There's also the need for the crossover region to consist of an omnidirectional source to balance a dipole source in order to create a reasonably accurate cardioid source - so any significant spatial displacement from "optimum" could detract from this. Actually though, I really doubt a couple of inches will make much difference.
The Neo8 isn't as far over the woofer as the MU10 in the original, a distance which looks at least a third of the cone diameter. But it is at least up to the edge of the woofer, perhaps as as far as the surround. I changed the straight supports from an earlier iteration to achieve this. Putting it even more forward would result in greater rear physical impedance unless I changed this mounting system altogether.
But I have my doubts whether the position change would be audible, and it might change the directivity pattern of the woofer significantly to have a sail-like dividing panel in front of its cone.
Need to think about this one, especially the how of central placement over the woofer.
Yes. Better from visual pov, easier to implement.
Still trying to figure out how to adjust the planar driver's z axis position as juhazi referred to it.
Maybe... as an experimental platform, 2 threaded rods running front to back over the woofer, 2 nuts on each rod with bottom edge of NEO8 clamped between them. To do it without marking up or permanent modifying the tweeter, might need to rigidly attach a piece of aluminum plate as an extension to the bottom lip or something along that line. Gotta scrounge around in the workshop to see what I have to work with.
Finally got together with Brent, who brought his champagne flute LXminis + 2 Seas L26RO4Y subs. His subs each have a Hypex FA123 3-ch plate amp; each LXminis is connected to a sub/amp.
It was an interesting comparison. The deep bass of the champagne LXm was expectedly way better. But my pentagon tower bass is OK especially a bit higher up, and through the middle & highs, the Neo8 provide better detail & clarity. AND I think the more symmetrical dipole & undiminished rear radiation of the Neo8 gives it greater spaciousness -- closer to the LX521s.
No changes were made to the champagne LXm DSP settings. Brent cautioned that he'd toned down the treble because the room it resides in is quite lively and reflective. In contrast, the slanted ceiling of my 22x19x9' studio has 12" of fiberglass insulation behind panels normally used for drop-down ceiling, supposedly 50% reflective. And the intersection of the back wall and ceiling is a 19x3' tube trap made from 8" thick UltraTouch™ Denim Insulation. The balance is nice, but significantly more damped (and quieter) than your typical LR.
Brent's champagne LXm look gorgeous.
There was a question whether the shorter height of my pentagons affect the seated listener's perception. Did some quick vertical FR measurements at 24". The band 400~1000 Hz changes quite a lot moving from 0, to 15, 25 & 37 degrees higher (it's what I roughly measured), but at no point does it sound bad or change enough to bother me. Whether seated or standing from 8~10' away the perception is of a lively, big, enveloping & transparent presentation of the music. Example: The album Dave's True Story: Unauthorized recorded live minimally in a NYC church sounds brilliant.
One other thing I did this morning was to add a Dayton PTmini-6 planar tweeter in the inside edge of each Neo8, using double-sided tape. At first, I tried 2 + 1 mfd for a ~9kHz 1st order high pass for the PTmini-6, but then changed it to just 2mfd for 13kHz. No attenuation -- the Neo8 is rated for 92 dB/2.83v, and the PTmini-6 is 90 dB. It sounds balanced to me. The result is sparkle at the top which was missing with just the Neo8, easily audible with most recordings from anywhere in front of the speakers. Yes, the improved extension and dispersion is quite audible & welcome.
It was an interesting comparison. The deep bass of the champagne LXm was expectedly way better. But my pentagon tower bass is OK especially a bit higher up, and through the middle & highs, the Neo8 provide better detail & clarity. AND I think the more symmetrical dipole & undiminished rear radiation of the Neo8 gives it greater spaciousness -- closer to the LX521s.
No changes were made to the champagne LXm DSP settings. Brent cautioned that he'd toned down the treble because the room it resides in is quite lively and reflective. In contrast, the slanted ceiling of my 22x19x9' studio has 12" of fiberglass insulation behind panels normally used for drop-down ceiling, supposedly 50% reflective. And the intersection of the back wall and ceiling is a 19x3' tube trap made from 8" thick UltraTouch™ Denim Insulation. The balance is nice, but significantly more damped (and quieter) than your typical LR.
Brent's champagne LXm look gorgeous.
There was a question whether the shorter height of my pentagons affect the seated listener's perception. Did some quick vertical FR measurements at 24". The band 400~1000 Hz changes quite a lot moving from 0, to 15, 25 & 37 degrees higher (it's what I roughly measured), but at no point does it sound bad or change enough to bother me. Whether seated or standing from 8~10' away the perception is of a lively, big, enveloping & transparent presentation of the music. Example: The album Dave's True Story: Unauthorized recorded live minimally in a NYC church sounds brilliant.
One other thing I did this morning was to add a Dayton PTmini-6 planar tweeter in the inside edge of each Neo8, using double-sided tape. At first, I tried 2 + 1 mfd for a ~9kHz 1st order high pass for the PTmini-6, but then changed it to just 2mfd for 13kHz. No attenuation -- the Neo8 is rated for 92 dB/2.83v, and the PTmini-6 is 90 dB. It sounds balanced to me. The result is sparkle at the top which was missing with just the Neo8, easily audible with most recordings from anywhere in front of the speakers. Yes, the improved extension and dispersion is quite audible & welcome.
For what purpose? I know I prefer wider horizontal dispersion, the Neo8 is decent over about a 90 deg front horizontal arc (similar in reverse I presume) but much narrower vertically. Ptmini6 would have much narrower horizontal dispersion that way too. You have your neo3 sideways in your gradient-alike, iirc, but can't recall why...
Yes, thanks. It was purpose built. The 600sf workshop was here, 200' from the house when we moved in 7 years ago. I knew I didn't want to haul speakers back and forth to the house to audition, measure and fine tune, so added an extension on the back of the workshop. Walkthrough from workshop through rescued French doors, mostly keeping sawdust out. Reused some cabinets from kitchen remodel; built most of the interior, workbenches & storage stands. It's a decent size but needs better organization.
Woofer cavity resonance?? Pls explain.
Somehow, I doubt this is an audible issue. The woofer's motion is mostly moving perpendicular to that of the Neo8.
After a day of listening with the PTmini-6 as a super-tweeter, I have to say it sounds better thasn without. But I wonder if rolling the Neo 8 off before its steep on-axis peak @12kHz would help. This would mean lowering the HP frequency on the PTmini-6 to around 8~9kHz. The 6.5 ohm flat impedance curve of the Neo8 means a 0.11 mH coil would provide 1st order HP @9kHz. But not sure I have that small an inductor. And my Dayton DATS V3 is flaking out so can't measure inductance even I wind the coils myself.
-you can also shift the axis of the 8 where the rising ON-axis response is far enough OFF-axis (effective “toe-out”) to achieve the desired low-pass result at the listening position.
..and at that point you are getting much closer to the reference I built for myself almost 15 years ago.
..and at that point you are getting much closer to the reference I built for myself almost 15 years ago.
Last edited:
That’s a bummer, maybe start a thread on PE’s board to see if anyone else has had similar problems w/ solutions.
As the acoustic wavefront from the neo radiates outward, it will encounter the cavity created by the woofer, and this will cause a peak/dip boost cancellation that reflects the depth of the woofer cavity. This is normal with a tweeter on a baffle next to the woofer, and even more pronounced with this alignment as the driver output is on axis and not attenuated. If the Neo is directly over the woofer, the radiation is null at 90 degrees, and any cavity resonance front and back would be offsetting.
Ah, interesting detail. I would prefer not to try and center the Neo8 over the bass driver... but I guess I'll have to sort out some way to try it, if only to see/hear what happens.
Setting a planar tweeter (or mid) horizontally, one gets wider vertical pattern. This might help in getting better response at listening spot (or even MMM) with long gating 500ms.
In addition to spl response, you should check decay - it reflects how you perceive tonality. "Dull" sound typically has falling treble response in decay.
Here the effect of added rearside tweeter to a tweeter with a horn. Mid is Neo8. Interval settings make graphs look very different, you must try this with REW, both with CSD and Decay
In addition to spl response, you should check decay - it reflects how you perceive tonality. "Dull" sound typically has falling treble response in decay.
Here the effect of added rearside tweeter to a tweeter with a horn. Mid is Neo8. Interval settings make graphs look very different, you must try this with REW, both with CSD and Decay
This seems counterintuitive. I've read in countless sources that the dispersion of any driver is primarily determined by its diameter. In the case of a rectangular planar driver, the wider directivity would thus be along the samle line as the shortest dimension. The slim version of the GRS Neo8, for example, is narrower than the standard Neo8 and it has better horizontal directivity when used in its standard (tall) config.
You yourself use the Neo 8 planar driver vertically in your aino-gradient.
BUT I'm willing to test this. Going to devise some kind of central mounting setup for the Neo8 per drewmc observation above about the effect of the bass driver's cavity. So the drivers will be off and I'll try some off-axis FR sweeps on its side.
Hmm righto -- too early in the morning when I read that.wider VERTICAL
IMO, narrow vertical dispersion does not strike me as a significant issue. As mentioned, there's variation in response 400~1000Hz at various vertical angles, but at nominal distance -- say 7-10' from the speakers, the subjective difference between sitting and sitting isn't big enough to warrant taking action.
OK, got the DATS V3 running again -- mysterious Windows driver conflict maybe. Got am inductor 0.11 mH in series with each Neo8. The result is a 1st order LP for the Neo8 starting @9kHz. PTmini-6's 2mfd cap is a 1st order HP at 13kHz. Hole in the middle? No, because of the peak in Neo8.
This crossover is so simple to implement because of the flat impedance curve of both drivers, at ~6 ohms. The little planar is only asked to play the last octave of usable sounds (in music), and the overlap between them is broad.
The FR of this system is difficult to acertain. No single sweep in any mic position gives a representive curve; I think the only thing that comes close is averaging at least half a dozen sweeps at different horizontal -- and probably more at different vertical -- mic positions. It is too hot & lazy for that!
There is a sense of sharper focus in the sound, and more precise imaging. Sibilance doesn't linger as long, is less noticeable, less "thick", somehow. A 3-dB Q2 bump around 1100Hz helps fill the gap between the bass driver & the Neo8, and a minor boost at ~7kHz keeps that range a little more linear. The overall level level for the HF band in the 2x4 HD app was dropped by 1 dB to -6dB relative to the LF band. Less PEQ is used now, and the sound seems more refined.
Need to stop as my perceptions are getting too swayed by hope & desire -- and the heat. It's been 31-32C all afternoon.
PS -- didn't come up with any good way to reposition the neo3 over the bass driver today. Maybe tomorrow.
This crossover is so simple to implement because of the flat impedance curve of both drivers, at ~6 ohms. The little planar is only asked to play the last octave of usable sounds (in music), and the overlap between them is broad.
The FR of this system is difficult to acertain. No single sweep in any mic position gives a representive curve; I think the only thing that comes close is averaging at least half a dozen sweeps at different horizontal -- and probably more at different vertical -- mic positions. It is too hot & lazy for that!
There is a sense of sharper focus in the sound, and more precise imaging. Sibilance doesn't linger as long, is less noticeable, less "thick", somehow. A 3-dB Q2 bump around 1100Hz helps fill the gap between the bass driver & the Neo8, and a minor boost at ~7kHz keeps that range a little more linear. The overall level level for the HF band in the 2x4 HD app was dropped by 1 dB to -6dB relative to the LF band. Less PEQ is used now, and the sound seems more refined.
Need to stop as my perceptions are getting too swayed by hope & desire -- and the heat. It's been 31-32C all afternoon.
PS -- didn't come up with any good way to reposition the neo3 over the bass driver today. Maybe tomorrow.
Last edited: