If you scaled the depth by 1.53x you will get the 0.63 meter depth. This is an approximate simulaton of the resulting response. It may still be sufficiently flat for you - I don't know.
Hi xrk971
Are you talking about this. I love it. It is more than good enough. Just keep in mind that I will be building these with the Leak Sandwich 13" woofers, which will have slightly lesser specs than the 15" AE Speakers. I just picked those because they have a low Qt, which is what I expect the Leaks to have. I have REW installed and I now still need an MMC (what is that, BTW? is it a USB mic?) to measure these drivers.
If I calculate 1.53x, the external width is 651mm, and this is too wide. For 1.53x the depth will be 690mm, and that can still work. I remeasured that wall, and the depth is 600mm, not 550mm, so 690mm will work. My problems is that my max width is 600mm, and even that is pushing it.
As soon as I have the Leaks measured I will give you the measurements. Then we can really get the best results. I am really looking forward to this!
Deon
Hi Freddi
It is a nice idea, however I need to XO at around 120Hz, and that sub looks UGLY up that high. It does go nice and low, though.
Deon
Then with any of the BP alignments you're limited to a ~two octave pass-band = a ~30 Hz Fb.
GM
Not with a ~120 Hz XO, at least not once room modes are factored in if you want good mid-bass 'slam', etc. unless your room is really large to push all its fundamentals down below ~35-40 Hz.
GM
Wow, just checking in on this thread, and it's 7 pages already.
Actually, on the current point, I kind of agree with GM. You don't want to push a pipe/BP 'subwoofer' too low if crossing over as high as 120. The slot helps a bit to smooth response, but don't expect miracles. I think subjectively there will be a bit of a gutless sensation in the midbass where that null region is. IME, for a bass TQWT/TL that is needed for actual midbass, it is better to tune a bit higher and EQ the bottom end if needed.
The original K15 Karlson can take tons of EQ without the cone flopping around. Being a pipe/Karlson/BP hybrid, I suspect the 'Karlsonator' can't take as much, but won't need as much either.
FWIW, I went through many iterations of folding to get the one used for Karlsonator. No one has noticed, but I attempted to approximate a conical expansion. Deon's proposal is linear expansion, more like a BIB/Transflex hybrid. BTW - cannot take credit for all pipe/K hybrids as 'karlsonators'. Others have worked on them too, Moray James for example.
Actually, on the current point, I kind of agree with GM. You don't want to push a pipe/BP 'subwoofer' too low if crossing over as high as 120. The slot helps a bit to smooth response, but don't expect miracles. I think subjectively there will be a bit of a gutless sensation in the midbass where that null region is. IME, for a bass TQWT/TL that is needed for actual midbass, it is better to tune a bit higher and EQ the bottom end if needed.
The original K15 Karlson can take tons of EQ without the cone flopping around. Being a pipe/Karlson/BP hybrid, I suspect the 'Karlsonator' can't take as much, but won't need as much either.
FWIW, I went through many iterations of folding to get the one used for Karlsonator. No one has noticed, but I attempted to approximate a conical expansion. Deon's proposal is linear expansion, more like a BIB/Transflex hybrid. BTW - cannot take credit for all pipe/K hybrids as 'karlsonators'. Others have worked on them too, Moray James for example.
Why moot? Surely I am not the only nut out there that will want to build a pair of these, am I? I hope not so.
Oh, I don't mean the idea is moot, the idea is positively un-moot. It's just that the size and tuning is impossible to predict for now, until we know the drivers' specs.
Then again, the Lambda drivers are probably more easily obtained for many people than vintage Leak sandwich woofers...
Still, my suggestion is to err somewhat on the side of higher tuning if these are going to be crossed at 120Hz. The lowest octave is going to get a lift from room gain anyway.
Another idea: staggering the woofers in a zig zag pattern would help clump them closer together. This will help them share a more equal load, as well as reducing the minimum height requirements.
It's just that the size and tuning is impossible to predict for now, until we know the drivers' specs. Then again, the Lambda drivers are probably more easily obtained for many people than vintage Leak sandwich woofers...
Still, my suggestion is to err somewhat on the side of higher tuning if these are going to be crossed at 120Hz. The lowest octave is going to get a lift from room gain anyway.
Another idea: staggering the woofers in a zig zag pattern would help clump them closer together. This will help them share a more equal load, as well as reducing the minimum height requirements.
I think GregB means that until we see your Leak's TS params discussing the box geometry and such is moot. But if someone likes the performance predicted and will get the AE TD15M's then it is good as is.
GregB has a point about pushing the tuning higher up if you will use this indoors. The 20 Hz comes for free if you tune to hit 30 Hz - and my idea is why stop at using these as subs? I think they will go all the way up to 1 kHz (at least with the TD15M). Get a good CD and horn mounted on top and have a cool 2 way. Shrinking the cabinet to circa 48 in tall should hit 30 Hz and make the horn on top not too high above ear level.
GregB has a point about pushing the tuning higher up if you will use this indoors. The 20 Hz comes for free if you tune to hit 30 Hz - and my idea is why stop at using these as subs? I think they will go all the way up to 1 kHz (at least with the TD15M). Get a good CD and horn mounted on top and have a cool 2 way. Shrinking the cabinet to circa 48 in tall should hit 30 Hz and make the horn on top not too high above ear level.
Last edited:
I have given thought to redesigning my front speakers so that they can get down to circa 80Hz - 70Hz. This will require an extra woofer on the fronts per side and a slightly more complex XO, but can be done. I will give more pics and illustrations tomorrow to illuminate my ideas, original and reworked. Then the MOAK subs can be repurposed for low bass. Until tomorrow then.
Deon
Here is what I as thinking of in terms of the front speakers. My original idea was for a Goodmans Axiom 201 + 6"x9" alnico fullrange on an open baffle, supported by a ribbon tweeter for all frequencies above 5kHz and baffle-step on the Goodmans. It would have been on a small baffle supported by a three-legged metal frame. The XO for this design would have been 120Hz or higher to keep power-handling high. It would have looked something like this:
An externally hosted image should be here but it was not working when we last tested it.
I have rethought my design and decided to add a 12" alnico woofer below the Goodmans in order to support the 6"x9" fullrange driver. The XO between the 6"x9" and the alnico woofer will be at bafflestep. The high-pass can then be lowered to 70Hz to 80Hz. Here are sketches for my ideas for this design.
The first one is the standard layout, with the 6"x9"-tweeter array on top, then the Goodmans and then the 12" alnico woofer. In the second design the Goodmans has been moved to sit above the 6"x9"-tweeter array, with the 12" alnico woofer sitting directly below the 6"x9"-tweeter array. Here are those sketches:
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
These ideas are great, but another reason I wanted the MOAK subs to go as high as 120Hz is to get that wonderful Karlson punch in the 100Hz region. So I decided to modify the two previous designs. I will do this by loading the 12" alnico woofer into a Karlson slot, similar to what Magnetar used to do. The overall speaker will still be open-baffle. I am still trying to decide if I just put a slot in front of the driver, or put slots in front and behind the driver (to equalize the load). The next two sketches are the same as the previous two, but with a slot added to the front of each speaker:
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
My preference at the moment is for the slot-loaded design with the Goodmans driver at the top. This should make for a more coherent speaker. Comments and suggestions are welcomed.
Thanks,
Deon
nice artwork - it might be useful to mock up the Karlson section from scraps, tape, cardboard, or foamboard to get an idea of how it will sound and perform.
there is one patent that doesn't reference Karlson that luses a double driver variant of the klam - one internal driver firing downwards, the other a fullrange direct radiator firing forwards.
Magnetar IIRC, used a 15" PHL woofer in "Rosie the klam" - and probably an Altec 421. Sero had a page of a clone of that system.
here's that patent
foirixi's Lowe's sound body used just the back wave
http://www.diyaudio.com/forums/full-range/212593-supreme-sound-body-s.html
large picture http://www.diyaudio.com/forums/atta...-supreme-sound-body-s-afbeeldingrosie-204.jpg
there is one patent that doesn't reference Karlson that luses a double driver variant of the klam - one internal driver firing downwards, the other a fullrange direct radiator firing forwards.
Magnetar IIRC, used a 15" PHL woofer in "Rosie the klam" - and probably an Altec 421. Sero had a page of a clone of that system.
here's that patent
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
foirixi's Lowe's sound body used just the back wave
http://www.diyaudio.com/forums/full-range/212593-supreme-sound-body-s.html
large picture http://www.diyaudio.com/forums/atta...-supreme-sound-body-s-afbeeldingrosie-204.jpg
Deon,
You have been busy! Nice drawings btw. I think a lot of your ideas can work well and I hope you are going DSP and quad or penta amping because designing and testing passive XO's for all of these would be a royal pain.
+1 on vertical spatial and time alignment. You will probably get a lot more constructive comments from experts in the multiway forum on this design. I like your last design with Karlson slot loaded woofer on the bottom. However, I have to wonder if you are making this unnecessarily complex. I'm lazy and like to keep things simple to the extent possible. It might be worth it to try having your original scheme but just keep the 6x9 and ribbon and XO at baffle step circa 1.2 kHz and let the MOAK handle below that. You will have a 3 way system. Maybe you are afraid the Leaks won't sound good in the vocal range from 100 Hz on up to 1.2 kHz? But certainly it might be worth a try and all you have to do is add your Goodmans next, and if still needed add the woofer below that. Take it a step at a time and make the system modular with two rails that let you add additional boards to the OB with each driver you want to add.
Also, offset horizontally the ribbon tweeter from centerline to reduce diffraction and don't make the baffle too wide as it ruins imaging.
I have a thread in the Fullrange forum called Cheap and FAST OB where Buildmesomething has done a great study on effect of baffle width on imaging.
Great ideas here though and I am really looking forward to seeing the build start.
Cheers,
X
You have been busy! Nice drawings btw. I think a lot of your ideas can work well and I hope you are going DSP and quad or penta amping because designing and testing passive XO's for all of these would be a royal pain.
+1 on vertical spatial and time alignment. You will probably get a lot more constructive comments from experts in the multiway forum on this design. I like your last design with Karlson slot loaded woofer on the bottom. However, I have to wonder if you are making this unnecessarily complex. I'm lazy and like to keep things simple to the extent possible. It might be worth it to try having your original scheme but just keep the 6x9 and ribbon and XO at baffle step circa 1.2 kHz and let the MOAK handle below that. You will have a 3 way system. Maybe you are afraid the Leaks won't sound good in the vocal range from 100 Hz on up to 1.2 kHz? But certainly it might be worth a try and all you have to do is add your Goodmans next, and if still needed add the woofer below that. Take it a step at a time and make the system modular with two rails that let you add additional boards to the OB with each driver you want to add.
Also, offset horizontally the ribbon tweeter from centerline to reduce diffraction and don't make the baffle too wide as it ruins imaging.
I have a thread in the Fullrange forum called Cheap and FAST OB where Buildmesomething has done a great study on effect of baffle width on imaging.
Great ideas here though and I am really looking forward to seeing the build start.
Cheers,
X
Thanks. Appreciated.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Not even close. I have asked Jim Nicholls of JWN Amplifiers to build me a nice EL84 PP amp. This is the amp that I will use for the main speakers. Problem is my salary is very low, so I am paying it off at $100 per month. I will however be using a valve-based active XO for the high-pass at 80Hz. The subs will have their own active XO, possible something like a Mini DSP to help tune them to my room, with a stereo digital amp for each sub.
As far as the XO for the main speakers is concerned, it is not going to be massive as I simply don't have the skills to do that. My thinking was heavily influenced by Lukasz Fikus' designs as well as some comments by Lynn Olson. Here is what I am thinking ITO the XO and overall design:
The 12" alnico will only have a 1st order low-pass at baffle-step. It will be about the same efficiency as the 6"x9", so this will not be for baffle-step comp, but I just chose it as that will be low enough to get out of the vocal range. This XO point can still change.
The Goodmans Axiom will only have baffle-step compensation and a 1st-order low-pass at 5kHz.
The 6"x9" will have the 1st-order high-pass at baffle-step and a 1st-order low-pass at 5kHz.
The ribbon tweeter will have a 3rd-order or 4th-order high-pass at 5kHz.
The idea here is partially based on Lukasz' P24 design. That is the reason for the ribbon being next to the 6"x9". Also in this way it will be close to both the 6"x9" and the Goodmans. And as Lukasz said the 6"x9" and the ribbon tweeter will also sort of try to mimick a point-source.
The reason for the Goodmans + 6"x9" is so that I get the tonal mix of a ceramic magnet and a Alnico magnet. That is also why the mix of ceramic and Alnico in the sub. Lynn Olson said that the mix of the grit of ceramic magnet and the lush tonal colors of Alnico should mix to make for excellent variety of tonal colors.
That would be nice, but would it be practical.
Complex, maybe, but I would like that punch in the 100Hz region. That is the main region for bass punch, and I want that Karlson kick there.
The problem is that the Leak woofers have a very nasty break-up somewhere between 1kHz and 2kHz. The other problem is that the woofers will not be in line-of-sight with respect to the listening position, so the wave would have to fold around the desk. Not so good for anything but very low frequencies. That is why I redesigned the main speakers to get them to go low enough to make that less of a problem.
The tweeter will be offset by the fact of it being next to the 6"x9", so this is a given. I will also try to keep the baffle as narrow as possible. I was thinking about the comments about time-alignment and a narrow baffle, so I can up with this idea:
An externally hosted image should be here but it was not working when we last tested it.
The Goodmans (the driver on top) will be mounted on a wooden ring, which will allow a measure of time-alignment and aid diffraction. Time-aligning the 6"x9" and the tweeter will be difficult, so I will let that one go. The bottom woofer will then still be in the Karlson slot (or slots, if I add another one in the rear like Magnetar's "Rosie the clam").
Well, that's all from me for now. Please add any more comments. I do read them all and appreciate every single one.
Thanks,
Deon
Deon,
Ok, I see your approach now - mostly 1st order passive XO's except for the valve active XO at 80 Hz. Are you really stuck on valve amps because you did mention miniDSP and class D amp for the subs? Anyhow, if you went with a single $80 minDSP board and a couple of class D amps per side (left or right) you can build a nice 4 way DSP XO and quad-amp system for $130 (at 50 watts/ch). Or $260 for both sides at 4 channels each. This lets you play with XO freq location and slope (and type LR or BW) on the fly and lets fine tuning and tweaking it very easy. Also, time alignment up to 7.5ms is also easy with this setup. Here is what I did for a 2x4 XO. You would basically get two of these. And if you need more than 50 watts/ch that is easy with an upgrade to more powerful class D amp. I am using the TPA3116D2 amp - which sounds very good. I have measured them putting out 75 watts into 4 ohms. Distortion is less than 0.1% at 40 watts and many people have said it sounds better than their tube amps. All for $15 to $20 a board.
I like your latest design with the narrow neck for the 6x9 and ribbon. The bass unit with the K-slot at the bottom may need rearranging: if you look a Klam, the driver is mounted in the deeper portion and the wide part of the K-aperture is next to the tapered narrow portion of the rear chamber. So you may have to build your woofer portion with more depth at the top tapering to zeeo depth at bottom. If you really want punch in th 80 Hz to 500 Hz range, a K15 enclosure is the ideal way to do this as it is much more efficient than an OB with a K-slot in front.
Ok, I see your approach now - mostly 1st order passive XO's except for the valve active XO at 80 Hz. Are you really stuck on valve amps because you did mention miniDSP and class D amp for the subs? Anyhow, if you went with a single $80 minDSP board and a couple of class D amps per side (left or right) you can build a nice 4 way DSP XO and quad-amp system for $130 (at 50 watts/ch). Or $260 for both sides at 4 channels each. This lets you play with XO freq location and slope (and type LR or BW) on the fly and lets fine tuning and tweaking it very easy. Also, time alignment up to 7.5ms is also easy with this setup. Here is what I did for a 2x4 XO. You would basically get two of these. And if you need more than 50 watts/ch that is easy with an upgrade to more powerful class D amp. I am using the TPA3116D2 amp - which sounds very good. I have measured them putting out 75 watts into 4 ohms. Distortion is less than 0.1% at 40 watts and many people have said it sounds better than their tube amps. All for $15 to $20 a board.
I like your latest design with the narrow neck for the 6x9 and ribbon. The bass unit with the K-slot at the bottom may need rearranging: if you look a Klam, the driver is mounted in the deeper portion and the wide part of the K-aperture is next to the tapered narrow portion of the rear chamber. So you may have to build your woofer portion with more depth at the top tapering to zeeo depth at bottom. If you really want punch in th 80 Hz to 500 Hz range, a K15 enclosure is the ideal way to do this as it is much more efficient than an OB with a K-slot in front.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.