for whoever is interested...
So far I am not too impressed with the curved panel. YES in side to side head moves treble and tone are maintained much better than the flat panel, However to my surprise it is quite difficult to position these to get them playing at their best AND even quite small side to side head movements alter image specificity. That was a surprise. I expected these issues to be better with the curve. The image btw was uncomfortable and restless. I found myself repositioning my head a lot. As well they often presented a one-dimensional wall of sound with little depth. It got a tad better when I got overall tone correct which btw was a linear falling response or about -5db at 10kHz, BUT the basic attributes were all still there.
It was a pleasant sound BUT as said above it seems to have issues and just wasn’t special. This surprised me. I played with radius changes but still the basic attributes ruled.
Im thinking at this point that I must have done something very wrong here. After all Martin Logan uses this method so it must be great right? So I went on a dive into reading any ML review I could find and to my surprise I found a few reviews (the more honest ones imo) that say exactly the same thing about the curved panel sound as I am saying here. Exactly the same.
I wasn’t expecting that and im sure the ML fan club will have a field day with this ha.
Not ready to give up on the curve just yet. I have some unorthodox ideas to try yet and will explore some more but again , so far, the flat panel wins hands down but damn that beaming!!!
So far I am not too impressed with the curved panel. YES in side to side head moves treble and tone are maintained much better than the flat panel, However to my surprise it is quite difficult to position these to get them playing at their best AND even quite small side to side head movements alter image specificity. That was a surprise. I expected these issues to be better with the curve. The image btw was uncomfortable and restless. I found myself repositioning my head a lot. As well they often presented a one-dimensional wall of sound with little depth. It got a tad better when I got overall tone correct which btw was a linear falling response or about -5db at 10kHz, BUT the basic attributes were all still there.
It was a pleasant sound BUT as said above it seems to have issues and just wasn’t special. This surprised me. I played with radius changes but still the basic attributes ruled.
Im thinking at this point that I must have done something very wrong here. After all Martin Logan uses this method so it must be great right? So I went on a dive into reading any ML review I could find and to my surprise I found a few reviews (the more honest ones imo) that say exactly the same thing about the curved panel sound as I am saying here. Exactly the same.
I wasn’t expecting that and im sure the ML fan club will have a field day with this ha.
Not ready to give up on the curve just yet. I have some unorthodox ideas to try yet and will explore some more but again , so far, the flat panel wins hands down but damn that beaming!!!
a little less curve , no baffle, and some trap work. Then another with a rake to get what sounds more correct tone
Attachments
Last edited:
they look great
yep to my ears open panel bass is great with enough area. Ive never heard a box sound as real down there.
yep to my ears open panel bass is great with enough area. Ive never heard a box sound as real down there.
So just a little update
I increased the radius a fare amount. Idea was to get closer to flat panel performance BUT have at least enough spread so that my head didnt have to be so tight in the vise to stay in the sweeet spot Ha.
An interesting point, with the larger radius curve on the diaphragm the "hump" in raw response is now centered around 3khz AND is quite symmetrical in shape. This means when I install the trap circuit to flatten response I can now play with the "glare" I often hear with panels by simply adjusting the traps resistor value. Now I can put a bit of a saddle in the response between about 2 and 4 Khz. Just starting with this so not sure what long term results will be but preliminary test seem promising.
Also noticing with all these different radiuses that the best FR as always about 20 degrees OFF axis.... Anyone know why that would be? Do the M Logans do something similar?
I increased the radius a fare amount. Idea was to get closer to flat panel performance BUT have at least enough spread so that my head didnt have to be so tight in the vise to stay in the sweeet spot Ha.
An interesting point, with the larger radius curve on the diaphragm the "hump" in raw response is now centered around 3khz AND is quite symmetrical in shape. This means when I install the trap circuit to flatten response I can now play with the "glare" I often hear with panels by simply adjusting the traps resistor value. Now I can put a bit of a saddle in the response between about 2 and 4 Khz. Just starting with this so not sure what long term results will be but preliminary test seem promising.
Also noticing with all these different radiuses that the best FR as always about 20 degrees OFF axis.... Anyone know why that would be? Do the M Logans do something similar?
I'm also very pleased with how the bass sounds. I did some tests, it reproduces clearly and very well down to 25 Hz.
May I suggest several tests: in-ear audibility of >10khz very high frequency (using either tone generator or a smartphone's built-in hearing-test/PEQ-generator); tilting the panels (or even the head/ears); play one channel only (look straight ahead not at speaker), is violin soloist imaged at a distance, does a tone-sweep move around?
To aid diagnosis of phase alignment maybe get a small fullrange-driver with good HF extension for direct comparison/reference (like SB65WBAC). Play a HF tone through both (one of each) and adjust offset until max combined volume (in phase); then change tone frequency -- does the "best" offset still hold (stay fixed)?
I think, somehow the panel curvature created/exacerbated comb-filtering and phase misalignment both. Are you able to try flipping the panel to curve the other way?
Last edited:
thanks wchang. may get to some of those tests eventually
At the moment the larger radiused version ( flatter profile) is performing better than the smaller radiused versions. Image is now more stable while keeping a reasonable listening window.
However they are still quite fussy of exact toe in. Also the 2-4 khz saddle adjustment of the trap circuit has proven to be important not only to overall tone but image depth as well. I have experienced this with ALL speakers Ive had but this one seems more critical of it.
Another thought is the number and placement of the foam strips that hold the diaphragm off the magnets. these create a small saddle curve in diaphragm between them . A deviation from a true curve surface that I have to wonder what its doing to radiation pattern. Not only is there a saddle shape formed on diaphragm between the strips, but where the diaphragm arcs down into the saddle it will be a tad closer to the magnets in the center. So ya end up with a curve AND a slight increase in drive force in the center between the strips. The larger radiused version ( flatter profile) is easyer to diaphragm and has less of this saddle effect.
As for the number of strips, I was using 3 which placed a termination right in the center. Not sure if that may be causing some phasy issue? Now using only 2strips so center of driver is open.
The diaphragm has a bit less control of low frequency movement with the larger cells and flatter profile and can no longer use a 6 db filter
At the moment the larger radiused version ( flatter profile) is performing better than the smaller radiused versions. Image is now more stable while keeping a reasonable listening window.
However they are still quite fussy of exact toe in. Also the 2-4 khz saddle adjustment of the trap circuit has proven to be important not only to overall tone but image depth as well. I have experienced this with ALL speakers Ive had but this one seems more critical of it.
Another thought is the number and placement of the foam strips that hold the diaphragm off the magnets. these create a small saddle curve in diaphragm between them . A deviation from a true curve surface that I have to wonder what its doing to radiation pattern. Not only is there a saddle shape formed on diaphragm between the strips, but where the diaphragm arcs down into the saddle it will be a tad closer to the magnets in the center. So ya end up with a curve AND a slight increase in drive force in the center between the strips. The larger radiused version ( flatter profile) is easyer to diaphragm and has less of this saddle effect.
As for the number of strips, I was using 3 which placed a termination right in the center. Not sure if that may be causing some phasy issue? Now using only 2strips so center of driver is open.
The diaphragm has a bit less control of low frequency movement with the larger cells and flatter profile and can no longer use a 6 db filter
"I think, somehow the panel curvature created/exacerbated comb-filtering and phase misalignment both. Are you able to try flipping the panel to curve the other way?"
BTW yes I have done this and to my surprise it seems to perform ( in listening) exactly the same as turned the "right way". Havent measured but Wahts up with that ha?
BTW yes I have done this and to my surprise it seems to perform ( in listening) exactly the same as turned the "right way". Havent measured but Wahts up with that ha?
Forum member @nfitch posted a new thread on the Bitches Brew Open Baffle design here: https://www.diyaudio.com/community/...ead-perry-marshalls-live-edge-dipoles.424917/ Has much-improved drawings with a good summary of the best current mods.
I have also find this "comb-filtering and phase misalignment" problem on OB panels with MTM speaker alignment, when panel is triangular shape.
Well all I can say is after about 60 hrs of testing various curve radiuses, then going back to the beamy flat panel, then going back to the concentric tweeter version the flat panel wins and not by a small amount.
I can get ruler flat freq response with the curve and the concentric , AND, the they both seem to neatly sidestep the larger diffraction issues the flat panel has However neither resolves detail nor image as well, nor have quite the dynamics that the flat panel has, and yes I payed particular attention to matching FR of all the units. I understand that small variations can have a big effect BUT after going back and forth as many times as I did and altering FR as much as I did the flat panel always more convincing overall, imaged better, and had a more dynamic quality.
The curve panel is beautiful to look at and it does give a slightly, and I mean only slightly btw, wider sweet spot than the flat panel BUT there is something wrong with the way the curve sounds. It doesn't sound bad, it just doesn't do what the flat panel does. The flat panel is more top class overall
So I went back and forth between curves, concentric tweeter, and flat. In the end I chose the flat and its narrow sweet spot hands down. Once ya get its freq response tailored to proper tone the band is in the room, properly placed in space, and the nuances and dynamics are simply more convincing. I got to the WOW stage with the flat.
I thought I would never get over the narrow sweet spot BUT I found that they all had a narrow sweet spot, and once ya get your speakers placed well and have your seat settled the beamy flat panel wasn't nearly the issue I had with it at first.
BTW this setup has seriously dynamic punch. Something ive never gotten out of any open back speaker of any kind in the past. What seemed critical was the woofer crossover. I landed on an 18db/oct at 300 hz and a 6 db/oct on the planer section. This with a 12 inch PE pro woofer.
I still want to try a woofer thats better behaved through the cross and another octave at least. I believe it may be that all these larger pulp woofers simply cannot match the planer all that well beyond about 250 hz. Sure they measure good BUT I believe they are flexing quite a bit beyond this point and that signature isnt as clean as the panel.
After much searching the only reasonably priced woofer I see that looks like a piston is the 12 inch PE aluminum woofer. Hardly a open baffle woofer but with some suspension mods it may be ok to 100 db??
I would love for this woofer assumption Im making to not be true but I suspect the next level will need a big woofer thats pistonic farther out??
I can get ruler flat freq response with the curve and the concentric , AND, the they both seem to neatly sidestep the larger diffraction issues the flat panel has However neither resolves detail nor image as well, nor have quite the dynamics that the flat panel has, and yes I payed particular attention to matching FR of all the units. I understand that small variations can have a big effect BUT after going back and forth as many times as I did and altering FR as much as I did the flat panel always more convincing overall, imaged better, and had a more dynamic quality.
The curve panel is beautiful to look at and it does give a slightly, and I mean only slightly btw, wider sweet spot than the flat panel BUT there is something wrong with the way the curve sounds. It doesn't sound bad, it just doesn't do what the flat panel does. The flat panel is more top class overall
So I went back and forth between curves, concentric tweeter, and flat. In the end I chose the flat and its narrow sweet spot hands down. Once ya get its freq response tailored to proper tone the band is in the room, properly placed in space, and the nuances and dynamics are simply more convincing. I got to the WOW stage with the flat.
I thought I would never get over the narrow sweet spot BUT I found that they all had a narrow sweet spot, and once ya get your speakers placed well and have your seat settled the beamy flat panel wasn't nearly the issue I had with it at first.
BTW this setup has seriously dynamic punch. Something ive never gotten out of any open back speaker of any kind in the past. What seemed critical was the woofer crossover. I landed on an 18db/oct at 300 hz and a 6 db/oct on the planer section. This with a 12 inch PE pro woofer.
I still want to try a woofer thats better behaved through the cross and another octave at least. I believe it may be that all these larger pulp woofers simply cannot match the planer all that well beyond about 250 hz. Sure they measure good BUT I believe they are flexing quite a bit beyond this point and that signature isnt as clean as the panel.
After much searching the only reasonably priced woofer I see that looks like a piston is the 12 inch PE aluminum woofer. Hardly a open baffle woofer but with some suspension mods it may be ok to 100 db??
I would love for this woofer assumption Im making to not be true but I suspect the next level will need a big woofer thats pistonic farther out??
Attachments
I hope they cross over ir very high then...
But what about an OB with a non dipolar tweeter? Does it sound not weird?
But what about an OB with a non dipolar tweeter? Does it sound not weird?
The widerange looks like Visaton B200 (with a bit modified dustcap). Maybe they crossed it around 10k to a ribbon tweeter...