Not so. Electrically wired In-phase diaphragms move in opposite directions on the closed box. On the driver axis the sound pressure is cancelled creating a null.
There is a small misalignment to perfect on-axis acoustical anti phase due to the finite distance between the drivers but at bass frequencies this is insignificant such as to make for near perfect on axis cancellation.
At 90 degrees off-axis (directly forward or rearward to side mounted woofers), the two electrically in-phase drivers are acoustically in-phase.
There is a small misalignment to perfect on-axis acoustical anti phase due to the finite distance between the drivers but at bass frequencies this is insignificant such as to make for near perfect on axis cancellation.
At 90 degrees off-axis (directly forward or rearward to side mounted woofers), the two electrically in-phase drivers are acoustically in-phase.
Here’s some data and graphics that may be helpful.
Dual opposed 8” woofers x2, on baffle 9.5” wide-
https://www.diyaudio.com/community/...elp-tpl-200hh-td10m-td15h.399746/post-7397098
Dual opposed 8” woofers x2, on baffle 9.5” wide-
https://www.diyaudio.com/community/...elp-tpl-200hh-td10m-td15h.399746/post-7397098
Thanks. Yes, opposing woofers would certainly be rather nice. But at the modest SPL expected here they wouldn't seem likely to gain quite as much as a large higher SPL speaker would, unless I needed it for directivity. It would also mean buying more drivers, in my case. So at least to start with I'll try just the one per side and see if there is any issue.
In the meantime I've done a couple of sketches for the kind of internal dimensions I would be looking at (assuming no EQ), and also how this might work with a deep tapering mid-range chamber - which isn't massively important but is something I would like to try. Nothing very accurate or specific, in fact the mid-range driver hasn't even been chosen yet (shown is for an alpair 7ms), it is just to get a feel for roughly what might be done:
Of course, just showing side views of the speakers. The traditional front-mounted woofer would make a more traditional ratio of height to depth (right hand image). However for desktop use I think the smaller height and deeper cabinet could be preferable, and is a bit easier to construct (the left hand images, either way up). It also isn't impossible that (with the woofer on the side) space could be left for a small midbass driver, should that be wished for in the future.
But there isn't a night-and-day difference in height between them, so it probably depends most on what clutter there is (or might be) to the sides of the speakers.
In the meantime I've done a couple of sketches for the kind of internal dimensions I would be looking at (assuming no EQ), and also how this might work with a deep tapering mid-range chamber - which isn't massively important but is something I would like to try. Nothing very accurate or specific, in fact the mid-range driver hasn't even been chosen yet (shown is for an alpair 7ms), it is just to get a feel for roughly what might be done:
Of course, just showing side views of the speakers. The traditional front-mounted woofer would make a more traditional ratio of height to depth (right hand image). However for desktop use I think the smaller height and deeper cabinet could be preferable, and is a bit easier to construct (the left hand images, either way up). It also isn't impossible that (with the woofer on the side) space could be left for a small midbass driver, should that be wished for in the future.
But there isn't a night-and-day difference in height between them, so it probably depends most on what clutter there is (or might be) to the sides of the speakers.
Last edited:
I still don't get it. Are you saying sound pressure or driver force impulse are cancelling out?
For low frequencies and driver distance small compared to wavelength the sound pressures from both (in phase) drives should add up, irrespective of driver orientation - as far as I understand.
Here’s the data for one per side.
7” midwoofer on 8” wide cabinet:
https://www.diyaudio.com/community/...elp-tpl-200hh-td10m-td15h.399746/post-7396863
7” midwoofer on 8” wide cabinet:
https://www.diyaudio.com/community/...elp-tpl-200hh-td10m-td15h.399746/post-7396863
Thank you! How curious; I suppose the side baffle is wider than the front baffle; if they were the same presumably results from 90-degrees on the front mounted woofer should be identical to those at 0degrees on the side mounted one?
Your conclusions on crossing at 150hz (maybe up to 200hz) seem to tally with what I've read elsewhere around the forum. Or at least one approach to the decision (as mentioned in the OP I also found others).
That could of course be another factor in my choice; if I decide the woofer can go somewhat higher without audible detriment then front-mounting may offer better flexibility/headroom for crossover point and slope. Between things like this and desk clutter, the front facing option seems somewhat more fool-proof, at least. Which may be important in my case 🙂
EDIT: especially if I'm not going to take advantage of dual opposing configurations
Your conclusions on crossing at 150hz (maybe up to 200hz) seem to tally with what I've read elsewhere around the forum. Or at least one approach to the decision (as mentioned in the OP I also found others).
That could of course be another factor in my choice; if I decide the woofer can go somewhat higher without audible detriment then front-mounting may offer better flexibility/headroom for crossover point and slope. Between things like this and desk clutter, the front facing option seems somewhat more fool-proof, at least. Which may be important in my case 🙂
EDIT: especially if I'm not going to take advantage of dual opposing configurations
Last edited:
@ stv.
Sound pressure. Consider an idealisation of a narrow box side-to-side which has woofers back-to-back and the diaphragm rear wave is perfectly contained inside the sealed box. Then when the electrically in-phase drivers move outwards simultaneously, each produces a pressure wave moving in the same direction as the diaphragm (outwards). Because of the speed of sound and the small separation of the drivers, the pressure waves are experienced almost instantaneously on the other side of the box but moving in opposition to the pressure wave from that side. Hence a cancellation on axis.
Same reasoning for inwards moving diaphragms (rarefaction wave).
This is an idealisation and in practice cancellation will be imperfect.
Sound pressure. Consider an idealisation of a narrow box side-to-side which has woofers back-to-back and the diaphragm rear wave is perfectly contained inside the sealed box. Then when the electrically in-phase drivers move outwards simultaneously, each produces a pressure wave moving in the same direction as the diaphragm (outwards). Because of the speed of sound and the small separation of the drivers, the pressure waves are experienced almost instantaneously on the other side of the box but moving in opposition to the pressure wave from that side. Hence a cancellation on axis.
Same reasoning for inwards moving diaphragms (rarefaction wave).
This is an idealisation and in practice cancellation will be imperfect.
That is not correct, at least for low frequencies. The pressure wave expands in all directions, even at the back of the box (for low frequencies). There will not be any cancellation.
Cancellation will start to occure once the distance between drivers is getting similar to the sound wavelength (due to time delay).
Last edited:
I've had a rare day off today, and so have been able to do some testing. I can't discern any great audible difference between front and side mounting at less than 200hz, which is what people earlier in the thread suggested was the woofer's upper limit, so that is great. In isolation then, that would be my preference; and of course could be improved upon (or taken advantage of) with dual opposing woofers.
Except... in practice when the test boxes are put on the desk with bookshelves next to it, there becomes a clear negative effect from facing the woofers sidewards at the bookshelves. Boominess is very apparent above all. I presume the immediate surfaces are causing havoc with reflections and dispersion. Conversely, when the driver is forward facing all remains well. Not sure if the bookshelves might even be acting as an extension of the baffle (the PC monitor certainly will be, as it is so close). There is a gap of several inches between them and the speaker, so they may not, but at least I don't hear any diffraction or early reflection problems from the gap (at <200hz, anyway).
So I could re-arrange the room, but am more inclined to consider forward-facing woofers as a safer bet for working in a variety of small cluttered spaces that might arise in the speakers lifetime. Which is not what I wanted to find but hey-ho; taller speakers aren't really a great problem for desk space, I just need to get the vertical listening center somewhere around ear height, irrespective of the total cabinet height.
Except... in practice when the test boxes are put on the desk with bookshelves next to it, there becomes a clear negative effect from facing the woofers sidewards at the bookshelves. Boominess is very apparent above all. I presume the immediate surfaces are causing havoc with reflections and dispersion. Conversely, when the driver is forward facing all remains well. Not sure if the bookshelves might even be acting as an extension of the baffle (the PC monitor certainly will be, as it is so close). There is a gap of several inches between them and the speaker, so they may not, but at least I don't hear any diffraction or early reflection problems from the gap (at <200hz, anyway).
So I could re-arrange the room, but am more inclined to consider forward-facing woofers as a safer bet for working in a variety of small cluttered spaces that might arise in the speakers lifetime. Which is not what I wanted to find but hey-ho; taller speakers aren't really a great problem for desk space, I just need to get the vertical listening center somewhere around ear height, irrespective of the total cabinet height.
As a point of interest, perhaps not to Kev06 because bookshelves, but to other perusers, the KEF Blade Meta seems to do quite well with side-mounted woofers crossing at 400 Hz: https://www.stereophile.com/content/kef-blade-two-meta-loudspeaker-measurements
That is a design constraint not related to side firing woofers.
KEF’s pure midrange coaxial cannot reach down to ~120-150Hz at significant SPLs, which is perhaps why the woofers have been tasked to play up to 400 Hz.
KEF’s pure midrange coaxial cannot reach down to ~120-150Hz at significant SPLs, which is perhaps why the woofers have been tasked to play up to 400 Hz.
Last edited:
I can certainly see the temptation to cross higher. Wanting (in my case) one driver to span the whole main range makes for quite a trade-off in cone size between low and high end performance; raising the crossover frequency (somewhat) at the low end is not too objectionable and allows for more dynamics down there. Though I'm sure I read a fairly informed article or post somewhere, which felt the KEF Blade's higher crossover went too far.
I'm less likely to be tempted, since it seems my woofers are best below 200hz, anyway. My modest SPL requirements also make the midrange's job less demanding; I'm currently testing some drivers with ~3" diameter cones, so will soon know. But I haven't entirely ruled out a small front mounted midwoofer instead, and using the current big woofers as subs.
I'm less likely to be tempted, since it seems my woofers are best below 200hz, anyway. My modest SPL requirements also make the midrange's job less demanding; I'm currently testing some drivers with ~3" diameter cones, so will soon know. But I haven't entirely ruled out a small front mounted midwoofer instead, and using the current big woofers as subs.
From a directivity point of view;
A side mounted 12" woofer crossed at 150Hz to a 6.5" woofer crossed at 4" at 500Hz, crossed to a tweeter a 2.8KHz (ie. 4 way OR 3 way + subwoofer; depending on how you look at it), like the old Infinity IL60 will work.
I wonder where the 80Hz idea came from; I mean, the THX or Dolby standard for LFE channels is 120Hz.
PS.
You can check low frequency sound localization for yourself here-
https://www.audiocheck.net/audiotests_basslocalization.php
A side mounted 12" woofer crossed at 150Hz to a 6.5" woofer crossed at 4" at 500Hz, crossed to a tweeter a 2.8KHz (ie. 4 way OR 3 way + subwoofer; depending on how you look at it), like the old Infinity IL60 will work.
I wonder where the 80Hz idea came from; I mean, the THX or Dolby standard for LFE channels is 120Hz.
PS.
You can check low frequency sound localization for yourself here-
https://www.audiocheck.net/audiotests_basslocalization.php
Thanks, that is again useful. It is often mentioned that below 80hz sound isn't localisable, though as you say up to 120hz is also commonly claimed as acceptable. Which is usually applied to using separately located subwoofers, but I found some people on here applying the same frequency threshold to side-mounted woofers. Which I think your directivity plots (and the above link) suggest is quite pessimistic for that application. So I'm now quite happy that below 150hz to maybe 200hz would be fine for me.
However that now makes two things from this thread that limit me to 200hz or less. The first is the directivity if side mounted, the second is people's opinion on my woofer's upper usable limit (regardless of where it is mounted). This low crossover point puts more compromise on the driver handling the extended mid-ranges, and likely on crossover slopes.
I think it can be achieved with a wide-range driver of 3" cone (say 4" nominal frame), without going to massive enclosures etc; I'm testing some currently. Though I might have preferred a smaller one crossed one or two hundred hertz higher, for such a nearfield application. So I might be tempted to instead get a pair of 8" bass drivers rather than the current sub-bass ones (and possibly EQ them). The lower distortion and higher breakup might be no bad thing, either. On reflection, it might be more sensible than my thought of adding another driver, to bridge a small gap; my SPL requirements probably don't warrant it realy.
However that now makes two things from this thread that limit me to 200hz or less. The first is the directivity if side mounted, the second is people's opinion on my woofer's upper usable limit (regardless of where it is mounted). This low crossover point puts more compromise on the driver handling the extended mid-ranges, and likely on crossover slopes.
I think it can be achieved with a wide-range driver of 3" cone (say 4" nominal frame), without going to massive enclosures etc; I'm testing some currently. Though I might have preferred a smaller one crossed one or two hundred hertz higher, for such a nearfield application. So I might be tempted to instead get a pair of 8" bass drivers rather than the current sub-bass ones (and possibly EQ them). The lower distortion and higher breakup might be no bad thing, either. On reflection, it might be more sensible than my thought of adding another driver, to bridge a small gap; my SPL requirements probably don't warrant it realy.
Last edited:
Lest I forget, another small note to self on practicalities: I've been testing drivers that work better on axis, and so prefer to be toed-in. If the cabinet is very deep, this means the back outside corners would be hanging way over the desk (if indeed there is space at the sides to do so).
I suppose it would be possible to design boxes with an angled front baffle, but that adds complication and might look a bit weird (especially if used in different future situations).
This doesn't preclude side-mounted woofers, but if shorter and therefore taller cabinets are to be made anyway then there is perhaps less to gain by having the woofers on the sides instead of below/above the other front driver(s).
I suppose it would be possible to design boxes with an angled front baffle, but that adds complication and might look a bit weird (especially if used in different future situations).
This doesn't preclude side-mounted woofers, but if shorter and therefore taller cabinets are to be made anyway then there is perhaps less to gain by having the woofers on the sides instead of below/above the other front driver(s).
My latest thoughts and tests have started to show some promise. If the woofer cabinet is made separate to the mid+high, then there is (of course) much more flexibility on positioning. Perhaps the most interesting arrangement is due to there being a 5" gap between the bottom of the mid+high cabinets (at their ideal height) and the desk. That isn't high enough to get the woofers in physically, however it offers enough space for them to fire through without causing audible artifacts.
So, if the woofers were set back behind the mid+high cabinet then this could work quite well (though perhaps a delay on the latter might be needed). Not completely ideal, but I'm warming to it because the widerange mid driver becomes harder to localise when it is allowed a nice baffle, uninterrupted by a woofer, plus the smaller mid+high cabinet feels noticeably less intrusive when close up (than a box big enough for all the drivers together).
This might appear to be headed towards a sub+sat arrangement. However my intention is still to relieve the smaller mid driver from heavy excursion by crossing around 150hz-200hz, and for the woofers to remain both stereo and within 1/4 wavelength distance. So IMO it is still being designed more or less as a multi-way, albeit with separate boxes and a fairly low crossover. Though it may become a WAW/FAST equivalent, if I decide that a tweeter doesn't gain enough in this case; I've been trailing an alpair 7ms and (at my hearing age) the top end might be considered good enough without tweeter assistance.
So, if the woofers were set back behind the mid+high cabinet then this could work quite well (though perhaps a delay on the latter might be needed). Not completely ideal, but I'm warming to it because the widerange mid driver becomes harder to localise when it is allowed a nice baffle, uninterrupted by a woofer, plus the smaller mid+high cabinet feels noticeably less intrusive when close up (than a box big enough for all the drivers together).
This might appear to be headed towards a sub+sat arrangement. However my intention is still to relieve the smaller mid driver from heavy excursion by crossing around 150hz-200hz, and for the woofers to remain both stereo and within 1/4 wavelength distance. So IMO it is still being designed more or less as a multi-way, albeit with separate boxes and a fairly low crossover. Though it may become a WAW/FAST equivalent, if I decide that a tweeter doesn't gain enough in this case; I've been trailing an alpair 7ms and (at my hearing age) the top end might be considered good enough without tweeter assistance.
@Kev06
It is spooky how your thinking is converging towards my approach. I scaled up my original single box near-field monitors to a free standing 2-box 4-way loudspeaker, with the side-firing (sub) woofers in a separate sealed enclosure. In one of the images, the baby brother is seen in the background. The opposed reaction cancelling 8" subs are mounted low to take advantage of room boundary re-inforcement. Sealed box alignment for the subs reduce group delay and bass phase anomalies. I aimed for 20Hz in-room and achieved this. The sub frequency range is 20Hz-80 Hz with LR48 crossover alignment at 80Hz with the aim to keep the low-mid (80Hz-320Hz) driver excursion and distortion low. My approach requires a dsp active crossover. I don't know how compact separate enclosures can be made for desktop duty.
It is spooky how your thinking is converging towards my approach. I scaled up my original single box near-field monitors to a free standing 2-box 4-way loudspeaker, with the side-firing (sub) woofers in a separate sealed enclosure. In one of the images, the baby brother is seen in the background. The opposed reaction cancelling 8" subs are mounted low to take advantage of room boundary re-inforcement. Sealed box alignment for the subs reduce group delay and bass phase anomalies. I aimed for 20Hz in-room and achieved this. The sub frequency range is 20Hz-80 Hz with LR48 crossover alignment at 80Hz with the aim to keep the low-mid (80Hz-320Hz) driver excursion and distortion low. My approach requires a dsp active crossover. I don't know how compact separate enclosures can be made for desktop duty.
Attachments
Last edited:
Oops. I forgot I have previously posted similar images. It was not my aim to hog attention.
Apologies.
I have included a bit more info about the configuration though.
Apologies.
I have included a bit more info about the configuration though.
Thanks Bon. No worries about duplication; I'm taking so long that people forget - including me, so it is a useful reminder! I get little bits of time to think and question, yet enough time for making and testing (progressing) is frustratingly rare these days.
In retrospect I might have been better off getting smaller but dual woofers so that they could oppose each other. Though it is harder to get smaller drivers that go so low (in a small sealed box) and these were something of a bargain. Perhaps at the modest desktop SPLs needed here, single/unopposed need not be awful. But at least their small sub-like nature is turning out to be very suitable, given the low crossover that I'm aiming for.
Yes, fitting things on a desk is an added challenge, at least to do very optimally with clutter on all sides. So I'd been coming up with some weird and probably over-complicated cabinet shapes. But seeing your neat results again, reminds me how much an appealing design matters. Especially for something one lives with every day. So I'm increasingly tempted to avoid awkward shapes and instead try to rearrange the room.
In retrospect I might have been better off getting smaller but dual woofers so that they could oppose each other. Though it is harder to get smaller drivers that go so low (in a small sealed box) and these were something of a bargain. Perhaps at the modest desktop SPLs needed here, single/unopposed need not be awful. But at least their small sub-like nature is turning out to be very suitable, given the low crossover that I'm aiming for.
Yes, fitting things on a desk is an added challenge, at least to do very optimally with clutter on all sides. So I'd been coming up with some weird and probably over-complicated cabinet shapes. But seeing your neat results again, reminds me how much an appealing design matters. Especially for something one lives with every day. So I'm increasingly tempted to avoid awkward shapes and instead try to rearrange the room.