2 Sets of ESL's for bigger sweet spot

MartinLogan Audio Owners Forum

Help Support MartinLogan Audio Owners Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Davioj

Well-known member
Joined
Oct 27, 2019
Messages
57
Reaction score
20
Location
On. Canada
Just wondering if you use 2 sets of ESL's for front stage, 2 left and 2 right along with the center, seen a fellow have this set up on a post not here and was wondering if anyone has had experience with this type of set up and does it work for a wider sweet spot. Could not find any posts on this type of set up'

Thanks
Dave
 
Roughly 70% of the sound is from the center channel. Hence the bulk of the sweet spot is created from that speaker.
I would say work on improving that for a wider sweet spot.

Good Luck
Brad
 
You will very likely end up with comb filtering where the waves from the two speakers interfere with each other and you have dead spots and loud spots as you move along the axis.
 
Maybe if they are placed next to each other the curve linear becomes 60 degrees instead of 30 - in that case I don't thing you would get any interferences - then this would also be an issue from the Neolits having a wide panel. For the lower notes two speakers placed close to each other this would also not be an issue - if so, would there not be a big issue for their Masterpiece centers where the woofers are well apart?
I find with only one speaker on each side, I have a fairly OK wide sweet spot (but also using the Dirac in the wide set-up configuration).
Anyway, Interesting if anyone has tried this. Maybe you can audit it with a dealer?
 
Last edited:
Roughly 70% of the sound is from the center channel. Hence the bulk of the sweet spot is created from that speaker.
I would say work on improving that for a wider sweet spot.

Good Luck
Brad
Brad I am already using the focus C18 so the centre is good, how can you improve that except to go to the illusion but I am using the ReQuests for L & R and I been reading the illusion matches better with the 15A or 13As a costly upgrade and not much of an improvement from ReQuests. May be I'll put my rear Ethos up front and try it and see how it sounds , Requests are bi amped and the Ethos cannot be bi amped see how 2 channel music sounds. I was thinking of another set of Ethos to do this with but then it might not work well with HT.

Maybe if they are placed next to each other the curve linear becomes 60 degrees instead of 30 - in that case I don't thing you would get any interferences - then this would also be an issue from the Neolits having a wide panel. For the lower notes two speakers placed close to each other this would also not be an issue - if so, would there not be a big issue for their Masterpiece centers where the woofers are well apart?
I find with only one speaker on each side, I have a fairly OK wide sweet spot (but also using the Dirac in the wide set-up configuration).
Anyway, Interesting if anyone has tried this. Maybe you can audit it with a dealer?

Nuri that's what I was wondering if it would increase curve linear even if it did to say 50 degrees which would still give theoretically a better sweet spot because right now the sweet spot with 2 channel is in a fairly narrow spot where my listening chair is, the wife's seat is a little out side of it but she don't really care but I do so that's why the question, maybe someone has done this. What is the cost of dirac, I only have ARC right now.

Dave
 
How would one measure the size of the “sweet spot” if experimenting with multiple sets of panels? I understand it’s a “know it when you hear it” kind of parameter, but the analytical in me likes to quantify.

It would not just be db levels, as imaging and other factors probably weigh in when defining the spot. Admittedly, I have done no research to find if a method already exists. Thoughts?
 
I do not think (passive) Bi-Amping will help or rather the Ethos woofer has its own amplifier (somehow Bi-Amped one could argue). In general, I do not find Bi-Amping very useful unless you have passive speakers whereby where the big power draw is for the woofer. This would mean power reserve to the amp driving the panels thus reduces the load created by the low impedance for high frequences. If you add a sub to equation and configure the LS as "small" say cross-over at 80 Hz or so you would obtain almost the same.

Yah,I finally gathered the idea. I looked at the geometry and it would result in a certain gap between the two panels - if that would mean anything I don't know. But for 2ch would you not always have the issue you describe as the sound level of the nearest LS always will be louder than the further? Maybe not as much as the ELS - I wouldn't know. In my setup I can move my head +/- 1.25 m (4 feet) without the sound changing much to my ears, except the rear channels, clearly suffer from being much closer to one ear than the other.

For Dirac, it came with my Arcam so I do not know the cost. But it seems to work. Their latest development is Dirac ART (Active Room Treatment) that uses all LS to compensate for the issues for each individual LS. Storm Audio has this as the first manufacturer (and only one for now) to offer it as far as I know. I haven't heard it yet, only read some reviews and Dirac's own promotion. But I am sure it will be released for all amps having the needed HW.
 
the wife's seat is a little out side of it but she don't really care
I think most of us are in that boat. ML determined the optimum dispersion pattern in many situations is 30 degrees. This is a compromise between optimal listening area--"sweet spot"--and minimum interaction with walls, floor and ceiling, except for that pesky back wave which requires breathing space behind. You'll lose some of the ESL virtues if you get too greedy with sweet spot. I am prejudiced, having put up with the far narrower sweet spot of the Acoustat 1+1's for many years.
 
The traditional reason for doubling up ESL panels is better bass performance. This becomes largely irrelevant with hybrids, unless you're prepared to use external crossovers and lower the crossover point.

Dispersion is the age-old problem with ESL's, especially full range or nearly full range ESL's. KLH's solution was to design their panels so that the best frequency response was off-axis. You would aim the beams, preferably from two panels per channel, away from the listening area. Many designers went for polygonal designs including Janszen (in his add-on tweeters), Acoustat and Sound Lab. These tended to suffer from multiple beams. The ML curved panel was, IMO, absolutely brilliant. They have always designed for a 30 degree dispersion pattern. I have enough of an engineering background to respect designers who know more than I do about a given topic, and stick with their design decisions.

Also, I don't think you would get comb filtering since ESL panels are far, far from point sources. You would, however, potentially get more side wall reflections except in a very large room.

I once heard at a show cylindrical ESL's. Can't remember the company now, and can't find it with a reasonable search. Probably no longer around. They did not have any of the ESL magic, in my opinion. They were probably not cylinders, actually, but poly-polygons. Maybe someone can recall.
 
Maybe if they are placed next to each other the curve linear becomes 60 degrees instead of 30 - in that case I don't thing you would get any interferences - then this would also be an issue from the Neolits having a wide panel. For the lower notes two speakers placed close to each other this would also not be an issue - if so, would there not be a big issue for their Masterpiece centers where the woofers are well apart?
I find with only one speaker on each side, I have a fairly OK wide sweet spot (but also using the Dirac in the wide set-up configuration).
Anyway, Interesting if anyone has tried this. Maybe you can audit it with a dealer?
How would you get the panels of Masterpiece series speakers together and at 30 degrees between them, considering how far back the woofer enclosures extend?
 
As mentioned in my previous post, after considering the geometry it would result in a gap between the panels, so right you are. But as also mentioned by others the 30 degree angles is surely not a coincidence but a fine compromise.

My setup is 12 feet between the fronts and 18 feet to my listning position which gives about a 6-7 feet wide sweet spot I'd say.

I by the way recall something about the ideal speaker would be the entire wall, guess in many sections, but that's another story.
 
I second Leporello's point that getting greedy with the sweet spot will sacrifice some of the virtues of limited dispersion. Essentially; you can have a wide sweet spot or a precise image but physics doesn't allow having it all. You can only choose a compromise that best suits you.

What you might try, with only two speakers, is crossing their beams. That is; aiming them at a point several feet in front of the listening portion so that their beams cross and separate in front of the listening position.

If you try it, let us know how it sounds.

If you're not irrevocably married to curved perf-metal panels, there are other options for tuning the dispersion pattern, but you'll have to roll your own:

I've built many ESLs. My earlier versions used un-curved flat panels with perf-metal stators. These gave big slam and truly magical imaging at their focal point, but were directional to the extreme--- basically one-person "head-in-a-vice" speakers. A friend described them as "remote head phones".

Then I started building flat-panel wire-stator ESLs with symmetrically segmented stators. Segmentation works something like what Quad did with their ESL-63 except that where the Quads simulate a point source projecting a spherical wave-front, segmented wire panels function as a line source projecting a cylindrical wave front.

With segmented wire panels, the dispersion pattern can be tailored by varying the number and widths of the individual wire groups (i.e. more/narrower wire groups give wider, smoother trending dispersion).

I once built a pair using TIG welding rod stators, configured with switch-selectable wide and narrow dispersion modes. The switch mode feature was fun for a while but the novelty wore off pretty quickly because I had to power down to switch modes (lest arcing destroy the switch) and then re-EQ the panel each time, because the two modes had different response curves.

I still use segmented flat panel wire stators, which I consider the current state of the art, but I abandoned the switch-mode feature as impractical.

My current ESLs don't have quite the slam and pinpoint imaging of my old unsegmented flat panels, but their dispersion is exactly tailored to my taste, they look great, sound wonderful and they don't lock my head in a vice :)

(I could hook you up with a pair)
 
Last edited:
I second Leporello's point that getting greedy with the sweet spot will sacrifice some of the virtues of limited dispersion. Essentially; you can have a wide sweet spot or a precise image but physics doesn't allow having it all. You can only choose a compromise that best suits you.

What you might try, with only two speakers, is crossing their beams. That is; aiming them at a point several feet in front of the listening portion so that their beams cross and separate in front of the listening position.

If you try it, let us know how it sounds.

If you're not irrevocably married to curved perf-metal panels, there are other options for tuning the dispersion pattern, but you'll have to roll your own:

I've built many ESLs. My earlier versions used un-curved flat panels with perf-metal stators. These gave big slam and truly magical imaging at their focal point, but were directional to the extreme--- basically one-person "head-in-a-vice" speakers. A friend described them as "remote head phones".

Then I started building flat-panel wire-stator ESLs with symmetrically segmented stators. Segmentation works something like what Quad did with their ESL-63 except that where the Quads simulate a point source projecting a spherical wave-front, segmented wire panels function as a line source projecting a cylindrical wave front.

With segmented wire panels, the dispersion pattern can be tailored by varying the number and widths of the individual wire groups (i.e. more/narrower wire groups give wider, smoother trending dispersion).

I once built a pair using TIG welding rod stators, configured with switch-selectable wide and narrow dispersion modes. The switch mode feature was fun for a while but the novelty wore off pretty quickly because I had to power down to switch modes (lest arcing destroy the switch) and then re-EQ the panel each time, because the two modes had different response curves.

I still use segmented flat panel wire stators, which I consider the current state of the art, but I abandoned the switch-mode feature as impractical.

My current ESLs don't have quite the slam and pinpoint imaging of my old unsegmented flat panels, but their dispersion is exactly tailored to my taste, they look great, sound wonderful and they don't lock my head in a vice :)

(I could hook you up with a pair)
Acoustat also went with the segmented concept for a while, with their "SPECTRA". I don't remember what all of the acronym stood for, but spatially something, electronically curved. Gordon Holt reviewed them in Stereophile. I also don't remember what he thought of them.

With the Acoustat 1+1's I always found, in every room I used them in, that the best compromise was to cross the beams slightly behind the listening position. Aiming the beams straight at the listener is rarely optimal, or recommended, for any ESL. ML always says "listening to the inner thirds" of the panels is the best thing to try out at first. Recommendations like that should never be read as set in stone. Listening to the outer thirds is certainly worth trying out in your situation.
 
How would one measure the size of the “sweet spot” if experimenting with multiple sets of panels? I understand it’s a “know it when you hear it” kind of parameter, but the analytical in me likes to quantify.

It would not just be db levels, as imaging and other factors probably weigh in when defining the spot. Admittedly, I have done no research to find if a method already exists. Thoughts?
Measurements rarely tell the whole story, but polar response has a lot to do with both imaging and sweet spot. You could measure polar response at home in your room using REW, but it would be tedious, time consuming, and something of a mess, due to the stochastic nature of any in-room response. It helps to have an anechoic chamber, a rotating stage, and software such as LabVIEW, to automate the procedure.
 
Acoustat also went with the segmented concept for a while, with their "SPECTRA". I don't remember what all of the acronym stood for, but spatially something, electronically curved. Gordon Holt reviewed them in Stereophile. I also don't remember what he thought of them.

With the Acoustat 1+1's I always found, in every room I used them in, that the best compromise was to cross the beams slightly behind the listening position. Aiming the beams straight at the listener is rarely optimal, or recommended, for any ESL. ML always says "listening to the inner thirds" of the panels is the best thing to try out at first. Recommendations like that should never be read as set in stone. Listening to the outer thirds is certainly worth trying out in your situation.

My speaker are less directional than other ESLs but I too aim my speakers to cross slightly behind the listening position, which I believe provides an optimal compromise between imaging and width of the sweet spot.

It's been pointed out that a wider sweet spot is a trade-off with imaging. Even so, I interpreted the questioner's post as giving priority to a wider sweet spot.

It seems to me that, without adding additional speakers, the sweet spot can be widened by either crossing the beams even further behind the listening position, or crossing them in front of the listening position. Both will compromise imaging, but if the priority is widening the sweet spot, it would be interesting to try it both ways and compare the results.
 
Just wondering if you use 2 sets of ESL's for front stage, 2 left and 2 right along with the center, seen a fellow have this set up on a post not here and was wondering if anyone has had experience with this type of set up and does it work for a wider sweet spot. Could not find any posts on this type of set up'

Thanks
Dave
this is very specific setup (not for electrostats) where you have 5 screen speakers - used for very wide screens. And what you do, is you have L-LC-C-RC-R, so basically adding 2 centers.

Another option is to use (0 dg being center) 30 dg Left Front and 60dg Front Wide. You can then use Front Wides with delay and lower gain to get “wide stereo” , but it will not improve hour sweet spot, rather opposite. On top, I would not do it with dipoles, becuase with all the reflections it will be hot mess probably, phase issues, lobing, cancellations. Not worth it.

It hink best option for 2Ch music is to use Auro 3D for upmix, if you have anything like 7.x.4 setup.
 
Well I just through out the question to see if anyone had tried this, I listen and sit in the best seat and the wife don't care so maybe i'll just keep my funds and look to put them somewhere else like an OLED TV, just have to wait till the price comes down, can't justify spending $6000 on a TV already have an 85" LED that is pretty good Samsung 90 series, thanks for all the imput though.

Dave
 
Back
Top