My point is that there are tons of theoretical words about the perils of beaming but I'd rather see the simple-as-could-be, A versus B or even just a single flat panel measurement in somebody's music room. Do you have a link to that sort of data?
My own set-up wouldn't be meaningful. My panels are composed of flat cells and are 3 cells wide and 2 cells high, set on a curve. So there's no way to distinguish the curve from the dispersion.
I don't know why you accuse me of laying some effort on bolserst (who always goes beyond the call of duty) and I didn't refer to him at all.
Ben
Last edited:
I'll just add my 10 cents worth...
In theory:
> the difference between segmented and single segment ESL begins only above a few hundred Hz or so, differences only perceptible probably at frequencies nearer 1 kHz. Segmentation will not change the bass at all (all other things being equal).
> on axis (at exactly 0 degrees) there should be no difference if the two are equalised properly.
> The onset of problems varies according to the width of the panel, but a single segment ESL starts showing fall-off at highest frequencies within 1 degree or so, while a segmented ESL will have a flat response over 20 degrees or more.
> a listeners head (150 mm at 3m) is about 3 degrees wide. So positioning for ideal listening with single segment requires a vice. No problem with multi-segment ESL.
> In my experience, perfect imaging with segmented ESL still requires careful placement of ESLs and listener, but listener position is quite forgiving - several degrees easy.
> if you only want a balanced sound (no imaging) while moving around in the room, then the segmented ESL is good over very wide range of listening positions - better than many conventional speakers.
> Properly balanced audio requires all the room reflections to be properly equalised too. Segmented slightly better in this respect.
> Charlie's experience is interesting - shows how sensitive bass is to matching of frequency response. With his ESL set to wide response, the midrange sensitivity will be reduced a couple of dB, but the bass sensitivity will stay the same.
regards
Rod
In theory:
> the difference between segmented and single segment ESL begins only above a few hundred Hz or so, differences only perceptible probably at frequencies nearer 1 kHz. Segmentation will not change the bass at all (all other things being equal).
> on axis (at exactly 0 degrees) there should be no difference if the two are equalised properly.
> The onset of problems varies according to the width of the panel, but a single segment ESL starts showing fall-off at highest frequencies within 1 degree or so, while a segmented ESL will have a flat response over 20 degrees or more.
> a listeners head (150 mm at 3m) is about 3 degrees wide. So positioning for ideal listening with single segment requires a vice. No problem with multi-segment ESL.
> In my experience, perfect imaging with segmented ESL still requires careful placement of ESLs and listener, but listener position is quite forgiving - several degrees easy.
> if you only want a balanced sound (no imaging) while moving around in the room, then the segmented ESL is good over very wide range of listening positions - better than many conventional speakers.
> Properly balanced audio requires all the room reflections to be properly equalised too. Segmented slightly better in this respect.
> Charlie's experience is interesting - shows how sensitive bass is to matching of frequency response. With his ESL set to wide response, the midrange sensitivity will be reduced a couple of dB, but the bass sensitivity will stay the same.
regards
Rod
The question I put to CharlieM about his segmented ESL could not have been answered by Floyd Toole's work (I have several of his articles) nor measured by REW ( which software I have) since it was CharlieM's opinion that I was interested to hear, given that he had worked with both types of ESL, panel and segmented. Thanks CharlieM for the full and detailed answer.
I agree with Wrinex's suggestion that there are many useful discussion of directivity in diyaudio. I found Bolserst's simulations very helpful in this respect: Attachment 1, which shows the directivity of an unsegmented v. segmented panel.: Post #47 http://www.diyaudio.com/forums/planars-exotics/234975-another-segmented-esl-5.html
Golfnut's details of the directivity are most useful magnitudes to bear in mind when considering directivity. Lest members of the forum be in any doubt, I had already made a decision to implement a symmetrically-segmented ESL design.
I agree with Wrinex's suggestion that there are many useful discussion of directivity in diyaudio. I found Bolserst's simulations very helpful in this respect: Attachment 1, which shows the directivity of an unsegmented v. segmented panel.: Post #47 http://www.diyaudio.com/forums/planars-exotics/234975-another-segmented-esl-5.html
Golfnut's details of the directivity are most useful magnitudes to bear in mind when considering directivity. Lest members of the forum be in any doubt, I had already made a decision to implement a symmetrically-segmented ESL design.
Stereophile has measurements of several flat panel speakers...here is one.
InnerSound Eros Mk.III electrostatic loudspeaker Measurements part 2 | Stereophile.com
Note that the Fig. 10 polar response matches the theoretical results ENCR linked to.
Just 5 few degrees off axis(one couch cushion from the sweet spot) and you have lost the whole top two octaves.
Not sure if you were aware, but the phase of the response lobes alternates between 0deg(in phase) and 180deg(out of phase).
This off-axis behavior does not rate highly with Toole methodology.
Fig. 14 showing in room response may be of more interest to you; same results.
The accompanying comment...
"The red trace in fig.14 demonstrates that listening on-axis will reveal an astonishingly flat response from the low bass through to the top octave. But, as LG noted, the tolerance of the listener's head position that gives this extraordinary performance is measured in millimeters. Turn your head slightly, and the Eros's high treble simply disappears.
Thought I might add a few other relevant links:
Measurements validating smooth broad coverage of segmented ESL polar response - http://www.diyaudio.com/forums/planars-exotics/245454-glue-wire-stators-3.html#post3708397
Theory showing polar map of 30 deg curved ESL - http://www.diyaudio.com/forums/planars-exotics/246846-first-time-esl-builder-14.html#post4163636
Data from rooms. Thanks.
Your sonograms show dramatic differences with frequency, as far as I understand these displays. But made in a home music room?
In the Stereophile link, there is fig 10 which is taken at 50 inches and I am not sure of the sample time window (they mention below 5 ms which eliminates all reflections and rear radiation). Yes, a good way to confirm the theory that in theory, up-close, there is beaming with a largish, non-segmented flat panel. I can't read the figure too well but it looks like a lot of variation and big angles. Hardly like the opinion they present that the sweetspot is just "millimetres" in width.
As for fig 14, hard to nail down the trend lines and not clear what the time window is again. But it looks like north of 10kHz, the loss is in the vicinity of 5dB, which doesn't seem like much and needing only minor sort-of first-order EQ, in the event your ears agree.
BTW, the editors comment on how deplorable it is to have reflections of any sort. Today, that would get a big laugh from many of us. But it may show how they tilt their testing to catch only direct radiation.
Nobody is doubting the theory of beaming. But in real rooms with dipoles, is the detriment of using largish flat panels due to beaming, worth the pain to a DIYer of trying to make them curved.
Your priorities may be different if your dog does a lot of listening.
Sanders' rebuttal is that he is creating a reference speaker that, I assume, sounds perfect on-axis. While my head is more than a few millimetres wide, on a good day, I too care only about a single sweetspot and other locations are of secondary importance to me. So beam on.
Ben
Last edited:
Now I think most have saw this..... but seeing how we all seam too post a lot of the same things over an over........
Sanders Sound Systems - Dispersion White Paper
From Sanders Dispersion White Paper
Many audiophiles believe that a good speaker should have a wide sweet spot. But this is a physical oxymoron.
The laws of physics dictate that all stereo speakers will have an infinitely small sweet spot, regardless of their high frequency dispersion. That spot is when you are exactly equidistant from both speakers. Only when you are equidistant from the speakers can the phase information arrive at your ears simultaneously from both speakers. Obviously, there is no hope of imaging well if the sounds from both speakers do not arrive at the same time as the phasing will be destroyed.
For a speaker to have a wide sweet spot simply means that the phase information from the room is confusing the sound so badly that you can't even tell when you are in the sweet spot and when you aren't. A wide sweet spot is a guarantee that a speaker has poor imaging and transient response.
I don't compromise. I want narrow dispersion in my speakers to minimize room acoustics so that I can get the best possible sound.
"Beaming" is not a fault. It is a huge advantage. It is the only way to achieve truly high performance in a loudspeaker.
Audiophiles sometimes say that narrow dispersion speakers require you to have your head in a vise. This is nonsense. You just sit in your listening chair and listen as you would to any speaker.
And what about the off-axis performance of a narrow dispersion speaker? Well, they sound just like wide dispersion speakers when you are off-axis.
That is to say that when you are off-axis, you hear the room acoustics!
After owning Adiostats ES100 , Acoustat Spectra an Martin logans all these speakers made to give wide dispersion .... thay all are a lot of fun .... but my ears, well have to stay with the physics.....like Sanders, who wonts too compromise...like it or not the best ,real sounding in home speakers.. talking ESL here...
Flat low dispersion panels ..is were we end up.
I guess this case, it is more than just one mans finding
Sanders Sound Systems - Dispersion White Paper
From Sanders Dispersion White Paper
Many audiophiles believe that a good speaker should have a wide sweet spot. But this is a physical oxymoron.
The laws of physics dictate that all stereo speakers will have an infinitely small sweet spot, regardless of their high frequency dispersion. That spot is when you are exactly equidistant from both speakers. Only when you are equidistant from the speakers can the phase information arrive at your ears simultaneously from both speakers. Obviously, there is no hope of imaging well if the sounds from both speakers do not arrive at the same time as the phasing will be destroyed.
For a speaker to have a wide sweet spot simply means that the phase information from the room is confusing the sound so badly that you can't even tell when you are in the sweet spot and when you aren't. A wide sweet spot is a guarantee that a speaker has poor imaging and transient response.
I don't compromise. I want narrow dispersion in my speakers to minimize room acoustics so that I can get the best possible sound.
"Beaming" is not a fault. It is a huge advantage. It is the only way to achieve truly high performance in a loudspeaker.
Audiophiles sometimes say that narrow dispersion speakers require you to have your head in a vise. This is nonsense. You just sit in your listening chair and listen as you would to any speaker.
And what about the off-axis performance of a narrow dispersion speaker? Well, they sound just like wide dispersion speakers when you are off-axis.
That is to say that when you are off-axis, you hear the room acoustics!
After owning Adiostats ES100 , Acoustat Spectra an Martin logans all these speakers made to give wide dispersion .... thay all are a lot of fun .... but my ears, well have to stay with the physics.....like Sanders, who wonts too compromise...like it or not the best ,real sounding in home speakers.. talking ESL here...
Flat low dispersion panels ..is were we end up.
I guess this case, it is more than just one mans finding
I think you have misunderstood Fig. 14. Both red and blue curves are on-axis “in room” measurements with long time windows, just different rooms. The point of the plot was that even with an “in room” measurement showing flat response on-axis at the listening position, small head movements still resulted in loss of upper treble. In other words, the windowed dispersion data shown in Fig 9 indicating loss of upper octaves with small movements off-axis is not helped with the addition of room reflections/ambience.
The reason for this is that the total acoustic energy emitted into the room in the upper 2 to 3 octaves by a flat unsegmented ESL is well down in level relative to the midrange and what is there is emitted in alternating in-phase/out of phase lobes. The end result is almost no room ambience or support above the midrange. As I mentioned in previous post, this is counter to the Toole philosophy concerning the benefit of smooth, tonally balanced room reflections that you have regularly indicated support for.
Another way to get a feel for the balance of room energy is to look at the falling response in Fig 9 which is the average response for -15, -10, -5, 0, 5, 10, & 15 degree data. You can appreciate that the larger the angular coverage included in the plot, the more the average of the radiated acoustic energy would show deficiency in the top octaves.
The first curve after the on-axis flat line is for 5 deg off-axis and is already -6dB down by 3khz.
In my experience, yes. Even if you only care about having balanced sound in the single imaging sweet spot and don’t mind holding your head still during extending listening sessions, unsegmented flat panels do not provide smooth balanced off-axis response required for optimal room ambience. Of course as soon as you consider wanting to listen to balanced sound outside the imaging sweet spot, something other than flat unsegmented panels becomes necessary.
If you are interested in further discussion on the pros and cons of flat panels perhaps start a thread on the subject?
Bolserst ....thanks for all your info on all thing ESL
Even if you only care about having balanced sound in the single imaging sweet spot and don’t mind holding your head still during extending listening sessions,
I see this a lot..but I find with the Acoustat 9" panels ...it the small squares that can make this....more ... than others
If I was will to give up the Magic of the ESL sound..or thought I could get more than one set of ears to hear ... more real sweet spots in one room....with one pr of Speakers...but this dose not work...
.About to take the ESL plunge.... I would think this is the place...
Even if you only care about having balanced sound in the single imaging sweet spot and don’t mind holding your head still during extending listening sessions,
I see this a lot..but I find with the Acoustat 9" panels ...it the small squares that can make this....more ... than others
If I was will to give up the Magic of the ESL sound..or thought I could get more than one set of ears to hear ... more real sweet spots in one room....with one pr of Speakers...but this dose not work...
.About to take the ESL plunge.... I would think this is the place...
Last edited:
With a half-angle of 5 degrees, at 10 feet the beam is about 21 inches (not millimetres). Or about 3 airplane seats wide.
Ben
Ben
In as much as the discussion of beaming started in post #2, I'll just add a bit of data here if nobody minds.
Figures made with REW close to my head when listening, ⅓ or 1/6 oct. smoothing to better show beaming problem. These show a single curved panel composed of 6 Dayton-Wright cells that I built around 1976* driven by DW transformers and no EQ. Runs made a few minutes ago.
The first picture shows FR and distortion (THD, 2nd and 3rd.... pretty orderly, eh) and is included mostly to just show off (better than 40 dB which includes a whole lot of old gear, airplanes overhead, and so on).
The second shows the average of three runs on either side of my head and centred, about 18 inches side to side*.
The third shows the three runs in the variable real world.
Not a whole lot of beaming arising from a rather variable set of FR curves. My point is (still) that whatever abstract truth there may be to beaming, comb filtering, and the like, in practice not something that hits you over the head, so to speak, in your music room. You can't hear beaming, you can only say, "Gosh, I think I heard the triangle a bit louder when standing over there...maybe."
About that Sanders write-up that tyu seems to like. Nothing could be more naive that to imagine textbook geometry can be applied to the wild and wonderful chain of dogey manipulations between the brass section and your ears. Instruments are mic'ed separately - even if done on the same day! - and any "depth" is cooked-up by the engineer. Just an example of naive pontificating characteristic of tyu's post as well.
Ben
*Yup, 1976 but might have been earlier
** My head is slightly smaller but I wanted to catch the beaming if present
Figures made with REW close to my head when listening, ⅓ or 1/6 oct. smoothing to better show beaming problem. These show a single curved panel composed of 6 Dayton-Wright cells that I built around 1976* driven by DW transformers and no EQ. Runs made a few minutes ago.
The first picture shows FR and distortion (THD, 2nd and 3rd.... pretty orderly, eh) and is included mostly to just show off (better than 40 dB which includes a whole lot of old gear, airplanes overhead, and so on).
The second shows the average of three runs on either side of my head and centred, about 18 inches side to side*.
The third shows the three runs in the variable real world.
Not a whole lot of beaming arising from a rather variable set of FR curves. My point is (still) that whatever abstract truth there may be to beaming, comb filtering, and the like, in practice not something that hits you over the head, so to speak, in your music room. You can't hear beaming, you can only say, "Gosh, I think I heard the triangle a bit louder when standing over there...maybe."
About that Sanders write-up that tyu seems to like. Nothing could be more naive that to imagine textbook geometry can be applied to the wild and wonderful chain of dogey manipulations between the brass section and your ears. Instruments are mic'ed separately - even if done on the same day! - and any "depth" is cooked-up by the engineer. Just an example of naive pontificating characteristic of tyu's post as well.
Ben
*Yup, 1976 but might have been earlier
** My head is slightly smaller but I wanted to catch the beaming if present
Attachments
Last edited:
Oops, forgot to mention, measured at 104 inches. So that forms a triangle 18 x 104 x 104 which is about 10 degrees (or a 5 degrees half-angle).Whatever beaming or none that can be discerned in the 3rd figure, the measurement window is 18 inches wide at 104 inches distant from panel.
B.
B.
By saying the blade made the biggest difference, were you meaning you purchased a new blade? or using the same blade but at the lower height shown.
Wow, remind me never to fly on Canadian airlines 😉
Excluding armrests the average airplane seat is about 17” - 18” wide, not 7”.
For best performance the panel axes will be centered (or nearly so) on the imaging sweet spot. 5 degrees @ 10ft. would then be 10.5” to the side. So, even with a significant other sitting quite closely next to you, they will be hearing rolled off top octaves from one or both panels just as the measurements show and people describe.
Don’t get hung up on millimeters…the statement was a hyperbole, not meant to be taken literally. The reviewer just meant that adjustment of seating position, posture, or tilting of the head, as is usual for most humans, may cause distracting changes in the tonal balance of the sound received by each ear; nothing like moving over a full seat though. The key thing the review was trying to point out was that it was quite noticeable to him and different from typical loudspeaker listening experience. I will add that the effect is more bothersome to me when listening to a stereo pair than when listening to just a single test panel. I think this may have to do with the increased ratio of direct vs. ambient sound. As mentioned in previous posts, room ambience from big flat panels has very little energy in the top octaves. So when you move your head and the direct response changes, there is no ambient energy to help keep the perceived tonal balance on an even keel as is the case with most other loudspeakers having broader HF dispersion characteristics. With a listening distance of 10ft, Acoustat 1+1s with their 7” wide HF radiating area is about as “beamy” as I can go before it really starts to bother me for solo listening in the sweet spot. However, there are some listeners that are perfectly happy with 12” or wider flat panels and grow accustomed to the effect much as many do with pops and clicks from LPs. But, this doesn’t mean that the pops and clicks aren’t there, and might bother some people.
I don’t think any measurement will fully address your skepticism that some listeners are bothered by the tonal balance shifts mentioned above. I base this opinion on your comments concerning the relation between the 5 degree incremental measurement data and listeners seated outside the imaging sweet spot…a much clearer case. Certainly all verbal descriptions of first-hand experience with either situation have failed. The only option remaining seems to be for you to personally listen to a pair of 12” to 16” wide un-segmented, un-curved panels. If the opportunity presents itself, I would certainly recommend giving it a try and see what you think. If you are so inclined, you could remount the panels from your curved arrays into temporary flat frame arrays.
Took me nearly three hours to read, but what a great thread! ...with boatloads of information! A lot of which has answered many of my pre-build questions. 🙂
bengal, I would love to hear your listening impressions, (and view pictures), once these are completed and set up with your system.
Congratulations on the build and thanks for all the details, including the setbacks and perils along the way, which I think are equally informative. 😉
Cheers
-Steve
bengal, I would love to hear your listening impressions, (and view pictures), once these are completed and set up with your system.
Congratulations on the build and thanks for all the details, including the setbacks and perils along the way, which I think are equally informative. 😉
Cheers
-Steve
Hello everyone... I'm back from the dead 🙂.....
Been so busy "other" stuff (you wouldn't believe) I haven't had serious time to put into completing my ESL's. But now I'm back at and looking to get these done ASAP.
So got my stretching jig built (see pics below). I'm attempting to get the stretch the mylar and duck tape doesn't seem to work too well holding the mylar in place. The tape will literally pull (shear) right off the mylar. Seems to hold down to the jig well enough after many layers but can't seem to get it to stay stuck to the mylar.
Any recommendations on tape you guys have used? Gorilla tape is my next try but not sure if it is going to work any better.
Been so busy "other" stuff (you wouldn't believe) I haven't had serious time to put into completing my ESL's. But now I'm back at and looking to get these done ASAP.
So got my stretching jig built (see pics below). I'm attempting to get the stretch the mylar and duck tape doesn't seem to work too well holding the mylar in place. The tape will literally pull (shear) right off the mylar. Seems to hold down to the jig well enough after many layers but can't seem to get it to stay stuck to the mylar.
Any recommendations on tape you guys have used? Gorilla tape is my next try but not sure if it is going to work any better.
Attachments
I use plain brown packing tape. I have also used the clear stuff but it tends to leave a lot of glue on the table.
PPS 48mm x 50m Mailing Packaging Tape Brown 6 Pack | Officeworks
Regards,
kffern
That way you can get all the creases out and stretch it evenly.
I use the ER method of small spring weight gauge.
kffern
I have used both pneumatic (bike tube) and mechanical jigs like yours and on both types I rounded over the top edges and wrapped the Mylar over the edges and taped it to the backside of the jig with double sided tape. This method works quite well.
I also use the method recommended by ER, lengths of duct tape at right angles to the edge, all tensioned evenly using a fishing scale. The four corners first, then the middle of each edge, then continue bisecting the length between the ones you've already put down until there is no more room.
You will need a very clean base to work on, the smallest sharp point will cause the mylar to rip. I used an old glass shower door thoroughly cleaned. The only time I had a rip was with a ragged fingernail - now trim nails beforehand. Since then - never a problem.
Use a hot soldering iron to cut the mylar away after the stator section has been glued down.
You will need a very clean base to work on, the smallest sharp point will cause the mylar to rip. I used an old glass shower door thoroughly cleaned. The only time I had a rip was with a ragged fingernail - now trim nails beforehand. Since then - never a problem.
Use a hot soldering iron to cut the mylar away after the stator section has been glued down.