Xstat vs. older panel output

[Rich]

Ok, ever since ML came out with the smaller panels I have been a little skeptical of their claims that the smaller panels are more efficient and have greater output than the older larger panels. So, since I was a little bored today and wasn't in a mood to take pictures of my system, I decided to test this theory in a very non-scientific way.

I have a pair of Ascents and a pair of Summits. I counted the rows and columns of the stator panel holes to figure out how many holes each one had in its stator (counting only the rows/columns that were not blocked by the sidewall construction) and then measured the diameter of the holes in each type of speaker. I don't have caliper measuring tools, so this was a very inexact eyeball measurement using mm. From this, I calculated the total actual panel output area for each in sq. cm based on number and size of holes (figuring area for the circle and then multiplying that by number of circles in each panel). Since the Summit is equal to the Prodigy, and I don't have a Prodigy, I extrapolated the output area for the Prodigy based on the difference in size of the Prodigy panel from the Ascent panel. For reference, the Summit panel has a total area of 497 sq. in. and the Prodigy panel is around 750 sq. in. So the Prodigy panel is about 50% larger in area. Now, I will be the first to say that this was all done quickly and with very basic tools. I am fairly confident in the math, but the measurements could have been much more precise. I'm sure I'm off. Maybe way off. I don't know how badly I'm off and whether I am off equally between panel styles. So this is highly unscientific. But I think I got some basic ballpark figures to compare.

So is ML feeding us a bunch of marketing BS or are they being real? According to my very basic calculations, the Summit has about 50% to 60% more output area than the Prodigy! Even though the Prodigy panel is about 50% larger by total area. Wow! So ML is correct in saying they are more efficient and have greater output capacity. The next question this raises to me is: what does this do for the sound? Obviously, it allows them to make the speaker much smaller and still produce greater output while being easier to drive. Hence the Summit's 1 db better efficiency. Also, it may be because of the smaller panel that the Summit has a slightly higher crossover point than the Prodigy. But what does it do for the sound? I have no idea, but my guess is it increases accuracy, dynamics and transient response. Which is something all ESLs do great, but the new generation of MLs seems to excel at. However, my concern has always been that the smaller panel affects the perceived size of images and soundstage. I don't know if this is true, but I suspect it may be. There may be other positive and/or negative impacts of a smaller, more efficient panel that I haven't even considered.

Since I always think that more of a good thing has to be better, my final question is: how good would an ML speaker sound that used the newer panel construction but was made the size of the old Prodigy panel (appx. 14.5" x 48" vs. the Summit's 11.3" x 44"), or just over a third more radiating area? I have no idea, but I sure would like to find out. :rocker: I am curious if ML has experimented with larger panel sizes using the newer panel tech and, if so, what they discovered. Would they be even a little more efficient? Would they allow you to lower the crossover point? Would they affect imaging and sound staging in a positive way? Lots of possibilities.

Edit: For comparison sake, the Sanders Sound Model 10 has a panel that is 15" x 42" for a total area of 630 sq. in. and doesn't use the metal stators, so probably has an even greater effective radiating area than an ML panel the same size. They are 94 db efficient vs. Summit's 92 db, and they have a default crossover point of 172 hz. vs. Summit's 270 hz. They also go lower and higher than the Summits in frequency range. And they cast a hell of a soundstage with great imaging. But they lack the curved panel, so the sweet spot is narrow.

What do you guys think? And if anyone from ML wants to chime in with info or ideas, that would be great too.

 

[beanbag]

This is a question better asked on the DIYAudio ESL forum.

To answer some of your questions:
My Vistas were not louder than my crusty old Scenarios, and maybe even quieter, although I didn't compare too much. I recall some other review of the XStat ML also disproved their claims of higher sensitivity.

The hole size has a very token effect on how much sound gets through because the wavelengths are much longer than the hole size. Similar to how cone speaker cloth covers don't really block the sound by very much. Also, once the membrane pushes outwards, that extra air has to go somewhere. (It's not like the stators absorb the sound)

"accuracy, dynamics and transient response" : I don't see what the crossover point has to do with any of these, as long as you have a good crossover.

If you poke around enough on the DIYAudio forum and read some of the ESL construction and theory guides, you can see that there are a couple of different schools of thought:

Sander's is tall wide flat panel that maximizes beaming for best sound in sweet spot.
Tall skinny panel for approximation to cylindrical line source.
Small short panels for best directed sound and some dispersion.

You'll notice that tall wide curved panel is not on this list. That is probably because even though ML likes to say 30 degree dispersion, the curve actually causes high frequency "lobing", which in a way is even worse than beaming. Also curved panels are hard to make, and therefore not DIY friendly.

Also, some of the more knowledgeable builders don't find it a good end goal to push the xo frequency as low as possible. I forgot why, but whatevs.

http://www.diyaudio.com/forums/planars-exotics/243508-building-large-curved-esl.html

 

[MPS]

Lower crossover frequency to prevent saturation requires beefier signal transformer, this adds cost of manufacturing. Also acoustic cancellation reduces efficiency unless panel is made bigger, again adding cost of manufacturing.
Panel hole size is just one parameter in electro-acoustical transformer which ESL-panels are. You can't judge by one parameter only..

 

[Rich]

Hi guys. Thanks for the comments. This is the type of discussion I was hoping to get started with this post. I wanted to respond to a couple of comments Beanbag made.

The hole size has a very token effect on how much sound gets through because the wavelengths are much longer than the hole size. Similar to how cone speaker cloth covers don't really block the sound by very much. Also, once the membrane pushes outwards, that extra air has to go somewhere. (It's not like the stators absorb the sound)

I'm not sure I buy the concept about frequency wavelength being smaller than the diameter of the holes and therefore doesn't effect how much sound gets through. Comparing metal screen with holes drilled in it with speaker cloth is apples and oranges. If it was solid metal, it would definitely impede sound transmission. So it makes sense that the more porous you make it (i.e. more holes per unit area), the more the sound can freely pass. After all, the diaphragm has to move air to create the frequencies. Fewer holes means more resistance to that air movement. And if the decibels go up, you have to move even more of that air to produce the higher spls. So surely the holes have an effect on that. By the panel basically being more porous (less resistant to air movement), it has to be easier for the speaker diaphragm to do its thing at higher spls with less effort. Resulting in a more efficient speaker with faster transient response and greater output. Again, these are just my thoughts and I am certainly no acoustical physicist. Would love to hear more thoughts on this.

"accuracy, dynamics and transient response" : I don't see what the crossover point has to do with any of these, as long as you have a good crossover.

I think you misunderstood what I was saying. I wasn't saying the crossover point had anything to do with any of these things. I was saying that by making the panel more efficient and able to move more air more freely, that you may get greater accuracy, dynamics and transient response as a result. I don't know that this is true, but this is just my gut feeling. I think the lower crossover point is just a byproduct of having a bigger panel that can move enough air at a lower frequency level to be able to have a lower crossover point. And I think a lower crossover point in a hybrid ESL is almost always a good thing when it comes to the realism of the sound.

Also, some of the more knowledgeable builders don't find it a good end goal to push the xo frequency as low as possible. I forgot why, but whatevs

I understand that there are limitations and points of diminishing returns to this. However, I think if you can bring the crossover as low as is reasonably possible, you get much more accuracy in more of the important human vocal range. I would rather have my ESL crossover at 170 hz. vs. 270 hz. if I had the choice. I agree with MPS that some of the limitations of the ML are probably due to cost of manufacturing constraints. The crossover point may be one of those things. He also makes a very good point that panel hole size is just one of many variables which affect acoustical production in an ESL. I was just curious about this parameter because of MLs marketing about how the Xstat panel increased the efficiency and output of their panels, allowing them to be smaller but have the same or more output.

Just for completeness, here is ML's official marketing take on it:

"Advanced MicroPerf stator technology gives MartinLogan's latest-generation CLS™ electrostatic transducer almost twice as much exposed diaphragm surface as a traditional electrostatic panel of the same size. This dramatic increase in effective radiating area translates to higher efficiency and an even more effortless, unlimited dynamic presentation. The XStat pulls no punches!"

 

[edwinr]

\

"Advanced MicroPerf stator technology gives MartinLogan's latest-generation CLS™ electrostatic transducer almost twice as much exposed diaphragm surface as a traditional electrostatic panel of the same size..."

As you suggest, Rich, I reckon the "...effective radiating area..." and the "...exposed diaphragm surface..." comments point to the more porous panel being the primary reason for ML's claims. Better build quality and adhesives plus tighter manufacturing tolerances would all contribute to more efficient panels.