chops said:
Chops, that is an apples-oranges comparison! A U frame with half the depth of a H frame has the same point where dipole cancellation occurs ie fequal. They are not the same thing. Sure if you make your H frame smaller in depth, then you get more roll-off.
The difference is documented clearly with measurements on John K's website where he discusses his Nao design which uses a U frame.
The downside to the U frame is that it lacks the figure of 8 polar response.
Can you show a link to that thread? It sounds very interesting.
paulspencer said:
It's not a downside, just a difference that is irrelevant in most cases, since no one places their dipole bass units in a way that figure 8 dispersion can even exist. John Sherrin's tests, though flawed in some ways, showed that the change in dispersion pattern wasn't significant, at least with relatively shallow U's. My experience with U-Baffle bass is the same, with only a slight decrease and change in position in the deep null at the sides, and it has not been enough to side wall and floor/ceiling nulls to rear their ugly head.
I'd also like to point out that virtually every H, U, W, or N baffle that I've seen is significantly flawed in that there is no bracing. With bass units the panel vibrations are quite significant without proper bracing, and this would never be acceptable in any other cab design, but for some reason it is totally ignored with OB.
Very interesting. Do you mean John Sheerin? Have you got a link to the tests? How are they flawed?
For an "all out" design, it would make sense to attach the drivers to a metal structure to support them, attached via their magnets. They could have a resilient connection to the baffle, but not a mechanical connection allowing significant vibrations to occur. I've seen this done with mid drivers, but not for bass.
For an "all out" design, it would make sense to attach the drivers to a metal structure to support them, attached via their magnets. They could have a resilient connection to the baffle, but not a mechanical connection allowing significant vibrations to occur. I've seen this done with mid drivers, but not for bass.
Paul,
Yes, sorry I misspelled John's name. Nearfield measurements of any OB are invalid unless you use them at that nearfield distance. In the nearfield, D is significant in comparison to the measuring distance, and especially with flat baffles (not relevant to John's work, but important to consider) D can be much greater in the nearfield than the farfield.
If D is significant in comparison to measuring distance, then the SPL of the 2 waves can be significantly different, which isn't true in the farfield. eg 2m wide flat baffle, so D is 1m and a 1m measuring distance. The rear wave must travel 2.4m to get to the mic, but the front wave travels only 1m. Since sound decreases by 6db per doubling of distance the rear wave, the front wave will be 7 or 8db louder than the rear wave greatly reducing the impact of the rear wave on the total sound. If you measure the same baffle at 5m, the rear wave will have travelled 6m and the front wave 5m, so the spl difference would be negligble. The bottom line is that in the nearfield OB's have significantly more bass than in the farfield, ignoring room effects.
John's test results are in a thread here on DIY that he started about the subject. He was trying to test whether it's true that U-baffles of the same size as an H baffle have a +6db edge over the H baffles. Apparently SL and JohnK had a big argument over this long ago, and is one of the reasons SL no longer posts here. The funny thing is that SL's formula for Fequal proves it, so why argue? If you take an H baffle and turn it into a U-baffle with the same overall dimensions, D doubles and Fequal is one octave lower or +6db.
Regarding bracing, I'm not talking about driver mounting structures. I'm talking about large unbraced panels. While the sound waves leaving the baffle don't exert much force on a flat baffle, the same is not true for other OB shapes. Once you constrain the waves the forces acting on the panels is significant and very similar to a box. With box speakers, it's not the change in absolute air pressure that makes bracing necessary. It's the sound/pressure waves acting on them that causes vibrations. The difference between boxes and OB is that boxes have reflections to deal with too. Most people believe that since OB's are open the forces on panels are minimal, but this simply isn't true. Plus boxes at least have their side panels reinforced by the rear corner, but you never see bracing at the rear edge of an OB. As a result, they "flap in the wind".
A prime example is Linkwitz's dipole woofer (not a subwoofer btw), which is copied so often http://linkwitzlab.com/woofer.htm . The 2 side panels have no bracing whatsoever and anyone who has built one can tell you that those panels vibrate quite significantly. A couple of braces across the back and a couple of rib braces running from front to back make it a much more solid design.
I never saw pics of Chops big H baffles, but I suspect that the side and top/bottom panels had no bracing, and I can just about guarantee that those panels were vibrating enough to audibly contribute to his output.
Yes, sorry I misspelled John's name. Nearfield measurements of any OB are invalid unless you use them at that nearfield distance. In the nearfield, D is significant in comparison to the measuring distance, and especially with flat baffles (not relevant to John's work, but important to consider) D can be much greater in the nearfield than the farfield.
If D is significant in comparison to measuring distance, then the SPL of the 2 waves can be significantly different, which isn't true in the farfield. eg 2m wide flat baffle, so D is 1m and a 1m measuring distance. The rear wave must travel 2.4m to get to the mic, but the front wave travels only 1m. Since sound decreases by 6db per doubling of distance the rear wave, the front wave will be 7 or 8db louder than the rear wave greatly reducing the impact of the rear wave on the total sound. If you measure the same baffle at 5m, the rear wave will have travelled 6m and the front wave 5m, so the spl difference would be negligble. The bottom line is that in the nearfield OB's have significantly more bass than in the farfield, ignoring room effects.
John's test results are in a thread here on DIY that he started about the subject. He was trying to test whether it's true that U-baffles of the same size as an H baffle have a +6db edge over the H baffles. Apparently SL and JohnK had a big argument over this long ago, and is one of the reasons SL no longer posts here. The funny thing is that SL's formula for Fequal proves it, so why argue? If you take an H baffle and turn it into a U-baffle with the same overall dimensions, D doubles and Fequal is one octave lower or +6db.
Regarding bracing, I'm not talking about driver mounting structures. I'm talking about large unbraced panels. While the sound waves leaving the baffle don't exert much force on a flat baffle, the same is not true for other OB shapes. Once you constrain the waves the forces acting on the panels is significant and very similar to a box. With box speakers, it's not the change in absolute air pressure that makes bracing necessary. It's the sound/pressure waves acting on them that causes vibrations. The difference between boxes and OB is that boxes have reflections to deal with too. Most people believe that since OB's are open the forces on panels are minimal, but this simply isn't true. Plus boxes at least have their side panels reinforced by the rear corner, but you never see bracing at the rear edge of an OB. As a result, they "flap in the wind".
A prime example is Linkwitz's dipole woofer (not a subwoofer btw), which is copied so often http://linkwitzlab.com/woofer.htm . The 2 side panels have no bracing whatsoever and anyone who has built one can tell you that those panels vibrate quite significantly. A couple of braces across the back and a couple of rib braces running from front to back make it a much more solid design.
I never saw pics of Chops big H baffles, but I suspect that the side and top/bottom panels had no bracing, and I can just about guarantee that those panels were vibrating enough to audibly contribute to his output.
John, some interesting points.
I would be more concerned about vibrations transferred to the panels from the driver itself.
Hmmm, I would have thought the 6db difference wouldn't be that debatable, also that SL wouldn't be concerned. I would have thought the real argument is what the different polar responses mean.
I would be more concerned about vibrations transferred to the panels from the driver itself.
Hmmm, I would have thought the 6db difference wouldn't be that debatable, also that SL wouldn't be concerned. I would have thought the real argument is what the different polar responses mean.
paulspencer said:
With a strong baffle of reasonable mass, the Mms (moving mass) is insignificant in comparison, just like with regular speakers. There's nothing wrong with addressing the mounting, but it should be the last thing on the list, just like with a box speaker. Panels vibrations are signficant when a baffle is folded, and just like with a box they should be addressed.
An example is my Joint Chiefs U-baffle. The maximum panel size is 16"Dx12"W. It has 5 sides to avoid parallel surfaces and is completely open at the back. It's constructed of 18mm ply and is quite strong. The panel vibrations were very bad and quite audible in the form of boomy bass. With such small panel sizes and construction, the resonant frequency of the panels had to be outside of the bass region affected, so it wasn't a matter of stimulating a resonance. It was sufficient forces causing very audible vibrations. Once I cross braced from each joint to just off center of the opposing panel the vibrations became inaudible, but it does still vibrate more than I like. I'll probably use 2"x12" or thicker to build my final form pair, and I may still need bracing.
Since 1" of depth in a U-baffle is equal to 4" of width for a flat baffle without considering the floor effect, I would never opt for a flat baffle. There are just a few considerations that you must consider in U-Baffle construction, but to me a room friendly size is important. Plus I feel that big flat baffles have some significant audible detriments, that my small U's don't have. You don't see Linkwitz or John K with big wide baffles. I'm just figuring out how to keep the narrowish front profile without using all the extra electronics between my source and speakers while retaining the open natural sound of OB.
I had started trying to wade through Chop's old thread, looking for a little enlightenment, but this thread all of a sudden became a nice digest for that information (thanks John 🙂 ).
So, to let me get this straight in my head . . . the best bet is to mount your woofer facing forward in an open backed box, right?
Assuming that, I have some other questions. First, is there really any point in using pairs of drivers over single ones? Second, optimal positioning should be close to the floor as possible, right? Third, how much of an effect will there be from removing the top of that open backed box, especially if the front baffle is 3'-4' tall? Fourth, what effect might their be in reducing the sides of the U from full depth at floor level to something less higher up (possibly contouring them for aesthetic appeal, like the side panels on an Orion, for example)?
Okay enough of that . . . John mentioned something about avoiding parallel surfaces in the back of the U. The five sided back on the U defnitely keeps that from happening, but seems as if it might be hard to impliment (my woodworking skills and tools are minimal, at best). What about a tube or cylinder, like a sawed off piece of Sonotube instead?
The more questions I ask, the more I think of . . . are there any arguments against having your woofers mounted facing to the side (I'm still assuming one driver, integrated as part of one structure with the rest of each speaker)? That way the front baffle could be say only 7-8" to accomodate my other driver(s), but the entire thing could be a U to a depth of at least the mounting dimension of the woofer.
Okay enough for tonight . . . thanks for bringing this back, guys.
Kensai
So, to let me get this straight in my head . . . the best bet is to mount your woofer facing forward in an open backed box, right?
Assuming that, I have some other questions. First, is there really any point in using pairs of drivers over single ones? Second, optimal positioning should be close to the floor as possible, right? Third, how much of an effect will there be from removing the top of that open backed box, especially if the front baffle is 3'-4' tall? Fourth, what effect might their be in reducing the sides of the U from full depth at floor level to something less higher up (possibly contouring them for aesthetic appeal, like the side panels on an Orion, for example)?
Okay enough of that . . . John mentioned something about avoiding parallel surfaces in the back of the U. The five sided back on the U defnitely keeps that from happening, but seems as if it might be hard to impliment (my woodworking skills and tools are minimal, at best). What about a tube or cylinder, like a sawed off piece of Sonotube instead?
The more questions I ask, the more I think of . . . are there any arguments against having your woofers mounted facing to the side (I'm still assuming one driver, integrated as part of one structure with the rest of each speaker)? That way the front baffle could be say only 7-8" to accomodate my other driver(s), but the entire thing could be a U to a depth of at least the mounting dimension of the woofer.
Okay enough for tonight . . . thanks for bringing this back, guys.
Kensai
For a single woofer I would say it's the most efficient use of space, although if you had a very narrow width or height (if your woofer is suited to horizontal mounting) to work with those cabs that Moray linked to could work well with added depth, especially if you're looking to lower Fs (but I believe an efficiency sacrifice comes with it).Kensai said:
Of course...more output at the bottom.
This floor is a big benefit, so I'd say yes as long as the very low mounting doesn't put you too far off axis like with nearfield listening. eg A U-baffle woofer at your feet won't do much good.
Mr. Linkwitz told me that the rear wave will search for the shortest path, so look at the overall pathway for the rear wave and visualize the effect. If one direction is much shorter than the others, more will flow that way and cause an earlier rolloff.
I strongly encourage different shapes for OB designs, since without constraints of boxes much more interesting designs are easy. You still have to take into account the answer above to #3.
Keep in mind that my 5 sided short pipe is for a 15" fullrange coax unit, so I had to think about standing waves in the higher frequencies. Sonotube could work very well, and oh so simple. I can't get it here in Costa Rica, or I would have tried it myself. With U-baffles you do have to think about 1/4 wave resonances. JohnK stuffs his U-pipe woofers to eliminate resonance, but it will depend on your XO point and U depth. Also, the deeper you go, the more you change away from dipole to a cardoid response pattern.
That is the approach with Ripoles (search here in the forum for more detail) and stacking the cabs Moray linked us to lengthwise would do this. There are some compromises. It's not a U-baffle, so the D is the depth of those, however, if you keep the driver as close to the front edge as possible, every inch you add in depth picks up 2" of D and the addition is U-baffle. As I said before, I've done a similar cab but with pairs of drivers facing cone to basket using 12" drivers and maintained a 12" width, but 4 drivers fit in what would only fit 2 with front mounting. With pairs, I picked up the benefits of mechanical vibration cancellation, push/pull operation, and lowering Fs. The compromise is depth and D. Mine have a 22" depth, but my D is only 31". Also, I should mention that you give up a lot of floor benefit with tall narrow forms, so while mine would equal a 62" wide flat baffle in free space, resting on the floor it wouldn't. Unfortunately I haven't found a formula to be able to quantify the floor effect, however, I've felt no need to turn mine on it's side to get more bass.
I won't disagree that an OB speaker should be solid to avoid coloration.
In a box you have pressure acting on the enclosure and vibration transferred from the driver, and both of these require the box to be solid. Without a box that builds up pressure, this is much less of an issue but the vibration issue remains.
Think of this: which vibrates more, a flat baffle attached to the driver or that same baffle held in the air near the driver with no mechanical connection. Isn't it fairly obvious?
Kensai,
Using two drivers lowers distortion in two ways:
1. half excursion required, thus reducing distortion that increases with excursion as well as power input
2. push pull mounting where the second driver is mounted with polarity reversed and magnet forward - this reduces distortion in a similar way to a shorting ring, but better because it addresses mechanical non-linearities in the suspension and motor
In a box you have pressure acting on the enclosure and vibration transferred from the driver, and both of these require the box to be solid. Without a box that builds up pressure, this is much less of an issue but the vibration issue remains.
Think of this: which vibrates more, a flat baffle attached to the driver or that same baffle held in the air near the driver with no mechanical connection. Isn't it fairly obvious?
Kensai,
Using two drivers lowers distortion in two ways:
1. half excursion required, thus reducing distortion that increases with excursion as well as power input
2. push pull mounting where the second driver is mounted with polarity reversed and magnet forward - this reduces distortion in a similar way to a shorting ring, but better because it addresses mechanical non-linearities in the suspension and motor
paulspencer said:
Paul, this is probably the most common misconception in all of audio and I'm quite surprised that you would spread it further. The air pressure inside a box speaker changes very little. SPL is sound pressure not air pressure. Think of even a subwoofer with a huge swept volume of 6L in a relatively small 100L box. That's just +/- 3% change. What about vented alignments?
It's the sound pressure waves striking the sides of a box that make them vibrate and these same waves (ignoring reflections inside a box) are trying to expand. With a flat baffle the waves trying to bend around the baffle edges are traveling parallel with the baffle, so vibrations aren't stimulated. Once you add wings which constrain the natural expansion, the forces can be very similar to a box depending upon the shape and dimensions of the baffle. You obviously haven't built many folded baffles, or you wouldn't so readily dismiss what I'm talking about.
No it's not as obvious as you are thinking. Build a relatively small cube out of 1/4' plywood and mount a driver. The driver baffle will likely vibrate less than the other 5 panels. Take the back off, and the top, bottom and side panels will most likely vibrate even more despite the back being open because, there is now no support at all at the trailing edges. If you took the top off instead of the back, that back panel would vibrate wildly because it is constraining the wave expansion even more and forcing it another direction.
OB's are surrounded by myths and misconceptions. The idea that pressures and forces go a way just because a speaker isn't closed is just one of them. Everyone already knows how to build strong, relatively vibration free driver mounting baffles, but what I see missing from every folded OB design is bracing for the other panels. This design flaw is significant and can be quite audible. With box speakers it's quite common to see overkill to prevent vibration, yet with folded OB's you see zero bracing and it's poor design, period.
Great stuff, guys.
Here's some other things that have been bothering me about using pairs of woofers like this. Why do we need to mount them facing opposite directions? Couldn't we get similar effect mounting them both facing forward but wiring them out of phase? First off, I can't believe that the rear wave sounds very good compared to the front wave. Second, its quite ugly having the rear of a driver facing into the room. The only point I could see in an H baffle over a U would be that you would have a frame around with to put a decorative grille to hide rear of the the back facing speaker (though I'm not convinced that the front cavity of the H wouldn't color or distort the front wave more than its worth in holding off the rear wave cancellation effects).
I guess another way to do it would be to put one driver (probably the top one) on the front baffle, and then put the other on a baffle built at the rear of the speaker, leaving the opening of the lower U at the front of the speaker. Is that called an S baffle? Are there issues with not having the two drivers mounted in the same plane like that?
I'm also wondering what effect the ripole sorts of configurations would have on the sound quality. It would seem that they loading both waves of the driver with those small spaces front and back. And the sonotube concept could be similarly loading the rear wave if not handled properly (what would be, in your opinion, a good length for the sonotube? probably not any longer than the driver diameter, eh?). I really want to cut down on size and construction effort, but I don't want to compromise on SQ. For most of my planned applications I'd probably be crossing around 100Hz, though possibly as low as 60-80Hz and even as high as 200-225Hz, so I'm not sure what effect any of these configurations would have at the top of this range.
I'm guessing the best way to get some serious bottom end in as small a footprint as possible is to just stack a bunch of downfacing drivers in a set of ripoles or an extended W frame (would that be a WW frame for 4 drivers ;-p) with front waves venting out the front and the rear waves venting out the back. This would take care of footprint and aesthetics (using 12" drivers, it could be as small as a 14" square if built right).
So much to think about.
Kensai
Here's some other things that have been bothering me about using pairs of woofers like this. Why do we need to mount them facing opposite directions? Couldn't we get similar effect mounting them both facing forward but wiring them out of phase? First off, I can't believe that the rear wave sounds very good compared to the front wave. Second, its quite ugly having the rear of a driver facing into the room. The only point I could see in an H baffle over a U would be that you would have a frame around with to put a decorative grille to hide rear of the the back facing speaker (though I'm not convinced that the front cavity of the H wouldn't color or distort the front wave more than its worth in holding off the rear wave cancellation effects).
I guess another way to do it would be to put one driver (probably the top one) on the front baffle, and then put the other on a baffle built at the rear of the speaker, leaving the opening of the lower U at the front of the speaker. Is that called an S baffle? Are there issues with not having the two drivers mounted in the same plane like that?
I'm also wondering what effect the ripole sorts of configurations would have on the sound quality. It would seem that they loading both waves of the driver with those small spaces front and back. And the sonotube concept could be similarly loading the rear wave if not handled properly (what would be, in your opinion, a good length for the sonotube? probably not any longer than the driver diameter, eh?). I really want to cut down on size and construction effort, but I don't want to compromise on SQ. For most of my planned applications I'd probably be crossing around 100Hz, though possibly as low as 60-80Hz and even as high as 200-225Hz, so I'm not sure what effect any of these configurations would have at the top of this range.
I'm guessing the best way to get some serious bottom end in as small a footprint as possible is to just stack a bunch of downfacing drivers in a set of ripoles or an extended W frame (would that be a WW frame for 4 drivers ;-p) with front waves venting out the front and the rear waves venting out the back. This would take care of footprint and aesthetics (using 12" drivers, it could be as small as a 14" square if built right).
So much to think about.
Kensai
Kensai,
First, I strongly suggest you gain an understanding dipole bass cancellation. The point where it starts depends upon the extra distance you make the rear wave travel to get to your ears. It the phase relationship of the front and rear waves when they reach your ears that matters. At the sides of a dipole, they net to zero because they have travelled the same distance in the same direction and retain their directly out of phase relationship. The 2 waves don't meet at the sides and cancel each other out. They travel right through each other.
You'll be able to answer your own questions if you take the time to draw what you are talking about and measure the difference in distance that the rear wave must travel to get to the listening position. Hint, to the extent you make it travel rearward first, then it must travel forward that distance too before the remaining distance is equal. This results in the most compact speaker size for a given bass roll-off point.
In the case of an H baffle, the wood in front of the driver baffle has zero effect on bass cutoff. All it does is make the front and back radiation equal, so it remains a pure dipole. You can cut off everything in front of the driver baffle and you'll get the same on axis bass response at the listening position.
First, I strongly suggest you gain an understanding dipole bass cancellation. The point where it starts depends upon the extra distance you make the rear wave travel to get to your ears. It the phase relationship of the front and rear waves when they reach your ears that matters. At the sides of a dipole, they net to zero because they have travelled the same distance in the same direction and retain their directly out of phase relationship. The 2 waves don't meet at the sides and cancel each other out. They travel right through each other.
You'll be able to answer your own questions if you take the time to draw what you are talking about and measure the difference in distance that the rear wave must travel to get to the listening position. Hint, to the extent you make it travel rearward first, then it must travel forward that distance too before the remaining distance is equal. This results in the most compact speaker size for a given bass roll-off point.
In the case of an H baffle, the wood in front of the driver baffle has zero effect on bass cutoff. All it does is make the front and back radiation equal, so it remains a pure dipole. You can cut off everything in front of the driver baffle and you'll get the same on axis bass response at the listening position.
Kensai said:
This is what a member on the Klipsch forum (DrWho) wrote me about this...
It is considered a push/pull or compound configuration. Distortions that are caused from a cone flexing while going in one direction is cancelled out by the distotions caused by the other cone moving in the opposite direction, thus cancelling out the distotions only.
It also has to do with one driver's VC leaving the magnetic gap as the other (inverted) driver's VC is entering the magnetic gap at the same time.
Chops, I don't know about the cone flexing issue. I have seen other factors such as asymmetrical BL and Kms curve and flux modulation mentioned, but not this. I guess it would probably be true as well. I have updated a wiki article I wrote on push-pull.
http://www.diyaudio.com/wiki/index.php?page=Push+pull+driver+mounting
http://www.diyaudio.com/wiki/index.php?page=Push+pull+driver+mounting
Hmm... Maybe that pertains more towards enclosed drivers and not OB. I would think it would still hold true though, no matter what the enclosure type, or the lack thereof. 😉
Re: Chops which ones?
http://www.partsexpress.com/pe/showdetl.cfm?&DID=7&Partnumber=292-218
moray james said:
http://www.partsexpress.com/pe/showdetl.cfm?&DID=7&Partnumber=292-218
Back in December when starting in the current OB speaker project I made a bunch of measurements of the Dayton IB385-8 15" drivers. One of the measurements was comparing the distortion of 2 drivers facing forward and then the same 2 drivers in PP. The drivers were mounted in separate 16x16x16 u-baffles without damping material on the back side.
The microphone position was 12" out from the front facing driver and centered between the 2 drivers or the 1 driver and the rear of the u-baffle. The path distance was not corrected for the 16" offset in the PP configuration and could explain the increase in 2nd harmonic above 80Hz (or not...). The rising response in the PP configuration is from the increasing output of the un-damped u-baffle, which may be another explanation for the increase of the 2nd harmonic at 80Hz (or not...).
The following images show the decrease in second harmonic distortion nicely.
First image is with both drivers facing forward
Forward facing drivers
Second image is with the drivers in PP configuration
PP drivers
Another note that is informative and amusing at the same time...
While playing with these drivers it was quite an experience to see just how much vibration these drivers generate at 20Hz and high excursion. The u-baffles were constructed from 3/4 MDF and or 3/4 baltic birch plywood. At the higher output levels the boxes were "drifting" around on the concrete floor. Grabbing the box and trying to stop the vibration would result in a great vibro-massage but not reduce the box vibration much.
Standing with 1 foot on each of the boxes noticeably reduced the 3rd HD measurement. Unfortunately, I did not record the measurement "when standing on the boxes".
Another example of the vibration generated came when I had 1 driver mounted in my infinite baffle test jig. The IB test jig is mounted through a doorway in my basement. The wall is standard 2x4 interior construction. At maximum output at 20Hz the speaker cone was moving over 1" peak to peak and the wall was moving almost 1/4" peak to peak in the opposite direction...
These experiences are part of the reason for going with the W frame. The vibration cancellation of the W frame is very good. The box vibration is based on how much air you are moving instead of the mass of the cones+air.
The microphone position was 12" out from the front facing driver and centered between the 2 drivers or the 1 driver and the rear of the u-baffle. The path distance was not corrected for the 16" offset in the PP configuration and could explain the increase in 2nd harmonic above 80Hz (or not...). The rising response in the PP configuration is from the increasing output of the un-damped u-baffle, which may be another explanation for the increase of the 2nd harmonic at 80Hz (or not...).
The following images show the decrease in second harmonic distortion nicely.
First image is with both drivers facing forward
Forward facing drivers
Second image is with the drivers in PP configuration
PP drivers
Another note that is informative and amusing at the same time...
While playing with these drivers it was quite an experience to see just how much vibration these drivers generate at 20Hz and high excursion. The u-baffles were constructed from 3/4 MDF and or 3/4 baltic birch plywood. At the higher output levels the boxes were "drifting" around on the concrete floor. Grabbing the box and trying to stop the vibration would result in a great vibro-massage but not reduce the box vibration much.
Standing with 1 foot on each of the boxes noticeably reduced the 3rd HD measurement. Unfortunately, I did not record the measurement "when standing on the boxes".
Another example of the vibration generated came when I had 1 driver mounted in my infinite baffle test jig. The IB test jig is mounted through a doorway in my basement. The wall is standard 2x4 interior construction. At maximum output at 20Hz the speaker cone was moving over 1" peak to peak and the wall was moving almost 1/4" peak to peak in the opposite direction...
These experiences are part of the reason for going with the W frame. The vibration cancellation of the W frame is very good. The box vibration is based on how much air you are moving instead of the mass of the cones+air.
Gary,
Interesting stuff.
So if you build a W frame, and you have both drivers mounted the same direction, they should be wired out of phase. That seems to give you the front wave from one driver and the back wave from the other from one side of the W at a time. Will these two waves be more constructive than destructive, then? I understand that because of the mechanical difference between the outward and inward strokes that the waves generated by each will be somewhat dissimilar, but when we're talking about front baffle mounting drivers, we're worried about the backwave interferring with the front wave. Guess I'm just a little confused.
I'm taking on faith for the moment that it works (plenty of anecdotal evidence around, so that's not too hard for me 😉 ). What if I were to build Ws as the lower part of a fullrange unit, and I set them on their sides (making the drivers horizontal instead of vertical). I'm assuming that it would be best for the drivers to be facing downward, right?
Kensai
Interesting stuff.
So if you build a W frame, and you have both drivers mounted the same direction, they should be wired out of phase. That seems to give you the front wave from one driver and the back wave from the other from one side of the W at a time. Will these two waves be more constructive than destructive, then? I understand that because of the mechanical difference between the outward and inward strokes that the waves generated by each will be somewhat dissimilar, but when we're talking about front baffle mounting drivers, we're worried about the backwave interferring with the front wave. Guess I'm just a little confused.
I'm taking on faith for the moment that it works (plenty of anecdotal evidence around, so that's not too hard for me 😉 ). What if I were to build Ws as the lower part of a fullrange unit, and I set them on their sides (making the drivers horizontal instead of vertical). I'm assuming that it would be best for the drivers to be facing downward, right?
Kensai
Gary,
Nice work on those measurements and I'm sure the distortion reduction of push/pull would show even greater in a uniform setup, but your measurements already show it is significant. I've only done P/P with some cheapie 12's that I have a bunch of, and I swear the difference is audible with them. I speculate that P/P alignment becomes more important with cheaper drivers, since I'm sure their operation is less linear.
BTW, it sounds like you did a lot of measuring before starting your OB project. What type of design are you using?
Nice work on those measurements and I'm sure the distortion reduction of push/pull would show even greater in a uniform setup, but your measurements already show it is significant. I've only done P/P with some cheapie 12's that I have a bunch of, and I swear the difference is audible with them. I speculate that P/P alignment becomes more important with cheaper drivers, since I'm sure their operation is less linear.
BTW, it sounds like you did a lot of measuring before starting your OB project. What type of design are you using?
Kensai,
That's the 2nd time you've brought up horizontal mounting, and I'm wondering why, since you do give up some Xmax. To answer your question, I would face them down. That way you also have gravity helping prevent banging the back plate.
If you're interest in horizontal for space savings, I'd like to note that each pair of my 12's results in essentially the same front profile whether they're horizontal or vertical. What are you trying to accomplish?
That's the 2nd time you've brought up horizontal mounting, and I'm wondering why, since you do give up some Xmax. To answer your question, I would face them down. That way you also have gravity helping prevent banging the back plate.
If you're interest in horizontal for space savings, I'd like to note that each pair of my 12's results in essentially the same front profile whether they're horizontal or vertical. What are you trying to accomplish?
- Status
- Not open for further replies.