Back to back electrostatic speakers

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I am familiar with the manner in which a standard dipole suffers in the low end when placed too close to a wall -- the out of phase rear wave largely cancels with the front wave when the reflected path length is small relative to a wavelength. DSP can mitigate this, I suppose within amplifier and driver limits.

If one were to try to create a "bipole" by using a pair of electrostatic loudspeakers placed back to back and wired in phase to a mono source, would the bass response suffer even further? It feels like even though the panels are flexing away from each other, which is in some ways similar to a bipole loudspeaker, the fact that the back waves aren't trapped in a sealed box would create a mess in the room. I'd love to try it out, but I have no access to any planar loudspeakers. I would think electrostats would be the worst candidate because of the relative transparency of each membrane to impinging sound.

Have any of you tried this? Any links/papers/reviews?
 
Moderator
Joined 2011
The two sound outputs would largely cancel through most of the range, from the LF lower limit,
up to where the panels' separation approaches a half wavelength. There would be some output
above that, but it would be very uneven.
 
Last edited:
Moderator
Joined 2011
Do planars play well in a sealed box? I don't recall seeing it done.

Their selling point has always been "boxless" sound. Someone, somewhere may have done it, though.
If you get a pair, try setting them in a doorway or entranceway to isolate the rear wave.
Electrostatic tweeters have been mounted on the baffle of box speakers, JansZen for example.
 
Last edited:
Their selling point has always been "boxless" sound.

Understood, but that's a pretty muddy distinction. They're different than boxed monopoles in many many ways, not the least of which is the driver technology.

I was getting at a PAIR of planars in a box, regardless, driven as a bipole array. I just don't know if (for instance) driver performance would suffer, perhaps due to back pressure or other effects.
 
Beveridge speakers used an enclosure, with the ESL running down reasonably low.

The newer Janszen models also use an enclosure for the ESL. Crosspoint is stated as 500 Hz for the Valentina.

As already discussed, reflections from any near surface back through the diaphragm have to be dealt with, or the sound is highly colored/unpleasant. With enough careful absorption it can be done though.
 
Great, thanks for all the thoughts everyone. Makes sense. I was definitely trying to wrap my mind around ways to extend the low frequency performance of ES panels, without resorting to conventional subwoofers, so solutions that stop at 500 Hz aren't interesting to me. I'm pretty sure plenty of designers would have done the back-to-back configuration, either in a box or not, if it had any merit.
 
It's a pressure source

Hi, from my limited understanding:-
I am familiar with the manner in which a standard dipole suffers in the low end when placed too close to a wall -- the out of phase rear wave largely cancels with the front wave when the reflected path length is small relative to a wavelength.

It's not the wall causing this, it's the rear wave wrapping around the speaker frame and cancelling the front wave.

In fact, when the anti phase rear wave bounces off the rear wall, it reverses phase and becomes in phase. In which case the closer to the rear wall the better! ;) So that is not why the speaker is kept away from the wall: it is an attempt to increase the delay, so the rear sound is perceived as more distinct and not a blur-causing part of the front wave direct energy -- and is more beneficial in the mid and high frequencies than the bass.

DSP can mitigate this, I suppose within amplifier and driver limits.

Driver limits at low frequencies are so severe for electrostatic panels that it would be IMO unwise to try using DSP for this purpose.

If one were to try to create a "bipole" by using a pair of electrostatic loudspeakers placed back to back and wired in phase to a mono source, would the bass response suffer even further? It feels like even though the panels are flexing away from each other, which is in some ways similar to a bipole loudspeaker, the fact that the back waves aren't trapped in a sealed box would create a mess in the room. I'd love to try it out, but I have no access to any planar loudspeakers. I would think electrostats would be the worst candidate because of the relative transparency of each membrane to impinging sound.

Have any of you tried this? Any links/papers/reviews?

ESL membranes are pressure sources and don't respond well to pressure on the back of the membrane, whether it be (imperfect) absorbent material, cabinet walls, or another membrane producing the pressure. They are quite different from dynamic drivers and cannot be treated in similar ways.

I don't recommend your approach at all.

cheers
 
It's not the wall causing this, it's the rear wave wrapping around the speaker frame and cancelling the front wave.

This is counter-intuitive. If it were true, then a dipole's bass response would not get worse as the speaker is placed closer to the wall, which is in fact what happens. (I'm not suggesting that there is no effect such as you describe, but merely that it doesn't tell the whole story.)


In fact, when the anti phase rear wave bounces off the rear wall, it reverses phase and becomes in phase.

Nope. The pressure doubles at the wall and the reflected wave comes off the wall in phase with the impinging wave. You are describing a pressure-release boundary condition, more akin to sound exiting a duct, but focused on that which reflects off the opening back upstream.
 
This is counter-intuitive. If it were true, then a dipole's bass response would not get worse as the speaker is placed closer to the wall, which is in fact what happens. (I'm not suggesting that there is no effect such as you describe, but merely that it doesn't tell the whole story.)

It's the fundamental principle of dipole radiation. It is the source of the figure-8 pattern.

Nope. The pressure doubles at the wall and the reflected wave comes off the wall in phase with the impinging wave. You are describing a pressure-release boundary condition, more akin to sound exiting a duct, but focused on that which reflects off the opening back upstream.

Quite right, I was thinking of sound in walls and cabinets. Confused myself.

Doesn't change the real reason why dipoles are kept away from walls, though, as I mentioned.
"....the (dipole) loudspeakers must be placed at some minimum distance from those large surfaces in order to delay specular reflections by more than 6 ms. This allows the brain to give primary attention to the earlier arriving direct sound...." Sigfried Linkwitz

Thanks for giving my comments some thought. The concept remains a bad idea, IMHO.

cheers
 
"....the (dipole) loudspeakers must be placed at some minimum distance from those large surfaces in order to delay specular reflections by more than 6 ms. This allows the brain to give primary attention to the earlier arriving direct sound...."


That quote is not limited to dipole loudspeakers. You added "(dipole)" but Linkwitz made no such limiting claim.​
 
You can get some insights into the effects of walls by imagining that the walls are mirrors. So, supposing that your speakers are a few metres apart and have a wall say 1 metre behind them. What you hear will be affected by the images of your speakers, a metre beyond the wall with their rear facing you.

We can draw the following conclusions

First, if a dipole speaker and its image are very close and back-to back, there will be a lot of cancellation and loss of bass.

Second, if there is a gap between the speaker and its image, then they will be in phase at some frequencies and out of phase at others, resulting in a comb-filter like frequency response with ripples. The peaks occur every c/2x, where c = is speed of sound, x = distance speaker is placed from the wall. For speakers 1 metre from the wall, the peaks occur every 170 Hz.

Third, you can exploit the figure-eight polar response of a dipole to lessen the cancelation effects by having the speakers toed in. If the planar speakers are placed at 45 degrees to the wall, the cancellation/reinforcement effects are minimized.

Fourth, distance helps, not only because of the delay and psychoacoustics, but also because speakers further away sound quieter, so the degree of cancelation/reinforcement will be less.

Fifth, you can reduce the reflection by putting absorbing material on the wall. Impossible at low frequencies but effective in the vocal range where your ears are most sensitive.

Sixth, when you start including reflections from the floor, ceiling and other walls, its all a mess, but in general, avoid corners. Floor-to-ceiling line-source speakers are less complicated , and tend to suffer less from corner effects.

So, my ESLs are about a metre from the walls, toed in, and have curtains behind them. The bass response is noticeably improved if I pull them further out from the wall but she who must be obeyed objects.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.