AVS Forum banner

1 - 4 of 4 Posts

·
Registered
Joined
·
861 Posts
Discussion Starter #1
Hi all


Am about to make a test of concept' bass unit, based on the one done by Shinobiwan in his Winter Sub project thread, albeit nowhere near as pretty ha ha.


Firstly, the drivers are the peerless xls 10inch (four off) and I'm pretty certain I will be running them sealed. I only wish/need them to go up to around 40-50hz. My current mains go flat in room to 29-30 hz, lower than that and they fall off very quickly. The mains bass units are sealed, with a lot of eq to get them that low. So my thoughts are that if I can get the subs working nicely, get them low without much strain or effort then I can take some of the load off the mains and end up with a less stressed set up that also goes lower. Not that I can hear much strain with the mains at the moment, but logic dictates that relieving the load (as they are so heavily boosted) should make everything a little more relaxed.


From what I can gather, Shinobiwan first and foremost designed them as vented, but they could also be run sealed for means of comparison. Being vented means that the box is bigger than it absolutely needs to be if sealed?? He has the drivers seeing a volume of 130 l, giving a Q of around .37. Is there any reason not to raise the Q to (say) .5 and also benefit from a smaller box?? From memory, that is the Q of my mains so perhaps they will integrate a bit better as well?? (will check that at a later date).


In his thread he linked to a review of a commercial sub that uses this driver, and it was sealed with a volume of 18l, so if I use four of them does it make sense to simply multiply that volume by four???


I have modelled these in Winisd, the graph doesn't look too flash, but then we have to factor in room gain etc and then whatever boost I may need. Hopefully not too much boost, as that is one of the reasons to use the sub as explained above, too much boost here and it may be out of the pan and into the fire!


The main thing I'm testing is how to integrate it into my system, and if that works then for symmetry I will build another for the other side. That should all help with the boost aspect. Regarding room gain, does that only manifest when the sub is close to the wall or corner? The reason I ask is that I want to have them next to my mains, but I have the mains around two metres off the wall, so maybe the sub won't get much help after all.


I had a computer glitch a little while ago, which meant I had to re-download winisd. It seems to be different than what I remember, I recall extra windows or whatever that allowed me to check on driver excursion etc, they seem to have gone??


Anyway, a few months ago I recall mucking about with those other options', and one was mass loading. By adding (10%?) mass to the driver, it seemed to remove the need for eq (which was another of the options' in winisd) and ended up with as flat a graph as with the eq.


Engineering being the compromise that it is, I'm assuming that mass adding has it's drawbacks?? What might they be, and why would someone choose eq over mass adding (or vice versa) to get the flat response??


My current system is tri-amped using the DEQX as the pre. So I hope that a suitable means of integrating the subs is to take a splitter from the bass out of the deqx and running that to the sub.


I will only correct the mains down to (say) 50 hz, and then being sealed they drop off at 12db/octave I think. In any case I can find that out by measurement. Conversely, with the sub I will run the signal thru a behringer dcx 2496 and have them roll off in a mirror image fashion from the mains. Hopefully as they will be physically closely located (something to do with wavelengths presumably) then they can be considered to be a single unit and the two roll offs should sum to a flat response.


Is there something I've overlooked in this approach???(I've never really done this before, and I'm not too flash when it comes to cable hookups etc ).


If this approach works then that has the advantage of still allowing the deqx remote control to be able to tailor the response if the recording is crap.


If the above doesn't work then I suppose I could put a passive network between the mid and tweeter (and treat them as one driver') and so free up another channel on the deqx for the sub. I'd rather not go that way unless I have to.


Thanks for that, hope I didn't bore your socks off ha ha, and please if you can spot any flaws in my approach or have any suggestions then fire away.
 

·
Registered
Joined
·
5,196 Posts
Is there any physical reason you're tied to the XLS10's? Do you already have them? The XLS12's are not much more expensive but have a lot more Vd, and in my opinion sound even cleaner than the 10's at all volume levels.


If for whatever reason the 10's it is, then I think you're on the right track with box size (a well-stuffed 50L box looks good for four XLS10's) and system integration, but I wouldn't mass load. Mass-loading is basically what was done before cheap EQ and massive power were available. It's basically just passive EQ, cutting efficiency to give the impression of more bass. Also, I'd be loathe to mess with such a precisely engineered machine as the XLS. Better to just use EQ.
 

·
Registered
Joined
·
861 Posts
Discussion Starter #3
I was considering using a servo from Rythmik, but for some unknown reason the american postal system no longer does surface shipping so only air freight. That kinda really blew the price out, and considering I already had the tens laying around couldn't quite justify the cost if you get my drift.


Yes, bad advice I suppose many years ago but I have the tens so I may as well put them to use. Hopefully the cleaner sound of the 12's will be reached with the tens simply due to the numbers used and so they will be only cruising.


Shinobiwan seemed to quote distortion figures of 10% ( or whatever the exact figure was) when at high volume levels, not sure if that is a lot of distortion or kinda normal distortion. The servos would have less, but as I say hopefully the numbers I may end up using will help keep the numbers down.


Forgot to say earlier I intend to simply have them one above the other in a vertical array, mainly for aesthetics I suppose. Would there be any problems from excess vibration ?, should they be on opposite faces?? Would rather not if all on one face would work.


I assume the wiring will simply be two pairs of drivers wired in series, in parallel?


At these frequencies I seem to vaguely recall from somewhere (Seigfried Linkwitz's site??) that no stuffing is required, would that be acceptable and if not what sort and amount of stuffing would would I use. Put it this way, stuffing wouldn't actually harm things I suppose. (sorry, just re-read the post and you already mentioned 'well-stuffed',,,dohh.)


Understand re the added mass, today I guess that eq is easy and amp power is cheap means no longer any need for old techniques.


well, seems that all I do now is go and cut some wood. thanks for the reply, and as no-one else has jumped in with dire warnings I guess I can't be doing too much wrong.
 

·
Registered
Joined
·
5,196 Posts

Quote:
Originally Posted by terry j /forum/post/0


Yes, bad advice I suppose many years ago but I have the tens so I may as well put them to use. Hopefully the cleaner sound of the 12's will be reached with the tens simply due to the numbers used and so they will be only cruising.

That changes the ballgame considerably. Using what one has is always better than keeping it in the closet and buying new stuff!


Quote:
Forgot to say earlier I intend to simply have them one above the other in a vertical array, mainly for aesthetics I suppose. Would there be any problems from excess vibration ?, should they be on opposite faces?? Would rather not if all on one face would work.

My preference would be two on opposing faces, perhaps even with one side wired out of phase and firing inside-out.


The reason for the former is that the boxes can be much lighter, due to the net force exerted on the box being zero.


The reason to do two inverted is because that theoretically lowers distortion due to suspension nonlinearity. However, the XLS series is so well designed that it may not amount to anything worth writing about, even though the XLS baskets look so cool!



Certainly, you can do them all on one face, but then the box has to be stouter and more extensively braced. My triple XLS12+XLS12PR sub, with the drivers firing 60deg from one another for net force cancelation, is going to be made out of two sheets of .25" MDF with a constrained layer (fiberglass) between and a few internal braces. Hopefully my cabinetmaker will finish it by late next week.

Quote:
I assume the wiring will simply be two pairs of drivers wired in series, in parallel?

That's right.

Quote:
At these frequencies I seem to vaguely recall from somewhere (Seigfried Linkwitz's site??) that no stuffing is required, would that be acceptable and if not what sort and amount of stuffing would would I use. Put it this way, stuffing wouldn't actually harm things I suppose. (sorry, just re-read the post and you already mentioned 'well-stuffed',,,dohh.)

You don't have to, really. The only difference is that box volume goes up a bit if you don't. I generally stuff with about 1lb per cubic foot (~16g/L), sometimes removing some after listening.

Quote:
Understand re the added mass, today I guess that eq is easy and amp power is cheap means no longer any need for old techniques.

Both work well, mind, but EQ has the decided advantage of not requiring one to modify (and possibly ruin) perfectly good woofers.

Quote:
well, seems that all I do now is go and cut some wood. thanks for the reply, and as no-one else has jumped in with dire warnings I guess I can't be doing too much wrong.

With such good drivers and (given your comments about postage) I assume a dwelling built to higher standards than the cardboard-and-spit construction of the typical new American McMansion, I'd be shocked if the sub was anything less than spectacular.
 
1 - 4 of 4 Posts
Top