The backwave is dead.Long live the back wave.

[Gregadd]

http://forum.audiogon.com/cgi-bin/fr.pl?vdone&1109799786
I chose this picture because I know Neil will not be offended I want to talk about the effects of a back wave on a speaker that IMO has such excellent phase response. Unlike the Magnepan that is a true dipole that is theoretically putting out an identical wave out the back with the potential to be bounced off the rear wall and have all kind of cancellation effects. The back wave is 180 degrees out of phase.
I've always tried to deal with the effect by moving the speakers 1/3 of the way into the room. As you can see Neil is using his software as room treatment.
M/Ls' curved panel means although there is mirror image it is very different from say the Innersound electrostatic panel. The Innersound course like the Magnepan, a true dipole.

The question is whether those whose M/L is close to the wall, say inside three feet needs to have some serious sound absorption to neutralize the back-wave?A device might be the Soundlab SALLIE. or some other sound absorption.
Your thoughts Please.

 

[kach22i]

I have played with absorption behind my original Aerius speakers, did not like it at all.

I would rather have an overly live room over a dead room which might help explain it.

Then again I was not using anything special or scientific, no mic's, computers, electronic measuring devices and so forth. God forbid I actually trusted my ears.

I do have absorption in between the speakers to focus the center stage.

Also, I built a insulated wall behind my speakers (face is 30" off wall) which acts like a diaphragm, in theory bending to mid-bass frequencies but not absorbing the high and real low frequencies. I have a collection of books on acoustics, one from my college course on the topic which helped me fake my way though it. Pretty good fake job, the sucker works!

Photos here: (pic below is just some more fooling around)
http://s184.photobucket.com/albums/x295/kach22i/Stereo/?start=40

[/IMG]

My wall compensates for the stairs being behind the speakers. Starting from the front face; perf metal, one inch airspace, 1/8" veneer plywood, 4" air cavity with 3" batt insulation on homosote backboard and oak trim all around. The wall is not secured or fixed to anything, and has acoustic foam between the stairs and it's back to snug it up and prevent it from tilting over.

I think that the first photo posted (Neil's) and mine rely mostly on diffusion of the rear wave.

 

[Gregadd]

Darn it? Why can't I post nice pictures like that?

 

[kach22i]

There is a tiny symbol at the top of the posting box when you click "reply" which looks like a picture of a tree in a frame. 3rd from the last, in between the film like thing and the earth like thing.

Make sure your hotlink is in between the two IMG's with brackets. You should get a pop-up when you click on the tree symbol. Just copy and paste using CTRL-C and CTRL-V.

 

[nsgarch]

You have to remember that the back wave from a ML (and a Soundlab) panel are focused, because of the concave diaphragm shape, to a (more or less) narrow vertical line, which occurs (more or less) about 3 to 6 feet behind the speaker, and then spreads out again further back. You don't need an absorber of much width to capture it (which is what Soundlab's Sallie thingy attempts to do -- and I suppose does, though I've never heard one in action.) And you'd better capture it, quick, before it passes further back from that focus point, because all kinds of interference things start to happen (like two guitar strings just slightly out of tune.) The point here is that simply pulling (curved) speakers well out into a room isn't such a hot idea, even if the space allows.

You definitely don't want a hard wall, close behind the speaker either. It will reflect that concentrated wave right into the back side of the diaphragm!! In the picture of my system, although it may not look like it in the photo, the back panel of the record shelf unit is exactly three feet from the rear of the CLS diaphragm, or roughly a foot and a half before the rear wave would come to its theoretical point of focus. Any 'bounce back' is too attenuated and busted up (diffused) to impact the back of the CLS diaphragm in any signficantly negative way.

 

[Gregadd]

Sallie

http://www.soundlab-speakers.com/accessories.htm

 

[kach22i]

What's the point of owning dipole or bipole speaker if you are just going to kill the ambiance of the rear wave?

Diffuse, don't destroy and absorb.

 

[nsgarch]

What's the point of owning dipole or bipole speaker if you are just going to kill the ambiance of the rear wave? Diffuse, don't destroy and absorb.

It's not about killing (or not killing) the rear wave. The rear wave must be controlled however. And much more so with MartinLogan panels (and Soundlab panels) as I pointed out in my previous post.

Completely gobbling up the rear wave won't destroy the presentation of an electrostat. Remember that except for bass reflex or horn loaded enclosures using conventional woofers, most modern speakers, of all types, contain the drivers' rear waves. However, stats are not terribly efficient, so as with ports and horns, if one can save at least some of the rear wave and integrate it phase coherently with the front wave, the speaker's efficiency is greatly improved.

 

[Gregadd]

Conduct your own experiment . Take your Logans and press them all the way against the the wall. Move them out about six inches at a time until they are 1/3 the way into the room. Note the results. At some point the will move from thin and diffuse to having slam and coherency. If you use some sort of back wave diffuser this will happen when the speakers are much closer to the back wall. This is why I am so concerned by seeing so many M/Ls so close to the back wall.

 

[Steve williams]

wouldn't the same apply to just about all speakers

 

[FrantzM]

wouldn't the same apply to just about all speakers

Interesting discussion,more later. Suffice to say that the back wave can play havoc in the performance of a dipole if not controlled carefully ... Later ...

Frantz

 

[nsgarch]

Conduct your own experiment . Take your Logans and press them all the way against the the wall. Move them out about six inches at a time until they are 1/3 the way into the room. Note the results. At some point the will move from thin and diffuse to having slam and coherency. If you use some sort of back wave diffuser this will happen when the speakers are much closer to the back wall. This is why I am so concerned by seeing so many M/Ls so close to the back wall.

Greg, it's possible to misinterpret one's observations from this experiment. Between zero feet and around five feet or so from a back wall, there is enough energy being reflected back against the rear of the diaphragm to actually move it -- turning it into what's called a "passive radiator". If that wave impacts the diphragm out of phase with the wave actively being produced by the diaphragm, then the resulting (compound) wave will be full of all kinds of nasty artifacts -- but primarily phase distortion. So unless there is a large open window right behind the speaker, then regardless of the treatment on that wall, or no treatment at all, it's virtually impossible for the speaker to "get out of the way" of that reflected energy.

If the speakers are sufficiently far away from the back wall, and toed-in just slightly, it's possible to bounce some of that energy off the back wall and around to the left or right of the speakers where it will meet with the front wave, pretty much out of phase with it, and cancel. Many don't realize it, but with all normal height/width proportioned panel speakers, there are two 20-degree "cancellation zones" extending out from each vertical edge (of each speaker), where the direct front radiation and rear wave meet and cancel. With curved panels, those zones face somewhat toward the rear and not straight out from the sides. Interestingly, this unique phenomenon of panel speakers reduces the (soundstage destroying) effect of early sidewall reflections.

If one spends a little time with a nice big sheet of paper drawing ray tracings on a scale plan of the listening room, it's quite amazing what you can learn about dipole speaker placement. Three common mistakes have become very clear to me, over the years, with regard to the way people place panel speakers in small to average sized (13 x 20 ish) residential rooms:

• Too close to the back wall -- well, of course; but don't pull them out so far that they excite standing waves in the room.
• People place speakers too far apart. In a 12 foot wide room, always allow a minimum of 30" from the outside edge of each speaker to the sidewall; and 36" from the outside edge of each speaker to the sidewall for a 13 or 14 foot wide room -- and just let the speaker-to-speaker distance fall where it may! In rooms 15 feet wide or wider, it's first-of-all preferrable to place the speakers on the longer wall -- but either way, in no case should the center-of-panel to center-of-panel distance between speakers be greater than 8 feet max. even if you live in a cathedral! (I prefer 7 feet, especially for the narrower panels like a Summit or Vantage.)
• People use WAY too much toe in. (Goes with WAY too far apart ;--) Always get the speakers close together (even if the room doesn't force you to ;--) and pointing straight forward; and perfectly VERTICAL (not tilting forward, or backwards -- and certainly not leaning to one side or another!! -- what do you think 6 foot levels were made for??) This position is the default starting point for toe-in, which should be carried out in half inch increments, meaning: moving each inside edge of each panel back a half inch from the outside edge of the panel. An easy way to do this is with a string, its ends securely taped around (and stretched between) the outside edges of each speaker (near the floor). As you toe each speaker in, you simply measure the distance between the stretched string and the inside edge of each speaker, a half inch at a time. Then listen. This toe-in offset should NEVER be more than 4 inches. If you go too far (and the soundstage starts to weaken) reduce the toe-in to a previous setting, and begin to increase or decrease the speaker-to-speaker distance in 2" increments. It's a little like dialing in a cartridge, but you WILL nail it . . . . eventually ;--)

Here is a picture of my 12 foot wide by 24 foot long room (8 feet of the length is the kitchen area behind the bookshelf unit.)
http://forum.audiogon.com/cgi-bin/fr.pl?vdone&1109799786&read&keyw&zznsgarch
The distance between the inside edges of the speakers is 31"! The center of panel to center of panel distance is only 4'-10"(58")! and the outside edges of the panels are 29.5" from the side walls. At my listening position, my ears are exactly 5'- 9" from the front of the panels. The soundstage extends well beyond the sides of the room, and the soloists are well focused. The speakers are elevated a foot, and braced, on Sound Anchor special CLS stands, and my listening chair is elevated so my ears are at mid-height of the speaker panels. This is what's called 'nearfield listening' (I call it "control-booth" listening ;--) and is the mandatory setup for rooms narrower than 14 feet, regardless of the type of speakers one uses.

 

[Gregadd]

Excellent work my friend. For true nearfield listening check this out.http://cgim.audiogon.com/i/vs/s/f/1261362990.jpg

 

[nsgarch]

For true nearfield listening check this out.http://cgim.audiogon.com/i/vs/s/f/1261362990.jpg

Very funny!! What would you call that Greg, a "poor-man's" Stax headphone?? The really hilarious thiing is that 95% of each panel's output is going right past the listener's head and running smack into the opposite diaphragm. Yikes!

 

[Gregadd]

He's a nice guy and I really don't want to make fun of him. May be he has a hearing problem. It gives new meaning to the term "killed by friendly fire."

 

[kach22i]

I remember that photo, silly thing this stereo nervosa is.

 

[nsgarch]

. . . . . . . maybe he has a hearing problem.

. . . . . if he doesn't now, he will soon!!

 

[Gregadd]

Moving the speakers closer together actually widened the sound stage instead of shrinking it. Making the sneakers plum also improved things. I also moved the couch about one foot form the wall. Great improvement. I'll begin toe in this weekend. I assume the goal there is to get the middle of the speaker to point at you.
I try to follow the rule of thirds. so for an eighteen foot room The speaker should be on th first line and you should sit on the second line.

 

[vinylphilemag]

I'll begin toe in this weekend. I assume the goal there is to get the middle of the speaker to point at you.

Not with MLs. With ML electrostats, you want the middle of the inner third of the stator pointing at you.

When I first set up my Spires (before I read the manual--I couldn't wait!), I did just as you said. The sound was much improved after setting them up using ye olde flashlight trick.

 

[nsgarch]

More accurate (because it's easier to read) is a little laser pointer. Sit in your listening chair and place/hold the laser pen on top of your head. Point the laser spot straight forward between the speakers and at eye level, while at the same time looking straight forward. (It's easier and more accurate than it sounds!!)

Then slowly rotate you head left then right (just in the horizontal plane; don't look up or down!) until you see the laser in the diaphragm. Note: if the speaker isn't perfectly vertical front to back, then you MAY NOT even see the laser; or you might, but it will be above or below your eye level. So the first adjustment (before toe-in) is to make sure the speaker panel doesnt tilt forward or backward. If you have an accurate carpenters level 3 feet long or more, just place one edge, straight up and down, against the middle of the front stator, and tilt the speaker back and forth until the level reads plumb. The reflection of the laser point should now be right opposite your eye and at the same height on both panels.

Vinylphile is correct, the toe-in should be such that the laser reflection falls on a vertical line (roughly) 1/3 the panel width from the inner edge of the panel (and of course identical for both speakers.) To really dial in the toe-in, play a female solist in mono and increase the toe-in until she forms a solid image between the speakers. Go back to stereo (don't worry if she shifts a little left or right in stereo -- it could be the recording, or worse ;-) When in stereo, the overall soundstage (forget the centered soloist for a minute) should NOT feel congested/squeezed between the speakers. If it does, reduce the toe-in until the soundstage opens up to beyond the outside edge of the speakers.

NOTE: None of this will work if the speakers are too far apart!