I preferred the TG9FD-10 over most other small Peerless drivers I tried. I never worked with the TC9 though. Compared to a high quality source, I found small full range drivers slightly lacking in treble refinement, no matter how I equalized them. A few are still enjoyable, but a tweeter normally seemed like a useful addition. I never tried Alpair though.
I think your previous ideas about a dipole 2' from the wall are going to prove correct: you're going to need something to try to deal with the primary reflection. I haven't tried it myself, so can't say whether absorption, angled reflection, diffusion, or some combination would be most beneficial.
Yes, the reflections from the wall behind at 2' are a bit similar to the problems you get (internally) from a box speaker. My solution for both would be some amount of surface treatment with closed-cell/absorptive material (to soften the transition between air and surface). Thick, dense felt seems to be a good solution for box lining, but not exactly ideal for wall covering...
Open-cell solutions are more the realm of anechoic chambers, as they require so much space (they have to be as thick as a quarter wavelength to work, it seems).
Large objects behind the speakers are going to help disperse reflections, but, as ever, the lower frequencies will be the most problematic...
Open-cell solutions are more the realm of anechoic chambers, as they require so much space (they have to be as thick as a quarter wavelength to work, it seems).
Large objects behind the speakers are going to help disperse reflections, but, as ever, the lower frequencies will be the most problematic...
Last edited:
I was thinking about mentioning cardioid myself - possibly on the lines of the LX-mini, in which an omnidirectional radiator would be added to the existing dipole radiator (like radio direction-finding in reverse).
But I see that people like Genelec are offering cardioid speakers which seem to work using a somewhat different principle?
But I see that people like Genelec are offering cardioid speakers which seem to work using a somewhat different principle?
Nothing is out of the question; the Lxmini was touched upon earlier. Its standard plastic pipe form is generally not acceptable cosmetically, but if the performance and other criteria meet my requirements, I'd have to take a close look at making it a more attractive form.
I was introduced recently to a local resident who has a working pair of Lxmini in a unique & attractive champagne flute glass shape finished in multiple coats of a white automotive paint. CNC used for construction. Haven't had a chance to hear them; he's away at the moment.
Yesterday's experiments with the bass bin makes me dubious about OB only 2' from the front wall. An upward rear deflector has to be tried as it may help lengthen increase the distance of first reflections. Maybe today. But I'm not that hopeful.
The Lxmini calls for a minidsp 2x4 + 4 ch of amplification. If the Dayton DSP 4x 100w amp works as it is supposed to do, it would be a far cheaper alternative.
The Lxmini conveniently uses a Seas MU10 and a 6.5" woofer crossed at 700 Hz. That's relatively easy. For the bass tube, I could build a simple square cross section tube from MDF with large radius external corners, and put it on a stabilizing base. (Is the end of the tube vented or closed, btw?) I just got a 1.25" radius router quarter round over bit. Will have to beef up the corners to make it work.
But I want to hear & examine this local pair before I veer off in that direction.
I was introduced recently to a local resident who has a working pair of Lxmini in a unique & attractive champagne flute glass shape finished in multiple coats of a white automotive paint. CNC used for construction. Haven't had a chance to hear them; he's away at the moment.
Yesterday's experiments with the bass bin makes me dubious about OB only 2' from the front wall. An upward rear deflector has to be tried as it may help lengthen increase the distance of first reflections. Maybe today. But I'm not that hopeful.
The Lxmini calls for a minidsp 2x4 + 4 ch of amplification. If the Dayton DSP 4x 100w amp works as it is supposed to do, it would be a far cheaper alternative.
The Lxmini conveniently uses a Seas MU10 and a 6.5" woofer crossed at 700 Hz. That's relatively easy. For the bass tube, I could build a simple square cross section tube from MDF with large radius external corners, and put it on a stabilizing base. (Is the end of the tube vented or closed, btw?) I just got a 1.25" radius router quarter round over bit. Will have to beef up the corners to make it work.
But I want to hear & examine this local pair before I veer off in that direction.
Btw, it turns out someone at the Linkwitz Lxmini support forum has tried one of those 4x100w DSP amps. DSP functionality seems good enough. Only issue is load dependent FR -- a 1.35 dB rise at 20khz with 4 ohms load. I can live with that -- it's inaudible.
https://oplug-support.org/viewtopic.php?f=33&t=5528
https://oplug-support.org/viewtopic.php?f=33&t=5528
Front wall reflection shouldn't be a big problem with dipoles in stereo setup. Angulation tames reflected energy. A cardioid won't give any rear radiation, so it for sure will sound different, not so airy. I have taken measurements of the effect, but I couldn't find it now.
Juhazi -- I wondered when you'd chime in, as the sealed woofer + OB hybrid borrows from your now-iconic design. 😉
Thanks for the insight. The Lxmini, afaik, is omni<700 Hz, and partially OB above. The loosely stuffed tube to which the full range 10cm driver is attached has a 1-1.5" air gap. It's secured by several screws that go in through the plastic tube to the side of the magnet. So a lot of the rear radiation goes directly out starting in a ring, and some through the semi damped tube.
The Lxmini is highly regarded by many whose opin I respect. Afaik, it sounds best a bit out into the room but doesn't suffer as much when brought closer to the front wall. It's also a simpler 2-way and the overall parts cost is quite low. It's likely more forgiving of placement than the design I conjured up, though it won't be as authoritative in the bass. All quite irresistible, now that I've examined more of the details.
I'm itching to hear my neighbor's pair!
Thanks for the insight. The Lxmini, afaik, is omni<700 Hz, and partially OB above. The loosely stuffed tube to which the full range 10cm driver is attached has a 1-1.5" air gap. It's secured by several screws that go in through the plastic tube to the side of the magnet. So a lot of the rear radiation goes directly out starting in a ring, and some through the semi damped tube.
The Lxmini is highly regarded by many whose opin I respect. Afaik, it sounds best a bit out into the room but doesn't suffer as much when brought closer to the front wall. It's also a simpler 2-way and the overall parts cost is quite low. It's likely more forgiving of placement than the design I conjured up, though it won't be as authoritative in the bass. All quite irresistible, now that I've examined more of the details.
I'm itching to hear my neighbor's pair!
I've never heard this tried.
Will you try to audition the best available locations first with a sealed sub?
I don't use it electronically myself, but will you implement some sort of phase alignment too?
If you're replying to my question, I'm aware of other installs. My question addresses finding the places that best support bass in the listening position, regardless of where the back of the driver protrudes.
Hence, getting an idea of where to locate the infinite baffle sub or subs by experimenting with.various placements of a sealed.sub.
Hence, getting an idea of where to locate the infinite baffle sub or subs by experimenting with.various placements of a sealed.sub.
I finally heard my neighbor's LXminis last evening. He had them not ideally but reasonably set up with 2 sealed subs, all driven with Hypex 3-ch DSP plate modules. Pulled out 3' from the front wall and a fair distance from any side walls, they have a strong family resemblance to my LX521: spacious, clean rich & full. I istened to a few jazz tunes I know well. The transition between the 10cm SEAS full ranger and the 6" main driver is surprisingly seamless. I could not atribute anything untoward to it. The subs were well integrated in that they did not make their presence known, and the bass was good if not remarkable -- which it is in my space with the LX521s.
His LXminis are far better than the start I made with the sealed bass & OP top design. It is possible that I simply havent spent enough time working on it, but it's easier to work toward an end result that is clearly successful and exhibits the big open spacious sound I'm seeking.
So I'm changing direction.
I'm going to combine the SEAS 7" I started with in an LXmini style system with the best 10mm cone drive I have -- either the SEAS MU10 from the orginal LX521 or the Scanspeak 10F I'm using in my version of the LX521.
Why not the original mid/bass driver in the LXmini? It can only provide reasonable bass to ~55 Hz. IMO, that's not a full range speaker system. I want to to extend down to 40Hz. I know this is feasible in a similar footprint/volume with this 7" driver. There many have been other factors for Linkwitz's original 6" choice, but since the crossover point is a low 500 Hz, the 7" driver's dispersion will be no different in actual use. With -3 dB @40 Hz, >90% of music will be covered and make a sub unnecessary for most users at up to modestl;y loud levels (say 95 dB peaks in situ).
I also want to try the option of OB for that 10cm driver, vs LXmini's complex half OB/half aperiodic tube design. Why? Why not?! If the comparison isn't going to be ridiculously complex. The mounting scheme for the 10cm drive in the LXmini is quite fiddly & if it can be changed without a drop in performance, great!
Minidsp 2x4 hd will be employed for DSP xover.
The target usage model will be 1' from wall for casual listening & 3' out for serious listening. Yup, I'm giving up on the notion of a speaker that sounds best against the wall in any room -- too many doubts that it is feasible.
Hitting the <CA$1500 parts cost target should be no problem. <$1000 is probably not unrealistic.
If all the above works out, I'll dispense with the plastic pipe & build a 5- or 6-sided MDF tube that should allow for a nicer appearance.
His LXminis are far better than the start I made with the sealed bass & OP top design. It is possible that I simply havent spent enough time working on it, but it's easier to work toward an end result that is clearly successful and exhibits the big open spacious sound I'm seeking.
So I'm changing direction.
I'm going to combine the SEAS 7" I started with in an LXmini style system with the best 10mm cone drive I have -- either the SEAS MU10 from the orginal LX521 or the Scanspeak 10F I'm using in my version of the LX521.
Why not the original mid/bass driver in the LXmini? It can only provide reasonable bass to ~55 Hz. IMO, that's not a full range speaker system. I want to to extend down to 40Hz. I know this is feasible in a similar footprint/volume with this 7" driver. There many have been other factors for Linkwitz's original 6" choice, but since the crossover point is a low 500 Hz, the 7" driver's dispersion will be no different in actual use. With -3 dB @40 Hz, >90% of music will be covered and make a sub unnecessary for most users at up to modestl;y loud levels (say 95 dB peaks in situ).
I also want to try the option of OB for that 10cm driver, vs LXmini's complex half OB/half aperiodic tube design. Why? Why not?! If the comparison isn't going to be ridiculously complex. The mounting scheme for the 10cm drive in the LXmini is quite fiddly & if it can be changed without a drop in performance, great!
Minidsp 2x4 hd will be employed for DSP xover.
The target usage model will be 1' from wall for casual listening & 3' out for serious listening. Yup, I'm giving up on the notion of a speaker that sounds best against the wall in any room -- too many doubts that it is feasible.
Hitting the <CA$1500 parts cost target should be no problem. <$1000 is probably not unrealistic.
If all the above works out, I'll dispense with the plastic pipe & build a 5- or 6-sided MDF tube that should allow for a nicer appearance.
Last edited:
Unlike some of the truly meticulous posters here, I don't even have any drawings or a fully fleshed out plan, but I starting making some sawdust. Nothing like some action to make one feel progress is happening. 
I was wrong about the angle of the edges for 5 boards to form a regular pentagon profile tube. It's 36 degrees. I cut 10 pieces of a 1/2" MDF sheet on hand for a pair. With a sliding table saw, I got good enough accuracy to make it work. The pentagon rings from 18mm BB plywood for internal supports were hard to get just right; this took more time than expected, but such are my woodworking skills.
1/2 MDF might be a touch thin, but with 3 internal braces (attached via silicone instead of glue for constrained-layer damping) + double thickness top baffle & cap on the bottom of the 30" tube, it should be rigid enough. Each panel is 5.25" wide internally, and the internal ring braces will ensure the longest unsupported length will be only about 7". A 5x7" panel's first resonance should be pretty narrow Q & high in freq.
The estimated internal volume is 24 liters, which is way more than the 18L needed for the WINISD predicted 0.7 Q for my Seas L18RN-TE driver. F3 @50 Hz, which is higher than I want but there's DSP PEQ, Linkwitz transform. The extra volume will give me options for things like maybe mounting a small amp in the bottom of the enclosure & help the deep bass (F10 @27 Hz instead of 30 Hz) The pentagon profile ensures there are no parallel surfaces (except the bottom, which a tiny portion of the total internal panel area) & thus no standing waves.
One of the edges will be the "front", and the tweeter assembly will be mounted on the opposite panel.
I was wrong about the angle of the edges for 5 boards to form a regular pentagon profile tube. It's 36 degrees. I cut 10 pieces of a 1/2" MDF sheet on hand for a pair. With a sliding table saw, I got good enough accuracy to make it work. The pentagon rings from 18mm BB plywood for internal supports were hard to get just right; this took more time than expected, but such are my woodworking skills.
1/2 MDF might be a touch thin, but with 3 internal braces (attached via silicone instead of glue for constrained-layer damping) + double thickness top baffle & cap on the bottom of the 30" tube, it should be rigid enough. Each panel is 5.25" wide internally, and the internal ring braces will ensure the longest unsupported length will be only about 7". A 5x7" panel's first resonance should be pretty narrow Q & high in freq.
The estimated internal volume is 24 liters, which is way more than the 18L needed for the WINISD predicted 0.7 Q for my Seas L18RN-TE driver. F3 @50 Hz, which is higher than I want but there's DSP PEQ, Linkwitz transform. The extra volume will give me options for things like maybe mounting a small amp in the bottom of the enclosure & help the deep bass (F10 @27 Hz instead of 30 Hz) The pentagon profile ensures there are no parallel surfaces (except the bottom, which a tiny portion of the total internal panel area) & thus no standing waves.
One of the edges will be the "front", and the tweeter assembly will be mounted on the opposite panel.
