PJ Screen Stage Sub Enclosure Design

[DrWho]

GPM wrote:

Since this is the OP's first attempt at DIY audio though, I hadn't planned to leave him 'swinging in the breeze' if he wants to pursue a TH solution and why I made the effort to find an appropriate driver/alignment for the usable 160 Hz BW required for an 80 Hz LFE XO, but I see he's had an overnight change of heart, so no sense in me wasting any more time on it.

Hornresp and Akabak are great tools, but like all engineering software they make some assumptions that may not be inconsequential to the intended performance, which is another way of saying that real measurements don't always (read rarely) correlate to the model.

With that in mind, I look at the posted hornresp prediction and I would only call it usable to about 40Hz....certainly not 160Hz (yikes). And even then, the frequency response doesn't tell the whole story.

All that craziness above 80Hz in your model is the result of reflections inside the horn that are causing resonances. There is no way in the world that you could expect to EQ that out as even the slightest change in temperature inside the room will shift all those frequencies around. , even slight build inconsistencies will change it too. You really can't EQ out reflections anyway because it's a transient problem.

I also think aiming for 15Hz (or even 20Hz) extension, given the constraints, is kinda throwing away a lot of available perceived performance. 25-30Hz for an F3 would be more reasonable in my mind, but that's probably opening a whole can of worms on this forum...let me just say there are reasons that commercial subwoofer manufacturers rarely try to dig into the teens (and it has nothing to do with not knowing what their customers want to experience).

I believe a single Peavey 18" Low Rider ($200 each) should be able to do a good 120dB down to 20-24Hz given the right enclosure. Two of them would add another 6dB.

For what it's worth, it might make sense to go with just a single 18" driver that costs about $300, but build both sides of the theater as if you were building two subs. Then when money permits later on, you could add a second driver and increase your SPL by 6dB (which would also be a nice way to give you some room for improvement when you upgrade the mains later on too).

In fact, just one AV15X should be good for around 120dB if my memory isn't failing too horribly:
http://aespeakers.com/shop/catalog/product_info.php?cPath=31&products_id=68
I really enjoy the AV drivers from AE....they sound great at all SPLs and their specs are very real. You might want to consider their lambda drivers later on down the road for when you DIY your mains.

I was also thinking about your room, and I think you might be better off putting the subs closer to the middle instead of out on the sides. You'll get better coverage across the couch and better overall midbass performance...especially when you've got two subs running. It basically comes down to reasons of polar response and off-axis lobing, and your room is wide enough that you'll get extra attenuation of the early side reflections which will also further reduce their impact on the sound.

[annunaki]

DrWho wrote:

Hornresp and Akabak are great tools, but like all engineering software they make some assumptions that may not be inconsequential to the intended performance, which is another way of saying that real measurements don't always (read rarely) correlate to the model.

With that in mind, I look at the posted hornresp prediction and I would only call it usable to about 40Hz....certainly not 160Hz (yikes). And even then, the frequency response doesn't tell the whole story.

All that craziness above 80Hz in your model is the result of reflections inside the horn that are causing resonances. There is no way in the world that you could expect to EQ that out as even the slightest change in temperature inside the room will shift all those frequencies around. , even slight build inconsistencies will change it too. You really can't EQ out reflections anyway because it's a transient problem.

I also think aiming for 15Hz (or even 20Hz) extension, given the constraints, is kinda throwing away a lot of available perceived performance. 25-30Hz for an F3 would be more reasonable in my mind, but that's probably opening a whole can of worms on this forum...let me just say there are reasons that commercial subwoofer manufacturers rarely try to dig into the teens (and it has nothing to do with not knowing what their customers want to experience).

I believe a single Peavey 18" Low Rider ($200 each) should be able to do a good 120dB down to 20-24Hz given the right enclosure. Two of them would add another 6dB.

For what it's worth, it might make sense to go with just a single 18" driver that costs about $300, but build both sides of the theater as if you were building two subs. Then when money permits later on, you could add a second driver and increase your SPL by 6dB (which would also be a nice way to give you some room for improvement when you upgrade the mains later on too).

In fact, just one AV15X should be good for around 120dB if my memory isn't failing too horribly:
http://aespeakers.com/shop/catalog/product_info.php?cPath=31&products_id=68
I really enjoy the AV drivers from AE....they sound great at all SPLs and their specs are very real. You might want to consider their lambda drivers later on down the road for when you DIY your mains.

I was also thinking about your room, and I think you might be better off putting the subs closer to the middle instead of out on the sides. You'll get better coverage across the couch and better overall midbass performance...especially when you've got two subs running. It basically comes down to reasons of polar response and off-axis lobing, and your room is wide enough that you'll get extra attenuation of the early side reflections which will also further reduce their impact on the sound.

Mike is on the right path here. I was also going to suggest a single high quality 18" sub vented and add another down the road. Something like the Maelstrom-X from Exodus Audio. You could tune to around 17hz-18hz and still have excellent response all the way up to 100hz if needed. AE also makes some very nice drivers and are a great option.

Set a path to do it as best as possible the first time around rather than compromise, compromise, compromise. You will be much happier in the long run and spend less money in the process.

Another option you may have is to build a "hollow stage" with a removable facade that has a flushed opening area where you could place different enclosures in at any time. This would open up many options on the upgrade path, rather than being tied to a specific design.

[chadcummings]

I am using the same subs, DPL15. I bought 8 of them to ensure I do not reach the max excursion while maintaining a high SPL. The one difference is I am running about 7.5x Vas, not a true IB but fairly close.

[GPM]

DrWho wrote:

Hornresp and Akabak are great tools, but like all engineering software they make some assumptions that may not be inconsequential to the intended performance, which is another way of saying that real measurements don't always (read rarely) correlate to the model.

With that in mind, I look at the posted hornresp prediction and I would only call it usable to about 40Hz....certainly not 160Hz (yikes). And even then, the frequency response doesn't tell the whole story.

All that craziness above 80Hz in your model is the result of reflections inside the horn that are causing resonances. There is no way in the world that you could expect to EQ that out as even the slightest change in temperature inside the room will shift all those frequencies around. , even slight build inconsistencies will change it too. You really can't EQ out reflections anyway because it's a transient problem.

True, and yet not so much as presented since it's too general a statement; otherwise how do you explain away that when properly measured, low inductance sealed and BR designs show close correlation to even the most basic of speaker box calculators in the alignment's pass-band where its HF corner is defined as 2*Fs/Qts?

Specific to HR and AkAbak and especially WRT TL, tapped alignment design, the latter is somewhat more accurate due to making fewer design assumptions, ergo more user input is required, but HR has been proven to be accurate enough to show a basic overall response that can be turned into a good performing speaker over the desired gain BW when the 'close enough' right driver specs are used, which I've done IF its published specs are ~accurate. Even if not, they can be 'forced' to work, trading efficiency for BW same as with any other type box alignment, though obviously this wouldn't be my first choice if there was another driver within budget that was closer to spec.

Hmm, your assertions seem to be based on a basic ~two octave TP/TH alignment which as you note doesn't have any useful BW above the pipe/horn's 3rd harmonic null whereas my sim is a three octave alignment where the 3rd harmonic should be acoustically damped enough to provide the design goal of a ~flat BW from ~15-160 Hz or at worst, its F3 points.

Regardless, if by 'reflections' you mean its half WL horn gain BW, then it is usable once damped same as any TL or horn alignment with the lower octaves being its quarter WL TL action. The spikes are reflections due to the too small mouth that again, once damped by bends and acoustic insulation won't be an issue, especially since the horn's relatively high acoustic damping compared to a two octave variant will flatten them much more that HR predicts.

Once 'critically damped' acoustically, its response/phase/impedance should be flat, extended enough for typical room EQ and XO solutions to apply same as any other speaker alignment with the caveat that if the TH isn't placed somewhere at/behind the viewing position an outboard time delay unit will probably be required due to the TH's long path-length as I doubt a typical HT processor or receiver has enough mains delay adjustability otherwise.

If you doubt any of this, you need only look to DSL's TH-Spud for proof and there's a wealth of info including some design insight from Tom Danley in the various TP/TH threads over on the DIYaudio Subwoofer forum which includes a good performing real world DIY ~clone that highlights the impact stuffing has on its response same as it does in a TL, so maybe comparing my design to its single driver conic horn variant (in shadow) will help clarify some of this.

GM

[GPM]

DrWho wrote:

I also think aiming for 15Hz (or even 20Hz) extension..........

For what it's worth, it might make sense to go with just a single 18" driver...........

I was also thinking about your room............

I agree that when forced to make significant BW Vs efficiency trade-offs, adding extension below 20 Hz (or even 30 Hz) isn't acceptable since with the exception of a relatively few movies and pipe organ symphony CDs there's not enough content to justify it and why bass shakers can be an acceptable alternative in a HIFI/HT app.

WRT this forum's seeming preoccupation with 18" driver alignments, there's lots to like about them when there's enough acoustic space available since at a glance it takes two 15" drivers to match its acoustic efficiency, especially now that there's enough selection to make them competitive price-wise with dual 15s. Then there's always the 'look'/'bragging rights' factor of big drivers, especially multiples, at least among the male population. I did it to get acoustic efficiency high enough to meet my dynamic range performance goals at vanishingly low distortion, but the latter's unintended added 'perks' were nice too.

Still, the needs of the app should be the prime consideration, so WRT Chris's chosen trade-off of screen height Vs stage height, the driver will have to be angle mounted. Historically though, no massive, high excursion driver has been able to perform well for any extended period of time loaded this way AFAIK, so unless any suitable spec'd 18s that have a proven record of being able to do so are presented, I strongly recommend he stick to his original multiple 15" solution and add more subs around the room later if either more peak SPL and/or better averaging out of room modes in the (sub)woofer's BW is desired. I agree though that a vented alignment will yield a more satisfying HT experience overall if an acceptable multiple sealed one can't be found, making the DPL-15 a so-so choice for sealed and unacceptable for vented.

While placing the drivers close together in the center has much merit, having them outboard, but toe'd in to the viewing position's 'sweet spot' seems a better choice overall since through the XO BW it will have the same advantages as it does for the mains, i.e. even less chance of early reflections and a more diffuse sound field behind the listening position.

GM

[elerium]

annunaki wrote:

Another option you may have is to build a "hollow stage" with a removable facade that has a flushed opening area where you could place different enclosures in at any time. This would open up many options on the upgrade path, rather than being tied to a specific design.

I'm now thinking something like this hollow stage idea would be the best route forward (or safest anyway). Based on some of the previous posts, it is apparent that I may not be satisfied with the SPL delivered by just two Exodus DPL-15s in a sealed stage box as originally proposed. I am working on a new design idea that will be similar to the original stage shape but will only be clad on five sides. The front side will only be covered with acoustically transparent fabric that could be removed when needed. This would allow me to re-position subs underneath the stage, add additional subs if needed and just provided more options in terms of enclosure designs (sealed/ported boxes). I will post the new stage design once it is completed to some feedback on it as well.

annunaki wrote:

Mike is on the right path here. I was also going to suggest a single high quality 18" sub vented and add another down the road. Something like the Maelstrom-X from Exodus Audio. You could tune to around 17hz-18hz and still have excellent response all the way up to 100hz if needed. AE also makes some very nice drivers and are a great option.

Set a path to do it as best as possible the first time around rather than compromise, compromise, compromise. You will be much happier in the long run and spend less money in the process.

I don't think I will be able to fit an 18" driver in the intended installation location which should be more apparent once I post the new "hollow stage" design. 15" drivers shouldn't be a problem so I'm thinking that two ported 15" subs might be a good starting point. Any driver recommendations would be appreciated. Enclosure size shouldn't be a problem - each sub box could be as large as 600L without much trouble (although height limitations still apply).

[elerium]

GPM wrote:

Since this is the OP's first attempt at DIY audio though, I hadn't planned to leave him 'swinging in the breeze' if he wants to pursue a TH solution and why I made the effort to find an appropriate driver/alignment for the usable 160 Hz BW required for an 80 Hz LFE XO, but I see he's had an overnight change of heart, so no sense in me wasting any more time on it.

I just wanted to say thanks for your effort and time trying to help me out and I apologize for being misleading on the tapped horn idea. I wasn't initially aware of the complexity of tapped horn sub designs and as I discovered more I realized it wasn't something that I wanted to take on as my first speaker project. I'm now thinking that two ported 15" subs would be the best and I am planning to modify the stage design accordingly (see above post). Any input in drivers and enclosure designs for ported 15" subs would be appreciated.

[Ricci]

I'm just curious but why would you have to mount a pair of 18's slanted to use them? The graphic shows a height of 24" for the stage. I've got an 80lb 18 driver in a baffle no bigger than 21.5x21.5. 24" would be more than plenty to mount them firing normally. The extra cone area and displacement of 18's would help to pressurize that large room. I'd suggest a pair of Fi Q18 D1's for a large ported alignment.

If you are intent on 15's then the CSS SDX15's are a solid 15" driver.

[GPM]

elerium wrote:

I'm now thinking something like this hollow stage idea would be the best route forward (or safest anyway).

Sounds like a plan as this was to be ~ my suggestion had you chosen to build the THs since it would be necessary to 'pop its top' to adjust their tuning with damping. Once dialed in, encase/'float' them in individual removable shells with a layer of foam sheeting like used under some flooring systems and finally, top off the 'stage' with one or more sections of removable, velcro'd 'flooring'. Having assembled/knocked down/auditioned several of these portable 'stages' at trade shows, I'm satisfied that it's a proven concept, though don't know who first came up with it.

GM

[GPM]

elerium wrote:

I just wanted to say thanks for your effort.......

I'm now thinking that two ported 15" subs.........

You're welcome! No big deal, I'd done the sim awhile back for someone else who'd had a 'change of heart' also, so just a bit frustrated. WRT complexity, I just view the divider boards that define its path-length as nothing more than the bracing required for a huge cab, though of course coming up with a folding scheme to fit a specific overall cab size can be tedious/time consuming compared to the simple two or three fold variety that's normally used by DIYers.

WRT ported and taking the minimal room gain available into consideration, a < 16 Hz tuning is desirable, so whatever driver within budget suitable for porting and has the highest Xmax and around a 20 Hz Fs should suffice, then use EQ to flatten them in-room. Again all this assumes you can't find a suitable sealed solution.

GM