Cabinet for electrostatic mini panel

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi,
I have these nice little electrostatic mini panels. I like them as dipoles, very clear down to a respectable 250-300Hz, but I prefer them high-passed at 400Hz. Because between 250-400 hz I hear that I am missing something.
Sensitivity: 90dB at 2.83V.
Membrane size: 135 x 485 mm.

I am thinking about making a sealed or ported enclosure for them, to check if I can get down to 250Hz, because the 15" woofer is not great at midrange.

Almost everyone advice against using an electrostatic speaker other than as a dipole, but I wonder if many actually tried it. Is there anyone with some experience that would like to share it, before I just build one from guessing?

One guess for a sealed cabinet would be to use the lowest wavelength which could be 300Hz, which translates to about 110cm in length. Then make the back chamber a third of this which is about 36 cm. So the back chamber will be 13.5 x 48.5 mm and 36cm deep. For ported I have no idea.
 

Attachments

  • 005.jpg
    005.jpg
    70.3 KB · Views: 355
Last edited:
Hi,

Forget boxes for such a driver, it will simply sound terrible.

What your missing below 400Hz and possibly higher is
the transition from line source to omnidirectional as the
wavelength approaches the panels length the efficiency
will drop off. Can be compensated by baffle size and the
right baffle hump to a degree, but you want your midbass
line to maintain line source characteristics for quality.

Open baffle, with a multiple line of smaller midbass units next
to the panels and decent dual * subwoofers is the way to go.

Trust me, I know, a 0.5m line will never x/o to a single
bass unit without sounding simply wrong. I have a pair
of Celestion 5000's, and boy do I know the problems.

rgds, sreten.

* They can be mono, one amplifier to drive two,
but you need two properly locates subwoofers.

As a further aside at 135mm wide treble dispersion
will suffer, and you could do worse than a line of
supertweeters as well. See the $1 specials at :
Speaker Stuff Audax, apparently and quite good.
 
Last edited:
Hi,

Forget boxes for such a driver, it will simply sound terrible.

What your missing below 400Hz and possibly higher is
the transition from line source to omnidirectional as the
wavelength approaches the panels length the efficiency
will drop off. Can be compensated by baffle size and the
right baffle hump to a degree, but you want your midbass
line to maintain line source characteristics for quality.

Open baffle, with a multiple line of smaller midbass units next
to the panels and decent dual * subwoofers is the way to go.

Trust me, I know, a 0.5m line will never x/o to a single
bass unit without sounding simply wrong. I have a pair
of Celestion 5000's, and boy do I know the problems.

rgds, sreten.

* They can be mono, one amplifier to drive two,
but you need two properly locates subwoofers.

As a further aside at 135mm wide treble dispersion
will suffer, and you could do worse than a line of
supertweeters as well. See the $1 specials at :
Speaker Stuff Audax, apparently and quite good.

Thanks, but I asked for advice on how to make them non-dipoles, i.e. how it calculate a sealed or ported cabinet.

I really want to try them as non-dipoles, since the rest of my system is non-dipole, and it would probably make them go down to 250 Hz properly. I have been told by the manufacturer that below 400 Hz cancellation from the rear wall may occur, but I think there is more to it than that. A cabinet would theoretically help it to go deeper, but I am expecting to lose some clarity and gain some distortion.

There is nothing wrong with the treble, but I have a Raal for that anyway and it does an even better job.

The panels will not be my only midrange channel. I want to use it go get a more versitile sound. Horns are good at one thing, and electrostatic another. But combining horns with dipoles will sound a bit wrong. I tried it before with a Saba Greencone 8" vintage widerange. It also had great clarity above 400Hz as a dipole, but in a sealed cabinet it was playing to down to 150Hz. It did not sound perfect, but my girlfriend liked it, mostly because of the green colour, I suspect. :)

Another example was my attempt to make a dipole ribbon tweeter of the same height as the mini panels into non-dipoles with a back chamber. Its aluminum ribbon was approx 45cm x 1" wide. I did get deeper upper midrange from the ribbon tweeter when adding a back chamber that was 45 x 2.54 and 3.5 cm deep. I used the same idea, when guessing the appropriate size of the back chamber. I wanted it to play down to 3000Hz and this is how I was told to calculate it.

3000Hz = 11 cm wavenlength => 11/3 = 3.5cm

But the aluminum ribbon was too thick and heavy to sound as good as the Raal ribbon tweeter, so they are now stored away until I have the time to manufacture a new ribbon and order a custom made transformer. Since they have the same sensitivity as the electrostatic panels I am now curious if I can use them on the same amplifier channel with a passive crossover. They are essentially like a Raal Lazy ribbon, with a great need of modification.

In any case, making a dipole into a non dipole is not always such a bad a idea, and you don't know how it will sound until you try it.
 
Last edited:
Thanks, but I asked for advice on how to make them non-dipoles, i.e. how it calculate a sealed or ported cabinet..

You talk as if a box brought some benefits. Aside from sequestering the rear wave, which is what you'd be talking about if you had an elementary grasp of the subject, there's not much good about a box behind the cell and various that's bad.

But sticking with your idiosyncratic line of thought, what you want is an open back box filled with sound absorbent. That will provide you with what you are asking for.

Ben
 
Hi,

one of the most prominent advantages of an esl is its mass-freeness.
A classic CB modifies the mass-spring system of a dynamic driver so that the typically insufficient LF-response is pimped to an acceptable response.
With an ESL the tuning frequency may easily become too high, which on the other hand means, that comparatively huge cabinet volumes are required, prefferably not closed, to not raise the fundamental resonance too high.
A second point is that one needs to ensure that no reflected sounds pass the ESL membrane from the cabinet's inners.
This requires efficient damping of the panels rear sounds.
Mr. Walker suggested a rather shallow deep cabinet with decreasing crosssection, ending in a small slot, dampened with loads of damping material.
In other words, a classic transmissionline.
The easiest would probabely be to build a stuffed-with-wadding, trapezoidal shaped casing with an open backside.
The required depth to be effective will be large though, maybe requiring to roll up the casing (similar to B&Ws Nautikus) or to fold it.
As practical Transmissionlines behave rather similar to BR-Boxes the dimensions of the TL need to be tuned to fit the panel-cabinet fundamental resonance (built-in resonance fb).

You could maybe solve the problem of acoustic phase cancellation (that starts already below ~1kHz with such a thin panel), but You'll need to equalize the panels response anyway.
So, equalizing slightly more but driving the panel as dipole will imho lead to a more satisfying result with less building effort.

As sreten pointed out already, it'll be almost impossible to xover a 15" to such a small, thin panel without massive problems regarding homogenity of sound.
Seamless transition from woofer to panel is imo impossible here, regardless of the panel beeing boxed or a open dipole.

jauu
Calvin
 
Thanks you for the information. The manufacturer had not provided any such data,but I can ask. Is there any way I can simulate this as a regular driver? I could get the Dayton Audio speaker measuring device to tune a TL or BR correctly.

I am thinking that a sealed design might put more strain on the thin membrane.

Again, for the third time, I already know that many people prefer to use elctrostatic speakers as dipoles. No need to point out that I am a beginner at the English language, audio engineering and electrostatic drivers. I stiIl would still like to try this and see how it turns out.
 
Hi,

unfortunately no.
The sim programs demand a set of the TSPs (Thiele-Small parameters) which You can't measure with ESLs as with dynamic speakers.
Basically You just need to know the fundamental resonance of the panel when built into the cabinet.
Knowing the fb and the volume of the enclosure You then use the formulas for a Helmholtz-resonator to design the BR-tube.
But You can only measure fb when built-in .... so its rather like a try and error game.
So it'd probabely best to just build the enclosure thin and deep with an detachable backside and stuff it fully and see what comes out ;)

jauu
Calvin
 
Last edited:
How about this enclosure made for the Feastrex 5" as a starting point? It would follow the edges of the panel, and then continue behind, depth unknown, and inside shape unknown.
I was trying to think hard what could work but I have no idea with such a driver. Nautilus could work, but so much work. I might try it if I like it in a regular box where I test ported and sealed. I was even thinking about back loaded horn like a Frugelhorn, but it might sound weird and take up space. I just need 250Hz, which I already get with EQ. To get a 5" high efficiancy speaker driver to play down to 250Hz does not take much at all. A few litres maybe.
 

Attachments

  • f30cro2.jpg
    f30cro2.jpg
    436.1 KB · Views: 309
Last edited:
Ive only built esls that were used as dipoles, but I have heard a set of modern Jantzen speakers with esl mid/tweet and cone woofer, both in a box. Sounded very, very good, clean and uncolored, (measured like that too) so it can be done.

Right. I forgot that in days of yore, that was the mode for ESL panels like any other driver. With the terminals and other hardware on the backside, made sense. Then there appeared the KLH9, a giant tower that promoted dipole operation.

Funny, for DIYaudio builders, only dipoles. Besides the extra power that can output into the room, the ambient sound of dipoles is certainly grand.

But so too was/is the crystal clear treble from Jantzen boxes.

Ben
 
Don't Baffle

ELS speakers are dipole and should not be put into a cabinet. Positioning the speaker to a back wall should begin at 1/7 of the room length and tweaked using a swept oscillator to measure the peak to dip measurements using a microphone at the listening position. This will allow you to quickly optimize the speaker location.

I suggest using the REW software and calibrated microphone to measure the frequency response of each component speaker. I used a 4-way crossover to optimize the signal to each of the speakers.

I used LTspice to model all of my crossover filters (using Riley-Linkwitz filters). I also used LTspice to measure group delay of each of the filters and found that I had to use two all pass filters to cross over from the transmission line woofers to the ELS panels to avoid a 90 degree phase difference.

I hope this will point a few of you in the right direction.
 
Footnote: not just "dogma", as suggested above, that leads nearly all of us to favour dipole ambient sound. Actually, rather the dismissal of point-source wannabee-engineering dogma that leads us there.

Well young folks, I remember well the early Jantzens (spelling?). Striking sound indeed. I think his issue was hiding all the HV circuitry and the garbage look of his ESL backsides (with damping from paper towel material?). Also, how else could he pass safety inspections?

But the key point to remember is that early sealed ESL boxes had multiple cells spreading the beam. So, you can make a sealed box (or better, leaky box) but not if using a single cell.

B.
 
Last edited:
Thanks Ben, good idea. I will put all the electronics inside. There are already a few subs around to take care of the low end, and under the ESL panel will be a mid-bass driver to cover 100 Hz to whatever I feel like.

The spelling is "JansZen". Some of their current speakers use single cell - like JansZen Electrostatic Active Hybrid Speakers.

Did you get the MFB subwoofers working well? I may get into this if I get in troubles transitioning to ESL panels.
 
Did you get the MFB subwoofers working well
Off topic. Yes, for several decades with a Klipschorn bass (prolly speaker that needs it the least in the world). If my clients will let me retire, I may make progress on a new accelerometer method.

On-topic. While I've never fussed about "matching" ESL sound to woofer sound, when you think about the aural charms of ESLs, MFB sounds like the right match.... and why those of us who favour dipole mounting think it is more natural to ESL sound than a box.*

B.
*if you really want a box, stuff with wool and leave the back off.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.