Closed box sub with Peerless SLS
Hi there,
I am planning to build a sub using a Peerless driver, but not the famous xls. I will use the sls range instead, because my plan is a closed box and xls has low Qts for that. (I don't want to build a passive radiator nor a reflex for a first diy sub project.) I'd like to get a system Q of around 0.6 (or 0.7 max), in a closed box. I'd like to get low group delay, high quality bass, not a big spl.
My questions:
- I made some calculations using several softwares (like the Winisd) and excel sheets. I have found that If I reduce the ideal box of, say, 150 l down to around 100 l then I get nearly the same results. Is this thing really this flexible? Can anyone (more experienced than me) run his software and advice a box size for me please?
- I made calculations with both the 10" and the 12" SLS driver, the results are almost exactly the same. Is it a mistake or this can really happen? If so, what is the reason? (Of course the ideal box is smaller for the smaller driver). Is it possible that the smaller driver is a better construction in some ways?
- Can I find any webpages about completed sealed box projects with this driver on the net?
- Can anyone advise other drivers of the same class to have a look at? (I haven't purchased the driver yet and I'm not stickin' to the type. By now you know what I'm searching for.)
Thanks,
m127
Hi there,
I am planning to build a sub using a Peerless driver, but not the famous xls. I will use the sls range instead, because my plan is a closed box and xls has low Qts for that. (I don't want to build a passive radiator nor a reflex for a first diy sub project.) I'd like to get a system Q of around 0.6 (or 0.7 max), in a closed box. I'd like to get low group delay, high quality bass, not a big spl.
My questions:
- I made some calculations using several softwares (like the Winisd) and excel sheets. I have found that If I reduce the ideal box of, say, 150 l down to around 100 l then I get nearly the same results. Is this thing really this flexible? Can anyone (more experienced than me) run his software and advice a box size for me please?
- I made calculations with both the 10" and the 12" SLS driver, the results are almost exactly the same. Is it a mistake or this can really happen? If so, what is the reason? (Of course the ideal box is smaller for the smaller driver). Is it possible that the smaller driver is a better construction in some ways?
- Can I find any webpages about completed sealed box projects with this driver on the net?
- Can anyone advise other drivers of the same class to have a look at? (I haven't purchased the driver yet and I'm not stickin' to the type. By now you know what I'm searching for.)
Thanks,
m127
Firstly I'd recommend the XLS driver over SLS. It is suitable for sealed applications as well. Peerless have released a new version which also has a higher Qts.
Two good examples:
Linkwitz Thor
Critical Q subwoofer
Another good driver is the Dayton Titanic MkIII
You don't need a big 150L box for a 12" sealed sub. 28 - 50L is more like it. Then you use a Linkwitz transform to get the response and Q right. There is a very interesting kit available at Rythmic Audio . It has a very good looking driver, and when used with the amp designed to go with it, which has a servo to reduce distortion, I suspect it's probably better than the XLS. This is probably the best quality small monopole sub a diyer can produce without XBL technology, which is quite a bit more expensive. Brian from Rythmic is very helpful.
Two good examples:
Linkwitz Thor
Critical Q subwoofer
Another good driver is the Dayton Titanic MkIII
You don't need a big 150L box for a 12" sealed sub. 28 - 50L is more like it. Then you use a Linkwitz transform to get the response and Q right. There is a very interesting kit available at Rythmic Audio . It has a very good looking driver, and when used with the amp designed to go with it, which has a servo to reduce distortion, I suspect it's probably better than the XLS. This is probably the best quality small monopole sub a diyer can produce without XBL technology, which is quite a bit more expensive. Brian from Rythmic is very helpful.
Thank you for the idea of the Linkwitz transform! My original idea was to use the "Parapix" amp and its filters, now I see I will combine the two. (I have to, since the Linkwitz circuit doesn't have an input circuit, volume control, high-pass or low-pass filters.) What I don't know is the order in which I should connect the "stages". My idea is (please correct me if I am wrong): 1. Input (High level), 2. volume adj., 3. Low pass filter (crossover), 4. High Pass filter (say, 15 Hz), 5. Phase control stage, 6. the Linkwitz circuit. I don't know if it's correct. Maybe the high- and low-pass filters should go to the end of the line.
I still haven't decided about the driver (xls or sls). And still couldn't find out why the 10" sls looks better (or not worse at least) than the 12" sls (according to winisd, in the properly sized boxes). Maybe the best solution is to use two pieces of the smaller one. 🙂 By the way, can I use the Linkwitz transform for two drivers in a box? How do I enter the parameters in that case?
I still haven't decided about the driver (xls or sls). And still couldn't find out why the 10" sls looks better (or not worse at least) than the 12" sls (according to winisd, in the properly sized boxes). Maybe the best solution is to use two pieces of the smaller one. 🙂 By the way, can I use the Linkwitz transform for two drivers in a box? How do I enter the parameters in that case?
Another thing.
Can anyone give me some advice how to build a high level input properly?
I don't want any crosstalk or any kind of an influence on the front speakers at all, but I want the sub to change its volume according to the volume of the front loudspeakers, that's why I need those high level inputs.
Can anyone give me some advice how to build a high level input properly?
I don't want any crosstalk or any kind of an influence on the front speakers at all, but I want the sub to change its volume according to the volume of the front loudspeakers, that's why I need those high level inputs.
The reason for using high level inputs is normally that you don't have a sub line level out or pre-outs on your amp. The advantage of speaker level inputs is that you can add a sub to any stereo amp that you might have.
I have no idea why anyone would consider the SLS driver over the XLS. It doesn't have enough excursion to cut it as a sealed sub.
For a single XLS a plate amp makes life easier! Others may be able to help you more with how to design active filters. 2 channel power amps become more economical when you are building 2 subs.
For modelling I suggest WinISD pro which is much more powerful than the standard version. It lets you play with filters and see excursion.
It also does linkwitz transforms. Make sure you make some allowance for room gain or even better eq for room modes with BFD.
Dual drivers allow you to push pull mount for lower distortion but a single larger driver gives you more bang for buck.
I have no idea why anyone would consider the SLS driver over the XLS. It doesn't have enough excursion to cut it as a sealed sub.
For a single XLS a plate amp makes life easier! Others may be able to help you more with how to design active filters. 2 channel power amps become more economical when you are building 2 subs.
For modelling I suggest WinISD pro which is much more powerful than the standard version. It lets you play with filters and see excursion.
It also does linkwitz transforms. Make sure you make some allowance for room gain or even better eq for room modes with BFD.
Dual drivers allow you to push pull mount for lower distortion but a single larger driver gives you more bang for buck.
Peerless XXLS
I have also been thinking of the SLS in sealed enclosures to fill in the bottom end of a Fostex Fullrange system i have.
Meantime, as paulspencer mentioned, peerless has recently come up with a new series caled XXLS, including 10" and 12" that are especially desiged for sealed boxes.
http://www.d-s-t.com/main/tech/xxls_intro.htm
http://www.d-s-t.com/peerless/dataover/ho_su_xxls.htm
It might be wortwhile to check them out..
LageB
I have also been thinking of the SLS in sealed enclosures to fill in the bottom end of a Fostex Fullrange system i have.
Meantime, as paulspencer mentioned, peerless has recently come up with a new series caled XXLS, including 10" and 12" that are especially desiged for sealed boxes.
http://www.d-s-t.com/main/tech/xxls_intro.htm
http://www.d-s-t.com/peerless/dataover/ho_su_xxls.htm
It might be wortwhile to check them out..
LageB
Well, my reason for the high level inputs is as follows:
- I have a HT amplifier with low level LFE sub out, but of course I do not listen to music on a Home Theatre amp! For that I have built a stereo amp (a pair of Natsemi LM1875s) which is a thousand times better than my Denon AVR. But the sub I am planning to build will be used for music as well, and my stereo amplifier doesn't have sub out. (Actually it doesn't have anything extra which is not abolutely necessary).
- I want the loudness of the sub to change automatically if I set the volume of the main speakers.
XLS/SLS question:
- XLS is somewhat pricey here in Hungary, much more expensive than in the rest of Europe. SLS is OK.
- Even more important than the previous, all the calculations I made with winisd pro said that I can get the best results with the 10" SLS. Don't ask me why. (The 12" SLS is also not better at all.) My calculations told me that the xls is made for vented enclosures or passive radiator systems. (OK, I haven't heard of the new version until today). If I have a 64 l closed box for starters (which will be corrected by Linkwitz later), the original box has a lower resonance with the sls than it has with the xls. If the box and the driver is closer to what I need then the Linkwitz circuit doesn't have that much to correct and it's also easier for the amp.
- I have a HT amplifier with low level LFE sub out, but of course I do not listen to music on a Home Theatre amp! For that I have built a stereo amp (a pair of Natsemi LM1875s) which is a thousand times better than my Denon AVR. But the sub I am planning to build will be used for music as well, and my stereo amplifier doesn't have sub out. (Actually it doesn't have anything extra which is not abolutely necessary).
- I want the loudness of the sub to change automatically if I set the volume of the main speakers.
XLS/SLS question:
- XLS is somewhat pricey here in Hungary, much more expensive than in the rest of Europe. SLS is OK.
- Even more important than the previous, all the calculations I made with winisd pro said that I can get the best results with the 10" SLS. Don't ask me why. (The 12" SLS is also not better at all.) My calculations told me that the xls is made for vented enclosures or passive radiator systems. (OK, I haven't heard of the new version until today). If I have a 64 l closed box for starters (which will be corrected by Linkwitz later), the original box has a lower resonance with the sls than it has with the xls. If the box and the driver is closer to what I need then the Linkwitz circuit doesn't have that much to correct and it's also easier for the amp.
Have you got quotes for shipping the XLS?
Better in what regard?
My guess is that the distortion will be higher and the output lower. To try to match the performance of the XLS, you would have to use so many of the SLS that it would end up costing more for an inferior driver. I can only assume that by better you mean having a better unequalised response. This is not relevant in a sub that you plan to eq!
64L is much bigger than you need for a sub using a LT. Even for a pair of 12" drivers. Just about any decent high excursion driver should be fine in a 28L box. The idea of using a LT is that you want a small box without it being boomy, but you must have sufficient excursion or the extension won't be matched with decent output. The SLS subs won't get much output in a small LT sub. You would have to give up considerable output or extension.
Better in what regard?
My guess is that the distortion will be higher and the output lower. To try to match the performance of the XLS, you would have to use so many of the SLS that it would end up costing more for an inferior driver. I can only assume that by better you mean having a better unequalised response. This is not relevant in a sub that you plan to eq!
64L is much bigger than you need for a sub using a LT. Even for a pair of 12" drivers. Just about any decent high excursion driver should be fine in a 28L box. The idea of using a LT is that you want a small box without it being boomy, but you must have sufficient excursion or the extension won't be matched with decent output. The SLS subs won't get much output in a small LT sub. You would have to give up considerable output or extension.
All right, I think you are right. I meant that the sls was better in unequalised box performance. By now I know that this is not a key factor if the sub will use a LT. I was just thinking that it's better if the performance is better so as the LT doesn't have to work that much. I will check the price of the new xls, or try to find a solution to get the xls from abroad at a reasonable price.
Anyway, how do I find the best box volume? I see that it's corrected by the LT anyway, but the help says somewhere that the situation is better if the box is better to start out with.
What I still don't know is:
-The order in wich I should connect the parts of the preamp (High pass, low pass, LT, volume control)
-How to make a high level input
Anyway, how do I find the best box volume? I see that it's corrected by the LT anyway, but the help says somewhere that the situation is better if the box is better to start out with.
What I still don't know is:
-The order in wich I should connect the parts of the preamp (High pass, low pass, LT, volume control)
-How to make a high level input
Keep in mind there is a LOT more to performance than the frequency response. You must also consider (in no particular order) transient response, distortion, power compression and the interaction with room gain and finally (even more important than all of that) room modes.
Size is not so critical as long as it is reasonable. If you put a single 12" XLS in a reasonably solid 15" cube it should be fine.
Is there any particular reason why you don't just use a plate amp? Normally they will accept high level inputs and also roll off the bass to the mains if desired (first order).
Another option is to look at Rythmic Audio's direct servo kit. I would say for musical performance in a compact sub this would be hard to beat.
Size is not so critical as long as it is reasonable. If you put a single 12" XLS in a reasonably solid 15" cube it should be fine.
Is there any particular reason why you don't just use a plate amp? Normally they will accept high level inputs and also roll off the bass to the mains if desired (first order).
Another option is to look at Rythmic Audio's direct servo kit. I would say for musical performance in a compact sub this would be hard to beat.
The two things I kept most important was freq response and group delay. (Though group delay is not that critical in closed subs as far as I know).
About the amp: I would like to build the amp, that's all. I am interested, this is not my first amp, I have some schematics that I trust, and it will come much cheaper this way. Another advantage is that I can play with the parameters to find what's best, which I can't do with a new amp (at least not in the guarantee period). Actually, my plan is to build the sub passive, and the amp will be built very near to the stereo amp (for the test period it will be mounted on the same wooden plate). If I don't manage then I will buy one of the ready-made plate amps. 🙂
About the amp: I would like to build the amp, that's all. I am interested, this is not my first amp, I have some schematics that I trust, and it will come much cheaper this way. Another advantage is that I can play with the parameters to find what's best, which I can't do with a new amp (at least not in the guarantee period). Actually, my plan is to build the sub passive, and the amp will be built very near to the stereo amp (for the test period it will be mounted on the same wooden plate). If I don't manage then I will buy one of the ready-made plate amps. 🙂
I used to worry about GD but I hear it is debatable whether or not it is audible. What I have found is that when a sealed sub is eq'd flat to match a vented box, the sealed box has:
* similar GD
* more excursion, therefore lower power handling and higher distortion
* is not as linear
than a vented box
I'm not sure whether the subjective tightness of a sealed sub is due to GD (esp when not eq'd) or the way in which it integrates with room gain. I suspect it is probably the latter.
* similar GD
* more excursion, therefore lower power handling and higher distortion
* is not as linear
than a vented box
I'm not sure whether the subjective tightness of a sealed sub is due to GD (esp when not eq'd) or the way in which it integrates with room gain. I suspect it is probably the latter.
Also, I have found that, in Australia at least, it can easily cost more to diy your sub amp but YMMV.
I don't worry about the costs of building the amp, I have most of the parts in "my stock". 🙂
I don't know if I've managed to catch what you mean by saying the things about the "too much eq'd" sealed sub. Are you saying that I can easily end up with a box worse than a vented one? If this is the point, then this is why I wanted to build a larger-than-minimal box, so as not to get to the "too much eq'd" state. Am I right?
I don't know if I've managed to catch what you mean by saying the things about the "too much eq'd" sealed sub. Are you saying that I can easily end up with a box worse than a vented one? If this is the point, then this is why I wanted to build a larger-than-minimal box, so as not to get to the "too much eq'd" state. Am I right?
The XLS is designed for a small box anyway. 28L should be fine. Double the volume and use a LT and you will not hear a difference IMO.
My point is that when you use eq, you increase GD as well as distortion due to the increase in cone excursion. This is true of any sub with a LT. Just have a look at the amount of eq you need to be sure it is not excessive. It's also good to make some kind of allowance for room gain. I'd allow for about 6db room lift at 20 Hz, which starts at about 40 Hz. This reduces the eq required.
You must keep in mind that as you go lower, while GD and distortion goes up, your perception decreases. GD may not be audible anyway, this is a subject of debate as far as I know.
My point is that when you use eq, you increase GD as well as distortion due to the increase in cone excursion. This is true of any sub with a LT. Just have a look at the amount of eq you need to be sure it is not excessive. It's also good to make some kind of allowance for room gain. I'd allow for about 6db room lift at 20 Hz, which starts at about 40 Hz. This reduces the eq required.
You must keep in mind that as you go lower, while GD and distortion goes up, your perception decreases. GD may not be audible anyway, this is a subject of debate as far as I know.
How about stuffing the box? That will increase the virtual size of the box so simply ruin the calculations. Or you just add an estimated percentage bigger box and calculate with that? Or you leave the box empty?
How do you count with the room gain when designing the LT? Do you just calculate with a box that's bigger than the one you actually build? How can you "tell" the system to have 6 db smaller gain at 20Hz than the ideal flat response? (As far as I see the LT calculators work to find the ideal flat curve.)
How do you count with the room gain when designing the LT? Do you just calculate with a box that's bigger than the one you actually build? How can you "tell" the system to have 6 db smaller gain at 20Hz than the ideal flat response? (As far as I see the LT calculators work to find the ideal flat curve.)
paulspencer said:
I would argue against this in the case of the LT. An LT is not EQ in the traditional sense where the phase and GD gets mucked up for the sake of better amplitude response. Because of the complex pole manipulation you get to have your cake and eat it -- you can achieve no worsening of GD if you want, in fact you can even (in the case of going from un-EQ'd box with high Q to EQ'd to low Q) make it better.
m127 said:
You first establish the room characteristics then design the LT to integrate with that. As a rough starting point, room gain starting frequency is 344/(longest dimension/2) (assuming metres) and rises at approx 12dB/octave. So if your gain starts at 40Hz then just make your LT go down to 40Hz.
Hm. Something wrong here. The longest dimension of my room is 5.4 m. If I use the formula I get 126 Hz of room gain starting point! That of course cannot be correct! Or is it misunderstood?
Care to demonstrate this?
My sims have indicated otherwise. I have an XLS in a LT sealed and vented box and the GD of sealed LT and vented are very similar.
I don't intend to be argumentative about this, but if there is a way that a performance edge could be given to the sealed LT box, I'd like to know.
I think there is some error in the forumula.
12db sounds a bit much as a general guide. I'd guess at more like 6db/octave starting at about 40 Hz for a medium room. Better than a guess is to actually simulate room gain with the FRD Room Reflection Response Calculator, which I find very handy.
- Status
- Not open for further replies.