strangely enough, I would assume the MORE rigid the box is made the worse this particular problem might become, usual 'swings and roundabouts' that crop up in engineering!

So, given your illustration of the concept, and assuming incomplete absorption and so at least some of the sound coming back through the cone, what if (what I take to represent the solid box, ie the white area outside your wedges) instead of going thru the foam, hitting the rigid structure and reflecting back thru the foam and out the cone (albeit attenuated) it simply carried on out of the foam because there is no solid structure there??

In that case have we gained anything at all over going straight dipole?? My only guess is that the sound from the rear of the driver will not have much of a reflection off the wall behind the speaker.

My mains will be 'a la' the Sonus Faber Stradivari, about 750mm wide. Theoretically if this attenuation idea had any sort of merit then it may function kinda like an IB???, an IB in the room if you will.

In any case, have you done any experiments along this line and if so what have you found?
 
Terry j,

Your idea to come up with a "lossy box" design makes sense to me.

I know of one design that attempts just that. It is based upon a transmission line.
Normally, TL's utilize taperd sections and are lagged and tuned for a balanced bass.
The "lossy" design uses parallel sections, the xsectional area = the cone area.Great
attention was given to deflecting the sound waves round the "bends"; for this the design
used concrete with ceramic tiles.The whole thing was then well stuffed with sound absorbant material.

The idea, is that, by the time the rear wave finds free air it is so attenuated that it
will have little or no effect upon the front wave. The designer also claimed minimal
effect on cone movement due to the use of deflectors and because the constant
xsectional area would not create any backward pressure.

I haven't heard of anyone who's tried it. So I've no idea how effective it is.


Cilla


Bud,

I found your post amazing!

Every time I read something amongst these pages I have my intuitions challenged.
This time they have been confounded!

If I understand you right: are you saying that if an EnABLed Lowther PM6A were to be
simply mounted on some kind of stand , presenting itself in all its naked glory, it would give "an amazingly full sound down to 80 hz or so" ?

Ye Gods! The amount of music that I have in my collection... well, 80hz or so would do just fine.
How about the EX3 ( what I have) , would it work?


Cilla
 
Hi Cilla,

I will know more about Lowthers as time goes on. Lowther America tells me that a pair of 15 ohm PM6A's and a pair of A45's are on their way to me for treatment. The thread below goes into some detail about EnABL and Mamboni standing wave elimination treatments and long about post 250 a fellow from Taiwan begins some testing that not only supports the practice but adds some facts about treating systemic ringing events that is new knowledge.

http://www.diyaudio.com/forums/showthread.php?s=&threadid=100399&perpage=10&pagenumber=9

The PM6A's that are here go back to Lowther America on Thursday and are alleged to be slotted in as the main display drivers for Lowther at the RMAF this year. They are pretty amazing sounding, but they had gotten so electron hungry, in coherent small signal propagation, that I had to put a couple of Litz wire electron traps out on the ground lugs of the speakers. The sound was changing abruptly from lushly rich to sharply thin as more and more sources entered the picture, ie. symphony recordings, but were remaining very lush for single piano recordings, like Jessica Williams "Gratitude" CD.

These are the stupidest things I have ever invented. A loop of true Litz cable, 140 strands of #40 AWG magnet wire, 6 inches long, with three 0.6" pieces of shrink tubing shrunk along that length, and then formed into a loop with both ends tied together. The things allow enough electrons to loiter on the ground side to provide a huge increase in tiny signal information and a resultant increase in coherency of all information, but especially reflections, hall sound and internal detail on strong transients, like a piano fortissimo.

This idea did get some input at this thread, much quite dismissive, but it is such a weird idea that anyone with a rigorous understanding of mathematical electronics must dismiss it, as idiocy. It is, unfortunately, very noticeable if the speakers can deal with the coherency it allows.

http://www.diyaudio.com/forums/showthread.php?s=&threadid=102180

I am not sure if this would be noticeable with an untreated set of drivers, but the EnABL process makes a driver seemingly infinite in it's ability to reproduce coherent, informative detail, along with providing greater transient signal headroom.

So, go read that first thread on EnABL and as you come to the descriptions of how to treat a Lowther, with patterns and step by step pictures, you can decide if the treatment is for you.

As for how it sounds, well the electron traps are still "filling up" so ever greater levels of detail are being reconstructed (takes about 4 hours or so to complete, just by playing music) and I may even find I have too much trap space. The quality is just stupifyingly good. These are without doubt the best sounding speakers I have ever heard, once EnABL'd, and not at all shabby before treatment. The bass is quite good, deep into the piano left hand and allows all of a cello's bass string when Yo Yo Ma attacks on his Silk Road SACD.

it is obvious that bass viola's need an auxiliary driver and the Eminence Beta 8 looks like a very good addition, but on an open baffle, down to 40 Hz or so. In addition I am using a pair of Wright Sound 7 watt 300 B prototypes, with our power and SE output transformers, so that is also upping the information content while allowing safe Lowther power levels.

So, yes, naked Lowthers, but with a little EDGE developed OB for the Beta 8 below it.

Bud
 
Re: Re: Omnipole, monopole, dipole and...nopole??

CeramicMan said:
I totally agree! People often seem to only focus on making the box solid, while forgetting the most obvious exit point: the speaker cone itself.

The Linkwitz Pluto has both drivers (the Aura mini-driver used as a tweter is open backed) mounted in damped transmission lines with a measured 40dB return loss for the mid-bass driver.

I own both Plutos and Orions and the Plutos sound surprisingly similar.
 
cool Drew, will check it out.

You obviously have some experience with both, and bear in mind I haven't checked out what you've just said, but the pluto is not a true dipole as the orions are?ie more along the lines of 'removing' the back wave etc.

It may simply be a matter of terminology, but I always think of a transmission line as a means of tuning and deepening bass response, does it have that effect in the midrange as well or is it used in a different type of sense?

I will however get on over there and have a read a bit more closely on the pluto.

Does anyone have any data on whether or not you can nilly-willy throw any old midrange driver into an OB alignment?? or indeed a no-pole :cool: :cool: :cool: ie could it lead to excursion problems etc etc etc? (ie is it only suitable for some mids and not others).
 
terry j said:
That was one way of trying to remove the rear wave, but I wonder how complete the absorption would be?

Hi Terry, yeah it's hard to say, I was looking for information about anechoic chambers and they generally seem to be rated for reflecting about 1% of the incident power, or about 20dB down. But that's still very good.


On another point, but still related I suppose, can the T-S parameters of a midrange be used in box size calculations etc just as the bass drivers are?? I would suppose so, and then pushing this a little further, just as there are recommended qts values etc for a bass driver used in IB, or indeed OB, are there corresponding q values for the midrange as well which make them suitable for any given alignment??

It's an interesting problem because a low Qts seems to be better as it means that the speaker's motor provides a higher back EMF, which provides a larger opposing force to external influences such as sound pressure from inside the box. A stiff cone should be very helpful in that regard too.

However, a stiff cone combined with a low Qts is also a stronger 'nodal point', giving resonances a higher q-factor and placing greater reliance on damping materials inside the box. So there's a bit of give and take.


Strangely enough, I would assume the MORE rigid the box is made the worse this particular problem might become, usual 'swings and roundabouts' that crop up in engineering!

That too!

So, given your illustration of the concept, and assuming incomplete absorption and so at least some of the sound coming back through the cone, what if (what I take to represent the solid box, ie the white area outside your wedges) instead of going thru the foam, hitting the rigid structure and reflecting back thru the foam and out the cone (albeit attenuated) it simply carried on out of the foam because there is no solid structure there??

Actually I thought that the wedges themselves could be made of wood. They could be layered with felt or something like that, but the idea is to have a smooth horn-like transition between the air and the solid edges of the box. My theory is that the backwave would be funnelled into infinitesimally small points, where the extremely high SPL pressures and tiny displacement would be easily converted into heat.


In any case, have you done any experiments along this line and if so what have you found?

I want to try it out in practice as soon as I can! But at the moment it's just researching and theorizing.

With lossy boxes I think the hard part would be creating a good acoustic filter. Maybe a similar 'wedge' design could work, but you'd have to tune it somehow.
 
The surest way I've found to get rid of the rear wave is to reflect it into oblivion and absorb the small remainder. Do this by constructing a tallish pyramid or cone and mount the driver in what would otherwise be the base. Add a little stuffing, especially as the tip, and you've got a very dead box that turns the rear wave into heat.
 
thanks lech and john. seems you are both saying the exact same thing, except lech has the idea of using quite a few inverted pyramids. My initial thoughts of using simply an absorbing 'mat' seems to have a few problems eh? I can see conceptually the concentrating via the pyramid funnel to a very high pressure and then absorbing it as a a valid method.

The above approach sort of has a different feel than the conceptual one I had, namely that there is still pressure behind the cone, whereas in a type of OB (with rear absorption) that back pressure is removed.Hmmm

John, seein' as how you're the OB man (wasn't he on star wars?? Ob'man Kenobi or something), do you know if I can automatically use my mid in an OB type arrangement??

Leaving aside the sound escaping thru the cone bit, I would suppose that the mere fact of pressure behind the cone must distort the sound somewhat, in addition to the escaping soundwave.

Like you Ceramicman, I must do some experiments myself. The worst thread of all is a talking whatif? thread which ultimately goes nowhere!

The first thing for me to check out is the comparison of how the driver sounds on a large baffle (around 750 by 1200 mm, as that will ultimately be the size of the baffle on the mains) with a sealed box behind one, no box behind the other.

That will be a more valid test of 'sealed vs OB' than any previous listening to OB I've done.

JohninCR, there is there any reason for the driver to be less detailed in OB than sealed is there?? If not, then the lack of detail I heard on the Orions I can put down to driver differences rather than alignment differences I would assume.

Also John, what is your gut feeling about the disadvantages (if any) about the interaction with the wall behind the speakers in the normal dipole arrangement. That, after all, is kinda the starting point on whether or not there is any advantage in removing that by this method of trapping the back wave.

Ceramicman, are you into the car races?? Come next Bathurst, a whole bunch of audio (and car) nuts will be coming for an alcohol fuelled (not talkin bout the cars here), music soaked exhaust fumes and fun GTG. Praps you could make it down?
 
I have drawn up a speaker plan after months of reading and modelling that has a damped open back. This would probably satisfy your original requirement, i.e. no back waves bounced back through the speaker cone while the back waves are significantly attenuated from the upper bass up. With this hybrid approach the power response should be more even than the typical dipole speaker.

The idea is to have the "front panel" as narrow as possible so as to minimize diffraction <10ms. A 2"+ round over to smooth out diffraction, and angled side wings extending the effective baffle width such that the dipole eq monopole frequency is brought down to just above 100Hz. Some felt may be applied to the panel to further reduce diffraction. That back of the speaker would allow acoustic stuffing material of up to 120mm thick and is adjustable.

With my limited knowledge, I guess it has the best of both worlds and through extensive modeling, I have not found any faults with it so far. Things are looking good.

The only trouble is that I have not found anyone to do the wood work for me.

Regards,
Bill
 
HiFiNut is on the right track. I prefer to attenuate the HF of rear wave too, and considering that potenial, open alignments give you the ability to tune your speaker to your tastes, room, and placement with changes taking only a few seconds. No box affords that flexibility.
 
frugal-phile™
Joined 2001
Paid Member
The kind of TL people are talking about are aperiodic TLs... the idea being to completely (or close to it) absorb the back wave (in the vernacular, "suck the back wave down the pipe") This can be done with a 1/4 wave line or with a half-wave line (this is what the B&W nautilus uses)

NautilusTLs.jpg


This one is for a bass driver (their lit has lots of good idea provoking pics), for a mid it doesn't have to be all that long (~1/2 the w/l of the lowest frequency of interest) You want to avoid having surfaces that reflect back at the cone,,, probably the biggest contributor is the driver itself. One has to be even careful with absorbtion material behind the driver as it is reflective at certain frequncies -- there should be none near the driver and get progressively denser as you approach the terminus...

Easiest implementations are a big PVC pipe firing out the back othe box or a simple folded & tapered TL with the terminus just by the driver.

dave
 
terry j said:
Ceramicman, are you into the car races?? Come next Bathurst, a whole bunch of audio (and car) nuts will be coming for an alcohol fuelled (not talkin bout the cars here), music soaked exhaust fumes and fun GTG. Praps you could make it down?

Sounds like a plan! :D
I'm quite new to Oz so I definitely need to see Sydney at some point :)

planet10 said:
One has to be even careful with absorbtion material behind the driver as it is reflective at certain frequncies -- there should be none near the driver and get progressively denser as you approach the terminus... [/B]

What about putting damping material only on the convex edge of the line so that the sound waves 'curl' inwards?
 
hi ceramicman

try this link http://www.stereo.net.au/forum/index.php?showtopic=2528, you may have to register, but if you do then welcome aboard!!

Dave, re your pic, does the TL pictured load the driver in anyway?? still trying to get my head around whether or not removing the back wave w/out pressure on the cone (a la OB or IB) will sound different than removing it with pressure (eg very heavily damped box). Time for my
experiments I would assume!

BTW, when you say terminus does that mean the exit of the TL? If the soundwave exits the 'tube', where in the picture does it exit (after the other skin is added).
 
frugal-phile™
Joined 2001
Paid Member
terry j said:
Dave, re your pic, does the TL pictured load the driver in anyway?? still trying to get my head around whether or not removing the back wave w/out pressure on the cone (a la OB or IB) will sound different than removing it with pressure (eg very heavily damped box). Time for my
experiments I would assume!

BTW, when you say terminus does that mean the exit of the TL? If the soundwave exits the 'tube', where in the picture does it exit (after the other skin is added).

This box could be considered sealed as far as loading goes, but i expect B&W designed a driver where they got very little pressure from the back-side.

This TL is half-wave so doesn't have an opening to the outside world, so terminus would refer to the very end of the spiral.

Personally i'd adopt a 1/4 wave line damped until it was aperiodic.

dave
 
terry j said:
cool Drew, will check it out.

You obviously have some experience with both, and bear in mind I haven't checked out what you've just said, but the pluto is not a true dipole as the orions are?ie more along the lines of 'removing' the back wave etc.


The Pluto is a sealed speaker which approximates a true monopole up to 3KHz.

http://www.linkwitzlab.com/Pluto/intro.html

I speculate that they sound similar because polar response is relatively uniform w.r.t. frequency and neither has a cabinet storing energy which is re-radiated through the drivers.

It may simply be a matter of terminology, but I always think of a transmission line as a means of tuning and deepening bass response, does it have that effect in the midrange as well or is it used in a different type of sense?

A transmission line directs transmission of energy (electromagnetic, acoustic) from one point to another. The Pluto uses a terminated transmission line like the B&W Nautilus where sound waves start at the driver, bounce off the end of the transmission line, and loose 99% of their energy by the time they make it back to the cone. It's just used to control the backwave and not provide bass output.


Does anyone have any data on whether or not you can nilly-willy throw any old midrange driver into an OB alignment?? or indeed a no-pole :cool: :cool: :cool: ie could it lead to excursion problems etc etc etc? (ie is it only suitable for some mids and not others).

Read Siegfriend's site http://www.linkwitzlab.com. Since dipoles roll-off at 6dB/octave, maintaining constant output requires boosting displacement by 2X for each octave and SL uses 8" midrange drivers with beefy tweeters and 1440Hz LR4 cross-overs.
 
aarrgh

had just talked myself into believing that there is no real danger in trying my mids in OB type baffle, you go and put the fear of the 'god that makes smoke come out of drivers' into me!

I put up a post on what parameters etc are suitable for a mid in OB over on the audio circle open baffle forum, received only one reply which is a bit of a bugger. So I had decided to just do it.

I don't mind doing a bit of research, but the trouble with Seigfreids site is it is so HUGE!:boggled:

Is there any reason that using WinISD would not work for midrange drivers?? I only ever see it being applied to bass drivers, quite naturally I suppose, but if I plugged in (say) 1000 l for a sealed box it should at least show it's IB behaviour, and that way I might be able to get a feel for the excursion vs extension of the driver, which is my main concern and which is the potential problem that was echoed by Drew.

If from my fumbling attempts at modelling look 'safe' for the driver, I will mock up a quick OB for the mids next week and have a quick listen.

thanks for replies so far, much appreciated.

I don't want to damage the driver, I only got a replacement yesterday would you believe, I've been music-less for the last three weeks, so to damage it as a result of my own stupidity would be even more annoying.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.