I was hoping to back up my thoughts on the XLS modelling discrepancy but after taking a quick look on BK's webpage and specifically the XLS200 page.
They're still selling that model but they've removed the measurements which were up when I bought mine, thats a shame as the results tallied well with measurements I took. Now they simply specify an F6 of 17hz which is about right considering I had an F3 of 20hz. This is a single XLS10 in an 18ltr enclosure *without* EQ.
Model this and the XLS10 looks like a terrible performer sealed when in reality that's not the case.
They're still selling that model but they've removed the measurements which were up when I bought mine, thats a shame as the results tallied well with measurements I took. Now they simply specify an F6 of 17hz which is about right considering I had an F3 of 20hz. This is a single XLS10 in an 18ltr enclosure *without* EQ.
Model this and the XLS10 looks like a terrible performer sealed when in reality that's not the case.
Member
Joined 2003
Assuming you are still at the same volume, I'd build the test mule vented...in a way that it can be easily sealed. That opens up playing with straws etc. I'm betting on sealed in the end, but I'd start with a vented box.
Is this a generally accepted truth or proven fact? Does higher excursion = higher distortion?
Also, why so big for the sealed version of this sub? It seems like you could do about as well with 80l as with 130l and it certainly would make the modules much more managable.
Just thoughts...
m0tion said:
I think it holds true in all cases where your dealing with traditional motors. Motor and VC design can lessen the effects but the reality is magnet force becomes less and less linear as the VC moves to the extremities of the field. Underhung helps greatly here as it at least remains fully within the field through the stroke, overhung is worst as the VC partially or even fully moves out of the gap, although new VC technology such as TC Sounds LMS coils achieve performance similar to underhung but with much greater excursion, the price paid is sensitivity and force because the VC gap clearance has to be much greater to accomodate this technology which is essentially a concave VC.
Of course there are other factors linked to distortion such as cone edge and broadband resonances, non linear suspensions, thermal issues and others. The underlying issue here is that as the driver moves more and more into the extremes of its operating range the less linear its behaviour and the result is distortion.
Primarily because I had the idea of simply trying out the two alignments using the same cabinet. You're right though, 130ltrs for sealed isn't needed.
Shin,
actually I think the 1x B&C holds up very well in the model, since you are comparing it to 4x XLS. At least you should add 6 dB to the B&C curves for a stereo output figure.
Also, SL in his Thor pages (12" XLS sealed and EQ'd) cautions that in a selaed box the XLS's useable Xmax is only half the reported technical Xmax before distortion becomes unacceptable. It may be that a driver optimized for a vented box will have a deliberately nonlinear suspension to "pull back" the cone from high excursions, which in turn will lead to distortion when using such drivers in a selaed box (where you have an air spring effect). See Dickason book.
In any case not only the models may be giving wrong ideas, the reported Xmaxes are also by no means a standard figure. B&C reports Xmax according to some AES standard, which specifies Xmax as the excursion where the driver reaches 10% THD , sounds awful but for a bass driver it's not so bad. They also specify another excursion parameter which they think is more realistic, this is Xvar where the driver is off spec by a certain percentage. Then, Adire for example considers Xmax as 70% of specc'd magnetic force left. And finally some manufacturers, such as 18Sounds, give Xmax als the purely geometrical value where the coil starts leaving the gap.
Bottom lines: beware models and numbers. And: the B&C may be a very good choice and design after all, and distort less than several XLS's . But full satisfaction will only occur if you do build a second one. One bass driver alone won't give you the feeling of bass envelopment, see Griesinger paper i referenced many posts above - for that you need two - and this may be what you've been missing with your single B&C.
Not that I've heard any of your setups, but just my $0.02.
actually I think the 1x B&C holds up very well in the model, since you are comparing it to 4x XLS. At least you should add 6 dB to the B&C curves for a stereo output figure.
Also, SL in his Thor pages (12" XLS sealed and EQ'd) cautions that in a selaed box the XLS's useable Xmax is only half the reported technical Xmax before distortion becomes unacceptable. It may be that a driver optimized for a vented box will have a deliberately nonlinear suspension to "pull back" the cone from high excursions, which in turn will lead to distortion when using such drivers in a selaed box (where you have an air spring effect). See Dickason book.
In any case not only the models may be giving wrong ideas, the reported Xmaxes are also by no means a standard figure. B&C reports Xmax according to some AES standard, which specifies Xmax as the excursion where the driver reaches 10% THD , sounds awful but for a bass driver it's not so bad. They also specify another excursion parameter which they think is more realistic, this is Xvar where the driver is off spec by a certain percentage. Then, Adire for example considers Xmax as 70% of specc'd magnetic force left. And finally some manufacturers, such as 18Sounds, give Xmax als the purely geometrical value where the coil starts leaving the gap.
Bottom lines: beware models and numbers. And: the B&C may be a very good choice and design after all, and distort less than several XLS's . But full satisfaction will only occur if you do build a second one. One bass driver alone won't give you the feeling of bass envelopment, see Griesinger paper i referenced many posts above - for that you need two - and this may be what you've been missing with your single B&C.
Not that I've heard any of your setups, but just my $0.02.
Adding the second B&C will only add 3dB SPL. This is not the big improvement needed IMO to give the extra he is looking for.
I see where you are coming from, but the reality is that he has tested the B&C and found it not to give the SPL claimed by the model. Also he has in the past tested a single XLS and been surprised at the large amount of output. These two factors combine to basically show that the models are not all that accurate/relevant so it really doesn't matter how the manufacturer quotes Xmax.
I see where you are coming from, but the reality is that he has tested the B&C and found it not to give the SPL claimed by the model. Also he has in the past tested a single XLS and been surprised at the large amount of output. These two factors combine to basically show that the models are not all that accurate/relevant so it really doesn't matter how the manufacturer quotes Xmax.
Hmm. A second B&C should add 6 dB SPL. I just suspect that the improvement will not be as large - the XLS may be equally mis-modeled fo ronce, and if both are mismodeled by the same amount then 2 B&C's could quite well do the trick. Also, if Linkwitz is correct and the XLS's distort heavily for half their Xmax then the perceived higher output may have come from abundant harmonics...
But anyway. since perfection can't be achieved, fair enough to choose the subjectively most pleasing solution.
But anyway. since perfection can't be achieved, fair enough to choose the subjectively most pleasing solution.
Adding another driver, or doubling input power only gives you +3dB SPL.
Again I see what your intentions are with the models etc, but as you should see from my last couple of posts and Shin's replies, it's now totally irrelevant what the SPL simulations and supposed distortion problems say, as he has effectively tried both in his room and found the conclusion that the B&C is lacking and the XLS has the potential.
Again I see what your intentions are with the models etc, but as you should see from my last couple of posts and Shin's replies, it's now totally irrelevant what the SPL simulations and supposed distortion problems say, as he has effectively tried both in his room and found the conclusion that the B&C is lacking and the XLS has the potential.
MBK said:
Oh dear, thanks for bringing that to my attention MBK.
It looks like the XLS is a bit of waste in a sealed box. I can only imagine that adding an LT will simply make matters worse here.
I've only had experience with the XLS10 in an 18ltr sealed enclosure and whilst the extension was greater than modelled prediction suggest its output was limited, it was an easy sub to drive but fell apart before even getting close to reference level. So based on that I deemed it overall a good performer if kept at sensible levels.
Looking at the TS its clear this is a driver that will obviously prefer vented and PR setups but I had no idea sealed harmed performance in anything other than the obvious side effects of that alignment. That makes me rethink my preference for going for sealed with LT now, I'm still trying it but I believe the best will come from vented, unless I've got some unknown desire for high THD.
Well I've got a clear place to start now: vented. I'm also going to do as Rob and MBK suggested and split the four drivers between two subs. This way I the usual option of either coupling output or combatting room problems. I'm figuring even if they are at completely opposite ends of the room they'll still couple output at low frequencies because of the 1/2 wavelength distance.
Vented. for music, is actually a preference of mine now but less good for more heavy duty infrasonic stuff unless you tune really low.
Only real problem with the XLS in a vented enclosure is the daft port lengths required, might have to look at PR's but I really don't fancy that because I don't believe they sound as natural and they're also £45 a pop and I'll be needing at least 4 - thats the equivalent cost of another two XLS10's and sealed would then start to make sense because were talking similar costs for dual subs each with four XLS's vs. two XLS's + PR's.
I need a cunning way to fit a 170cm port into a cabinet that doesn't have abnormally proportioned dimensions. Its looking like some kind of folded vent is in order but I not loving that idea, perhaps if I keep the turns limited to a single 90 degree bend that I smooth out a little?
Any idea's?
richie00boy said:
I'm judging the XLS purely on modelled responses, the single XLS10 sealed I had a good while ago now was pretty poor in clean output terms although it did go fairly loud if you didn't mind audible distortion, in other words I was less impressed with the output but if you kept things suited then it did sound good from what I remember, TBH it was a real long time ago now and I've listened to and built better stuff since then, so maybe I was more easily impressed back then.
The B&C sounds very well with music, I think it will be hard act to follow. I've got the UcD700 up and running although I still don't have the case for it yet but its moved things on a notch or two in quality terms, although output at the low end didn't increase much over the ~200w I fed it before, Surprisingly its near impossible to bottom, I've only heard it do it a couple of times when really pushing it on the end of inadequate power. It simply starts to sound compressed so I'm thinking the suspension is designed to make it an unlikely event, that also means a chance of ripping the spider or surround apart.
If the XLS10's hadn't come up on ebay I wouldn't be bothering TBH. However I think they're worth trying out considering that folks almost hold this driver as a benchmark for all other 10's. I'm not sure about that myself but probably has much to do with SL's Orion. Its common knowledge that even average drivers in an exceptional design can sound good. Based on previous experience, modelling, opinions and measured data that out there I think it could be good in a vented enclosure.
ShinOBIWAN said:
OK now with this new information I accept that my recent comments are probably not so right now.
But before you write off the XLS in sealed, be aware that adding LT does not change distortion really. To illustrate, if you were going to drive it to a certain SPL at say 20Hz, when you add the LT the cone does not move any more at that frequency, merely the input power profile across the band is changed to give the desired response.
richie00boy said:
Agreed, it doesn't change distortion at a given SPL ie. 20hz @ 95dB is still the same with or without the LT.
But the LT does change the distortion profile in respect to frequencies that have less gain applied. So an LT assist with a flat response has a higher THD than one without the LT. All things relative I guess. My real point is that LT will highlight distortion whilst vented will actively assist to lower THD where its needed most and is more amp friendly.
Still giving sealed a go though 🙂
To quote, or paraphrase, SL's full assessment, he thought the XLS's were better open baffle than sealed, wrt distortion - and in light of Dickason's remarks on suspension design it makes sense to me. Basically if the design intent was vented or IB, then the suspension becomes intentionnally stiffer at excursion extremes to prevent positive feedback (less force -> even less pullback at high excursions). If the intent of the driver designer was sealed, then he'd try to make the suspension as linear as possible. It's horses for courses.
IIRC B&C state their preference for vented use for the TBX series, and talk of double spider and "optimized compliance", to me that sounds like there is a good reason why the B&C should sound clean in a vented box... and likely the XLS will also sound cleaner there than sealed.
In addition - why are the XLS famous and why dos SL use them - 1) they are the Xmax masters of the DIY universe, 2) open baffles need Xmax above all, and low air noise, and 3) SL's intent is / was to present a project to DIY enthusiasts and therefore tries to choose drivers that are easily available worldwide, this is what he states - this may be an additional choice criterium for the XLS other than measured performance. What I'm trying to say is, it's quite possible some more "exotic" or less easily available pro drivers such as your B&C may well better the XLS, depending on application.
All this is pure speculation on my part of course, I admit. Just trying to connect the dots.
IIRC B&C state their preference for vented use for the TBX series, and talk of double spider and "optimized compliance", to me that sounds like there is a good reason why the B&C should sound clean in a vented box... and likely the XLS will also sound cleaner there than sealed.
In addition - why are the XLS famous and why dos SL use them - 1) they are the Xmax masters of the DIY universe, 2) open baffles need Xmax above all, and low air noise, and 3) SL's intent is / was to present a project to DIY enthusiasts and therefore tries to choose drivers that are easily available worldwide, this is what he states - this may be an additional choice criterium for the XLS other than measured performance. What I'm trying to say is, it's quite possible some more "exotic" or less easily available pro drivers such as your B&C may well better the XLS, depending on application.
All this is pure speculation on my part of course, I admit. Just trying to connect the dots.
Been playing around with idea's and something struck me.
In that rough sketch I've got the port in the middle of the enclosure, the port is a slot type and runs the entire width of the cabinet(the sketch show a side perspective of the cabinet so this isn't shown). The top section of the cabinet is therefor only open to the drivers from the passage above the port, am I correct in assuming that the volume of air in this area will be fully 'seen' by the drivers and treat the same as the air volume directly around the drivers or will the port somehow create a lower pressure in this area?
I worked it like this primarily because the port provides cabinet bracing. However if it looks problematic its a quick fix to allow the port to run upto the top right corner of the cabinet.
EDIT: Actually thinking about more closely, its obvious that the whole volume of the cabinet is considered equal any pressure differences will be very small when dealing with low frequencies. Sorry for the senior moment 🙂
An externally hosted image should be here but it was not working when we last tested it.
In that rough sketch I've got the port in the middle of the enclosure, the port is a slot type and runs the entire width of the cabinet(the sketch show a side perspective of the cabinet so this isn't shown). The top section of the cabinet is therefor only open to the drivers from the passage above the port, am I correct in assuming that the volume of air in this area will be fully 'seen' by the drivers and treat the same as the air volume directly around the drivers or will the port somehow create a lower pressure in this area?
I worked it like this primarily because the port provides cabinet bracing. However if it looks problematic its a quick fix to allow the port to run upto the top right corner of the cabinet.
EDIT: Actually thinking about more closely, its obvious that the whole volume of the cabinet is considered equal any pressure differences will be very small when dealing with low frequencies. Sorry for the senior moment 🙂
BlackCatSound said:
Yeah its a problem. Family and friends tell me I'm crackers and I've started to get jokes about preferring speakers to women
Which is in a roundabout way true if I'm being honest. 😀
Ahh, true, but he also chose them for the Thor subwoofer, which really could have been anything (sealed, vented, PR, I guess even OB) and also chose to put this driver in a sealed alignment. So, this would at least seem to suggest that placing the XLS in a sealed alignment is a good idea for a subwoofer, according to SL anyway.
Also, I'm not trying to rain on parades, but at first glance the vented design drawing you showed does a lot of things that I'm not saying are "wrong", but they seem like they could be troublesome. Down firing drivers, port bends, port that changes shape halfway through, and the portion of the cabinet that houses the drivers seems like it could possibly be prone to some sort of "rocking" because it's being coulped to the floor in an odd fashion. Like I said, I don't actually know that any of this is "wrong" per say, but I'm always a fan of the KISS philosophy (Keep It Simple, Stupid). Too many question marks with that design for me to feel comfortable building it.
Side note: I know you already have some 10" original "XLS", but since then Peerless has released a new line of subwoofer drivers, the "XXLS" (XLS 2.0 I guess) and they actually make one that seems designed specifically to be used in a sealed alignment. Maybe something to consider.
http://www.tymphany.com/datasheet/pdf/pl/830843.pdf
http://www.tymphany.com/datasheet/appview.php?id=27
Has anyone tried the LAT drivers? Might be interesting to pack a few of those into a sub each side!
What do they sound like?
What do they sound like?
Tenson said:
I asked about the LAT's a good while ago now. Apparently the first generation is more for car audio rather than 20hz and down. I've never heard of anyone using these for home use yet, although there's a couple of nice car installs with them.
Hi Shin,
I think you prefer speaker building to everything else, including speakers - you'd be devastated if your system at some point actually sounded perfect to you 😀
Seriously though, you might want to consider spending more time on planning a new addition thoroughly, and less money on buying new components. With that I mean not so much box modeling (and in another thread I just saw derogatory comments specifically about WinISD's results and why it may be off - for example, displaying peak Xmax instead of RMS xmax - here goes a few dB). It's planning for what you want.
Consider the whole system holistically: what is most important to you - which trade offs do you prefer - how does the room play into this. I know this thread has had plenty of discussion on these subjects, but little in terms of a long term consensus. Make a spec sheet so to speak, and then try and see if you can achieve it using different design philosophies, ideally using the same components: a technical goal easier to work for than swapping drivers and find yourself with a new and unknown animal each time. And this would satisfy your curiosity for something new while being actually achievable.
So the novelty would come from trying to reach a new spec, or a new design approach, and not from trying a new component. Read: you can keep the components and put them in new boxes. Example: same component, as IB (would that not be a much better solution for your small room??), as vented aligned, as vented misaligned deliberately (Scott's design), as cardioid... and this would tell you more about the merits of specific design specs and how you like them, than changing components (because then you change two variables: component and application).
Maybe I'm still surprised how quickly you got tired of the vented B&C box without much trying to, say, changer the alignment to a more conventional one (that would give you mor output per Xmax at the expense of some distortion), or building the second box (even as a simple test mule without the bells and whistles of the first one, just same volume same alignment, to try stereo). I thought it was a very interesting design approach, not often used, to get distortion down and keep some excursion benefits.
Anyway, sorry to sound like a therapist but that was the intent
I think you prefer speaker building to everything else, including speakers - you'd be devastated if your system at some point actually sounded perfect to you 😀
Seriously though, you might want to consider spending more time on planning a new addition thoroughly, and less money on buying new components. With that I mean not so much box modeling (and in another thread I just saw derogatory comments specifically about WinISD's results and why it may be off - for example, displaying peak Xmax instead of RMS xmax - here goes a few dB). It's planning for what you want.
Consider the whole system holistically: what is most important to you - which trade offs do you prefer - how does the room play into this. I know this thread has had plenty of discussion on these subjects, but little in terms of a long term consensus. Make a spec sheet so to speak, and then try and see if you can achieve it using different design philosophies, ideally using the same components: a technical goal easier to work for than swapping drivers and find yourself with a new and unknown animal each time. And this would satisfy your curiosity for something new while being actually achievable.
So the novelty would come from trying to reach a new spec, or a new design approach, and not from trying a new component. Read: you can keep the components and put them in new boxes. Example: same component, as IB (would that not be a much better solution for your small room??), as vented aligned, as vented misaligned deliberately (Scott's design), as cardioid... and this would tell you more about the merits of specific design specs and how you like them, than changing components (because then you change two variables: component and application).
Maybe I'm still surprised how quickly you got tired of the vented B&C box without much trying to, say, changer the alignment to a more conventional one (that would give you mor output per Xmax at the expense of some distortion), or building the second box (even as a simple test mule without the bells and whistles of the first one, just same volume same alignment, to try stereo). I thought it was a very interesting design approach, not often used, to get distortion down and keep some excursion benefits.
Anyway, sorry to sound like a therapist but that was the intent
- Status
- Not open for further replies.