I found a post from @planet10 for an old B&W design:
Would something like this be viable? Or maybe something 3d printed similar to this? If I went with the cuboid design, I think I would have a couple of inches or so to play with around the two woofers. It may be possible to accommodate something like that. Then again, I think about the law of diminishing returns and wonder how much of an effect it would actually have. I also still need to figure out the ACTUAL size of this. I would imagine that while force canceling is a neat technique, I'm sure physics has something to say about it not being perfect.
Another idea I had was maybe just using the stand as a port. I figure if it's going to be part of the speaker I may as well put it to work. I've considered putting the electronics in it too, but i would imagine there would be added delay with longer wires.
Would something like this be viable? Or maybe something 3d printed similar to this? If I went with the cuboid design, I think I would have a couple of inches or so to play with around the two woofers. It may be possible to accommodate something like that. Then again, I think about the law of diminishing returns and wonder how much of an effect it would actually have. I also still need to figure out the ACTUAL size of this. I would imagine that while force canceling is a neat technique, I'm sure physics has something to say about it not being perfect.
Another idea I had was maybe just using the stand as a port. I figure if it's going to be part of the speaker I may as well put it to work. I've considered putting the electronics in it too, but i would imagine there would be added delay with longer wires.
Last edited:
I just didn't know if it was any better or not. Oh well.
Next thing to be overly anal retentive about; diffraction! The Harry Olson study states that a cylinder is not a great shape for flat response, but others say it's good for mitigating baffle step loss. A little confused on that one...
DM302.
This is an example of sufficient differences in cabinet depth to actually make a difference. The depth would spread the reflections of the back would be spred over almost an octave. B&W would have had to expend a whack of cash for the injection molds to make the back.
With 3D print tech it would be fairly simple to emulate. But as GM suggests there are ways to achieve much the same result using different tricks.
Something like this for instance. Achhievesthe almost 2 octaves of relection spreading, gives room for damping which will remove some reflections creates a short closed TL, and can act as a driver brace. Much more efficent with gross box volume.
dave
Sweet! Well that gives me a little design wiggle room then.
I REALLY like this. Nice and simple, the center brace would probably help support the dual subs, and it would allow me to use a simpler shape. So what about baffle stepping? That's kind of the big issue I'm trying to solve. Losing 6db is a lot of power when you're starting to talk hundreds of watts. Also, with that design, would I be able to chamber it if I extend the front and back a little bit? One of the ideas I had was to chamber off the MTM section so that the bass doesn't start to come out where it shouldn't.Something like this for instance. Achhievesthe almost 2 octaves of relection spreading, gives room for damping which will remove some reflections creates a short closed TL, and can act as a driver brace. Much more efficent with gross box volume.
In room loss is more like 3-4 dB, never 6 unless the room is huge and the speakers are WAY out from the walls.
Any EQ to level the on-axis also rasies the room response (not good).
Of the hundreds of loudspeakers we have built only 1 had an explicit BSC correction, and that was cured by replacing the $10 drivers with SEAS FA22.
dave
Any EQ to level the on-axis also rasies the room response (not good).
Of the hundreds of loudspeakers we have built only 1 had an explicit BSC correction, and that was cured by replacing the $10 drivers with SEAS FA22.
dave
And here I thought I would need to make some exotic shape. For the subs I'm going with the SEAS L22ROY2, so that should give me my low end extension. I was just worried it would cause issues using a conventional shape like a box. Which means simply tapering the sides should give me a much flatter response curve. http://usenclosure.com/Olsen/olson_direct-radiator-loudspeaker-enclosures.pdf Shape J looks to be the best compromise of the bunch so I was thinking of going with that for the front baffle. So say the subs were at center. Would I build that triangle shape twice (one for each side, forward and aft)? Or would that one center bracket do everything I need it to do?
Wider box pushes the BSC down, and with a wall not too far effectively allows the wall behind to act to restrict the radiation to 2∏. BS BTW is th etransition from 2∏ to 4∏ Steriradian radiation.
Shape is much more important when it comes to how the transition happens. Add that to any edge diffraction higher up in frequency and you have a loudspeakers diffraction signature.
This research was published in the 50s, more has been done.
Why oi typically put out easy-to-build,harder, and even harder to build versions. As they get closer to a teardrop shape the diffraction signature decreases and the boxes have a greater tendency to disappear.
dave
Shape is much more important when it comes to how the transition happens. Add that to any edge diffraction higher up in frequency and you have a loudspeakers diffraction signature.
This research was published in the 50s, more has been done.
Why oi typically put out easy-to-build,harder, and even harder to build versions. As they get closer to a teardrop shape the diffraction signature decreases and the boxes have a greater tendency to disappear.
dave
Would you happen to know a good updated version of this I could read on?
Well... I did have one idea, but it's a weird shape I don't think I could make on my own. I was thinking something along the lines of the devialet phantom. Except instead of the electronics going in the back portion, that would be somewhat flattened out into a baffle for the MTM. The thing I would have to consider then is what I would use as a stand... as well as how to make the damn thing lol
Last edited by a moderator:
Yes, as they smooth out the driver/baffle transition, reduces amplitude of the driver/baffle eigenmodes, ditto the cylinder's theoretically infinite eigenmodes and if you round over the baffle edges enough and suspend the speaker in sufficient space you wipe them all out (spherical). 😉
https://en.wikipedia.org/wiki/Harry_F._Olson
Acoustical Engineering (New York: D. Van Nostrand Company, 1957). Dynamical Analogies (New York: D. Van Nostrand Company, 1943).
More modern research is scattered about, B&W did a bunch. There is a long thread on the subject (at least one),
dave
Last edited:
I've read a few of theirs. A bunch of KEFs. Their literature on the LS60 and the KC62 is what convinced me to go with force canceling woofers for bass frequencies. The LS60 one also has some stuff on diffraction.
If you want to get serious, study land speed record (LSR) vehicles that theoretically can/have broken the sound barrier, ditto supersonic aircraft.
I mean... to be fair, I just came back from DC a little bit ago and visited the Udvar Hazy and saw the SR-71. I would say that had more to do with two giant engines than just aerodynamic design 😛
I would say on vehicles though you have to take things like draft, drag, and lift into account, but I get what you're saying. Sound is still a wave and therefore interacts with different objects in different ways depending on geometry.
Sorry for the radio silence. Life has a funny way of derailing everything, but I had a free moment to spare so started working on this again.
Here's the revised design concept (disclaimer this was in powerpoint because I didn't feel like putting it in CAD yet so here's a quick and dirty):
dark blue = enclosure
light yellow = purifi 4" drivers
purple = raal tweeter
light blue = x2 seas 8" subs
dark orange = hypex amp
Went back to a simple cuboid enclosure. Although @planet10 's enclosure shape would get me closer to a tear drop shape, the cancellation of opposing forces from the subs wouldn't be symmetrical and would defeat the purpose of dual opposing subs. The red shapes are dividers in order to split it into chambers. This separates the electronics and the mid/high freq sections from the woofers just like the original design. The dark yellow is a center support brace. On the back of the subs are a small metal protrusion (only thing i can really think to call it) as opposed to just a flat surface. The center brace will allow me more surface area to connect the woofers with, as well as providing more support for the enclosure. Still trying to decide on which CLD product to use to line the case with (still need more data), but for the dead space, was probably thinking rebond foam. Only point of contention there is if it will attenuate the higher frequencies or not. If not, will probably just use some shredded denim fill. As always open to thoughts and suggestions.
Here's the revised design concept (disclaimer this was in powerpoint because I didn't feel like putting it in CAD yet so here's a quick and dirty):
dark blue = enclosure
light yellow = purifi 4" drivers
purple = raal tweeter
light blue = x2 seas 8" subs
dark orange = hypex amp
Went back to a simple cuboid enclosure. Although @planet10 's enclosure shape would get me closer to a tear drop shape, the cancellation of opposing forces from the subs wouldn't be symmetrical and would defeat the purpose of dual opposing subs. The red shapes are dividers in order to split it into chambers. This separates the electronics and the mid/high freq sections from the woofers just like the original design. The dark yellow is a center support brace. On the back of the subs are a small metal protrusion (only thing i can really think to call it) as opposed to just a flat surface. The center brace will allow me more surface area to connect the woofers with, as well as providing more support for the enclosure. Still trying to decide on which CLD product to use to line the case with (still need more data), but for the dead space, was probably thinking rebond foam. Only point of contention there is if it will attenuate the higher frequencies or not. If not, will probably just use some shredded denim fill. As always open to thoughts and suggestions.
Last edited by a moderator:
Found a moment for a quick update. Not much, but I'm starting to play around in CAD a bit.
It looks like there isn't a whole lot of clearance so I may need to bring this back to the drawing board again. Additionally, I've looked around and found a few works stating how the RAAL 70-20 works a lot better with some sort of horn. If that's the case, while it wouldn't be too bad to recess the driver to make the horn, it would force me to pick different drivers for the side firing LF drivers because of how far I would have to pull it in.
Another idea is I could revisit the "electronics in the stand" idea and just take the DSP out of the equation for the enclosure. That would probably allow me to use my current pick for the LF drivers, but would probably have to look at using a different layout for the internals.
Also, shout out to Seas. They didn't have the step file for the L22ROY2 on their website, so I shot them an email and asked them if they would post one. Looked at my email this morning and found a CAD file waiting for me lol
It looks like there isn't a whole lot of clearance so I may need to bring this back to the drawing board again. Additionally, I've looked around and found a few works stating how the RAAL 70-20 works a lot better with some sort of horn. If that's the case, while it wouldn't be too bad to recess the driver to make the horn, it would force me to pick different drivers for the side firing LF drivers because of how far I would have to pull it in.
Another idea is I could revisit the "electronics in the stand" idea and just take the DSP out of the equation for the enclosure. That would probably allow me to use my current pick for the LF drivers, but would probably have to look at using a different layout for the internals.
Also, shout out to Seas. They didn't have the step file for the L22ROY2 on their website, so I shot them an email and asked them if they would post one. Looked at my email this morning and found a CAD file waiting for me lol
So a few things:
1. It is with a heavy heart, but I think I'm going to switch out the tweeter. I notice I'm having to jump through too many hurdles in order to accommodate the RAAL. I remember a while ago a post from @xrk971 ( https://www.diyaudio.com/community/threads/raal-70-20xr-and-ptt6-5-compact-tl.356906/ ) a lot of the issues. Directivity especially since making a horn on a small(ish) speaker kind of defeats the purpose of it being small. Another big issue is simply its size. It would be too difficult to create a separate MTM compartment within the enclosure using the 70-20. All this considered, I will most likely be switching to the Bliesma T25B tweeter for its size and performance characteristics.
As you can see here, it's far easier to design with the Bliesma than it is with the RAAL. It's a bummer, but that's kind of what engineering is about: finding optimal solutions to a given problem. In this case, the Bliesma is simply the better option.
2. The enclosure design I would still like to have very few diffraction points. I'm still scribbling away trying to figure out an overall design for it. One thought that crossed my mind is swapping out the L22ROY2s with a set of Purifi 8" woofers. This would allow the enclosure to overall have a smaller cross-section, but I'm not quite sure on the swap. The Seas woofers are made specifically for LF work, where as the Purifis can perform up to much higher frequencies.
3. Since I don't have all the tools available to me like I used to, I'm still trying to figure out how I'd like to proceed on assembly. I've had a few ideas on how to approach this, but all of them still require me to have tools on hand. Maybe I should just suck it up and start from square one again and get some basics like hand tools, clamps, etc. Another idea I had was going the 3d printing route and do it in parts, but I'm not so sure on that one for a lot of reasons. Again, still just spitballing on this, but first thing's first I need to finish the enclosure design in order to figure out how I want to make it.
1. It is with a heavy heart, but I think I'm going to switch out the tweeter. I notice I'm having to jump through too many hurdles in order to accommodate the RAAL. I remember a while ago a post from @xrk971 ( https://www.diyaudio.com/community/threads/raal-70-20xr-and-ptt6-5-compact-tl.356906/ ) a lot of the issues. Directivity especially since making a horn on a small(ish) speaker kind of defeats the purpose of it being small. Another big issue is simply its size. It would be too difficult to create a separate MTM compartment within the enclosure using the 70-20. All this considered, I will most likely be switching to the Bliesma T25B tweeter for its size and performance characteristics.
As you can see here, it's far easier to design with the Bliesma than it is with the RAAL. It's a bummer, but that's kind of what engineering is about: finding optimal solutions to a given problem. In this case, the Bliesma is simply the better option.
2. The enclosure design I would still like to have very few diffraction points. I'm still scribbling away trying to figure out an overall design for it. One thought that crossed my mind is swapping out the L22ROY2s with a set of Purifi 8" woofers. This would allow the enclosure to overall have a smaller cross-section, but I'm not quite sure on the swap. The Seas woofers are made specifically for LF work, where as the Purifis can perform up to much higher frequencies.
3. Since I don't have all the tools available to me like I used to, I'm still trying to figure out how I'd like to proceed on assembly. I've had a few ideas on how to approach this, but all of them still require me to have tools on hand. Maybe I should just suck it up and start from square one again and get some basics like hand tools, clamps, etc. Another idea I had was going the 3d printing route and do it in parts, but I'm not so sure on that one for a lot of reasons. Again, still just spitballing on this, but first thing's first I need to finish the enclosure design in order to figure out how I want to make it.