Open baffle speaker design

[rebbiputzmaker]

Stacked with subs and rear wall heavily damped was the best iteration I have heard, I am sure in the 1950’s they were really something exceptional.
Keith

Yes, and Yes!

 

[Fitzcaraldo215]

What's the matter with your head, did your mother drop you on it?
Panel arching on early ESL's isn't anyone's fantasy, it's a known fact that has already been documented here in just a few minutes.
I personally owned a pair of speakers with Janszen tweeters and watched one of them sizzle to death around 1976.
If your going to run your mouth make sure you do it about something you have knowledge in. Otherwise you only end up making yourself appear the fool.

Not on topic, but, yes, electrostats once a long time ago WERE known to arc, especially the Quad 57s. But, that is stale old news. Sorry for your loss in 1976, but how is that relevant today? Electrostat technology has not just stood still and manufacturers have not just ignored the well known problems of old Quads in successive newer designs.

I have used Martin Logan full range, then hybrid 'stat dipoles since 1986, plus Fulton J-Modulars with Janszen 'stat tweeters for 12 years before that, driven by a then powerful Dynaco 400 amp, and I have never experienced any arcing whatsoever, as in zero. During that time, one set of early ML CLS I panels was replaced, but that was due to a manufacturing defect, not arcing. ML was very supportive to my situation then, and I have been a loyal buyer of their products ever since. I have 7 of them in my current Mch setup. Like everything, they ain't perfect, but I continue to enjoy them greatly with exceptional reliability. Incidentally, my main MLs are driven by a Spectron Class D amp with over 900 wpc into 4 or 2 ohms.

I have blown out a few speakers in my lifetime, but never electrostats. I was aware of the issue took careful note of the ML speakers' design to avoid arcing when I first bought them way back.

Personally, like our friend @RayDunzl, and, though it be heresy in this forum, I very much like what dipoles can do, especially electrostat hybrids. But, of course, I cannot speak to arcing or reliability issues for all electrostats. But, it seems an historical issue, not an ongoing problem.

 

[Cosmik]

The '63s I've heard give an image and soundstage that is different than what I get from forward-firing box speakers. If that's not your preference, fine. For me, that presentation gets more of my "live music" synapses firing. The '57s do have more volume limitations, but have a "rightness" to my ears on vocals and acoustic strings. I have not heard anything quite like this from DSPed box speakers. Again, none of this is "real," it's all illusion-creation, so we really are down to preference rather than objectivity- speakers and rooms are NOT the same thing as amps and DACs. When we compress a complex 3-D soundfield into two 1-D datastreams, then plop two transducers into a different space, the data loss is profound, and we can only deal with hedonics, not simply measurables.

It seems I differ from most people here in thinking that speakers are like amps and DACs: there is a basic 'ideal' that we can get reasonably close to these days. If I had never heard 'the live music' thing from DSP'-ed box speakers but had heard something special from dipoles, panels, etc. I would hesitate in saying that: maybe the 'data loss' is too great. But the illogicality of that position would be that by fiddling around by making holes in a speaker, removing the box, adding extra drivers and so on, I can (re)create data i.e. information. That's impossible.

For sure, sticking extra drivers around the room, under your seat or on the back of the speaker and driving them in anti-phase, or delayed by 50ms and shifted in pitch a bit, might give an amazing effect that triggers some synapses. But all the time..?

If you stick a seashell to your ear it sounds like a live recreation of a seaside scene. Two gives an amazing surround-sound effect. But no information has, in fact, been created (just the opposite), and that immersive experience has a very limited repertoire.

 

[Fitzcaraldo215]

Stacked with subs and rear wall heavily damped was the best iteration I have heard, I am sure in the 1950’s they were really something exceptional.
Keith

Well, the Mark Levinson HQD system, with stacked Quad 57s, Decca ribbon and Hartley subs was someting of a sensation at shows in the '70's. I heard them then at a dealership, as did Gordon Holt, the very same setup. Neither he nor I were overwhelmed, frankly. And, JGH did not place the speaker on his recommended list or displace his then reference speakers.

 

[sergeauckland]

Here you go Serge - plenty more on the site, too.

http://www.linkwitzlab.com/Loudspeaker-Room/tests&measurements.htm

Here's a distortion measurement from my LX521s. I think I have posted FR and step response somewhere on the site also. I'll try and find them

https://audiosciencereview.com/foru...sion-thread.2615/page-6#lg=post-76334&slide=1



This is not really valid anymore IMO. DSP, inexpensive high quality Class D amps, a bass driver such as the SEAS unit in the LX521 with 28mm of linear travel (56 mm max) that doesn't cost the earth can make for a relatively inexpensive build - and you don't have to be a skilled carpenter to build a perfectly sealed box, either.

I've no doubt that they can be made to work, but again, I come back to my main point, why bother? What do OBs do better than conventional loudspeakers? Horns are efficient if mostly coloured, Reflex loaded loudspeakers can be more sensitive and/or go deeper than sealed boxes, albeit with some energy storage at LF, sealed boxes and (real) TLs have the most accurate LF at the expense of sensitivity and size. What do OBs do that's technically better than other methods of loading? Even using Class D, they still need very large amounts of power due to the bass EQ that's necessary.

To me they're like Valve amplifiers, in that they too can work well, but why would anyone bother when SS amps work so well? Other more conventional loadings work without the need for massive amounts of EQ and the consequently high power amplifiers, even if Class D.

S.

 

[RayDunzl]

My Martin Logan reQuests are 20 years old. The panel is 15" x 48". The woofer is a 12" sealed box.

The membrane is charged to 2500V (or thereabouts). They draw a watt or two from the wall to generate the charge. If you unplug them, they slowly lose charge, it's not noticeable for a while (several minutes).

The stators (front and back metal sheets) are driven by the amplifiers via a step-up transformer, ratio unknown, possibly 1:100 (or thereabouts). The differential musical signal on the stators attract/repel the membrane. The stators have an insulating "paint" coat on them. I keep a "sock" over them.

The loudest I have measured was 116.9dBz peak, using a short drum solo. It was exceptionally "too loud". Normal "too loud" here gives peaks of 105 to 110dBz.

---

I also have a pair of JBL LSR 308 (the daily drivers), and have had a pair of Infinity P-363 in the room.

Can you tell which is which (no EQ, 1/3 octave, position in room not exactly the same, but close)? Not quite level matched, measured at different times.

I listen to the JBL daily for "economy operation", and to the MartinLogans for more "critical listening", due to the difference in imaging (untreated room), and harmonic distortion . ML wins on both of those. They just sound "cleaner".

Here is an unsmoothed SPL - the ML is blue (foreground). I interpret the extra "hash" in the other two as an indication of the reflections off the walls and ceiling. The cancellations are very narrow, so, as The King of Audio says, "You can't hear that".

I say, "I hear something".

 

[Blumlein 88]

Should I mention the ESL-63s were designed to work as a point source. I think they called it a quasi-point source as it was directional. It was somewhat effective at this and had different directional and imaging characteristics vs other dipoles for this reason. The sound acted as if it emanated from a point 9 inches behind the actual panel. I've always thought using delay lines a floor to ceiling line source would be a very good design for a panel speaker. With modern DSP one might even control directivity as desired.

Like many things modern materials tech and experience fixed any serious arcing issues.

The low efficiency and bass impact hasn't been fixed so much. Using subs helps with that if done well.

 

[Blumlein 88]

My Martin Logan reQuests are 20 years old. The panel is 15" x 48". The woofer is a 12" sealed box.

The membrane is charged to 2500V (or thereabouts). They draw a watt or two from the wall to generate the charge. If you unplug them, they slowly lose charge, it's not noticeable for a while (several minutes).

The stators (front and back metal sheets) are driven by the amplifiers via a step-up transformer, ratio unknown, possibly 1:100 (or thereabouts). The differential musical signal on the stators attract/repel the membrane. The stators have an insulating "paint" coat on them. I keep a "sock" over them.

The loudest I have measured was 116.9dBz peak, using a short drum solo. IT was exceptionally "too loud". Normal "too loud" here gives peaks of 105 to 110dBz.

---

I also have a pair of JBL LSR 308 (the daily drivers), and have had a pair of Infinity P-363 in the room.

Can you tell which is which (no EQ, 1/3 octave, position in room not exactly the same, but close)? Not quite level matched, measured at different times.

View attachment 13440

I listen to the JBL daily for "economy operation", and to the MartinLogans for more "critical listening", due to the difference in imaging (untreated room), and distortion . ML wins on both of those.

JBL's dark red.
ML's blue.

 

[Burning Sounds]

I've no doubt that they can be made to work, but again, I come back to my main point, why bother? What do OBs do better than conventional loudspeakers? Horns are efficient if mostly coloured, Reflex loaded loudspeakers can be more sensitive and/or go deeper than sealed boxes, albeit with some energy storage at LF, sealed boxes and (real) TLs have the most accurate LF at the expense of sensitivity and size. What do OBs do that's technically better than other methods of loading? Even using Class D, they still need very large amounts of power due to the bass EQ that's necessary.

To me they're like Valve amplifiers, in that they too can work well, but why would anyone bother when SS amps work so well? Other more conventional loadings work without the need for massive amounts of EQ and the consequently high power amplifiers, even if Class D.

S.

Even though you may not agree with him you may find Siegfried Linkwitz's paper interesting and for me answers your Why bother? question. He also makes a case for why he thinks a dynamic driver dipole has better bass articulation than a sealed box.

http://www.linkwitzlab.com/IJAETv2n2a2-Linkwitz-1.pdf

In my opinion the LX521 has better soundstaging and imaging than most box speakers I have heard and also better than most large panel dipoles. They simply disappear in a way that very few speakers do. The bass drivers require no more power than say a Dutch & Dutch 8C.

 

[Burning Sounds]

Stacked with subs and rear wall heavily damped was the best iteration I have heard, I am sure in the 1950’s they were really something exceptional.
Keith

By rear wall are you meaning behind the speakers? - if so for a dipole it should be diffusive, not damped. The wall behind the listener should be damped.

 

[DonH56]

The wall behind my Maggies is heavily damped. Diffusion (unless very, very large/deep) won't help midbass and below cancellation. But of course opinions vary and it depends upon the room. Killing the back wave helps reduce comb filter effects and such so imaging is very precise but also costs some of the sense of "space" (ambience) in the sound. I played with both ways (and many others) over the years and always ended up with absorbers behind except when I had them in a large room, though at least some of the time that was because diffusers cost too dang much and I was too lazy to build them.

Heard an HQD system a number of times; one of those systems that sounded impressive to me at first but lost its allure with continued listening. Despite the big 24" woofers bass was not all that great in one system; the other was dialed in a bit better but still more "boom" than true deep bass. And was aesthetically-challenged, at least the ones I saw, though other speakers were worse IMO (I never did manage to appreciate the aesthetics of the original Wilson Watt/Puppy system, or for that matter the look of the first generation of my current Revel Salon speakers).

 

[rebbiputzmaker]

Heard an HQD system a number of times; one of those systems that sounded impressive to me at first but lost its allure with continued listening. Despite the big 24" woofers bass was not all that great in one system; the other was dialed in a bit better but still more "boom" than true deep bass.

Did not like the woofers either. I used Janis woofers which were the only woofers at the time I felt could match well.

 

[Burning Sounds]

The wall behind my Maggies is heavily damped. Diffusion (unless very, very large/deep) won't help midbass and below cancellation. But of course opinions vary and it depends upon the room. Killing the back wave helps reduce comb filter effects and such so imaging is very precise but also costs some of the sense of "space" (ambience) in the sound. I played with both ways (and many others) over the years and always ended up with absorbers behind except when I had them in a large room, though at least some of the time that was because diffusers cost too dang much and I was too lazy to build them.

I used heavy damping for a number of years with my Maggies. When I built the LX521s I decided to follow Siegfried Linkwitz's recommendation of diffusion. It's just a large floor to almost ceiling bookcase that extends across the entire wall. It's about a foot deep and has randomly placed books of varying sizes. There's nothing scientific about it, but it seems to work well enough when it comes to imaging and spaciousness. Of course, it also depends how far you can get a dipole out into the room to ensure the backwave is delayed enough.

 

[Fitzcaraldo215]

Did not like the woofers either. I used Janis woofers which were the only woofers at the time I felt could match well.

Yes, Hartleys were big, but not that good. Also, I do not ever recall seeing any measurements on the HQD.

I used a pair of Janis subs with my ML CLSs. The deep bass was good, and John Marovskis at Janis was one of the few to publish some modicum of useful measurements. But, the ML CLS panels were still quite deficient and too thin in the mid/upper bass above the sub xover.

 

[Sal1950]

Not on topic, but, yes, electrostats once a long time ago WERE known to arc, especially the Quad 57s. But, that is stale old news. Sorry for your loss in 1976, but how is that relevant today?

https://audiosciencereview.com/forum/index.php?threads/open-baffle-speaker-design.2325/post-85941
https://audiosciencereview.com/forum/index.php?threads/open-baffle-speaker-design.2325/post-85948
Go back and read the posts.
It was relevant because they were what the putz was talking about and I made the statement in reference to, old Quads., So you do also agree old Quads could be great light shows, that was a fact no one could deny but putz..
Nothing was said in reference to anything more modern.
You really should keep up on the train of conversation.

 

[Fitzcaraldo215]

I used heavy damping for a number of years with my Maggies. When I built the LX521s I decided to follow Siegfried Linkwitz's recommendation of diffusion. It's just a large floor to almost ceiling bookcase that extends across the entire wall. It's about a foot deep and has randomly placed books of varying sizes. There's nothing scientific about it, but it seems to work well enough when it comes to imaging and spaciousness. Of course, it also depends how far you can get a dipole out into the room to ensure the backwave is delayed enough.

Yes, exactly. Moving dipoles out into the room rather than trying to get rid of the back wave through absorption or by trying to diffuse it has worked rather well for me over the years.

Possibly, the time delay from greater distance to the wall behind seems to allow our ear-brain to recognize that it is reflected energy and automatically compensate for it. Shorter time delays from a shorter path length seem to be inextricably comingled by our hearing with the direct sound. That would be true as well, I expect, for monopole speakers and side or ceiling reflections, of which dipoles have little.

I read about this years ago, and I use a rule of thumb of about 10 msec of delay, equal to about 10 ft. I therefore have been quite happy to keep my dipoles at least 5 ft. from the wall behind, allowing for the 10 ft. via travel of both the direct back wave and its reflection to the listening position. Comparative trials on the distance have reinforced this idea, unscientifically and anecdotally, of course.

 

[Fitzcaraldo215]

https://audiosciencereview.com/forum/index.php?threads/open-baffle-speaker-design.2325/post-85941
https://audiosciencereview.com/forum/index.php?threads/open-baffle-speaker-design.2325/post-85948
Go back and read the posts.
It was relevant because they were what the putz was talking about and I made the statement in reference to, old Quads., So you do also agree old Quads could be great light shows, that was a fact no one could deny but putz..
Nothing was said in reference to anything more modern.
You really should keep up on the train of conversation.

Sal, old buddy, you said it "is" a known problem and implying it was true of all electrostats. We agree that it "was" a known problem with old electrostats, and maybe it still is for some who use old designs. I merely tried to add some further detail so that all had a larger context to consider.

My train of thought is firmly on the tracks, in my opinion, old friend. But, agreed, we are both OT, you as well as me.

 

[SIY]

The e-stats with which I have personal long-term experience are '57s (eminently arc-able) and Acoustat 1+1 (just about impossible to arc).

FWIW.

 

[freddi]

on the reasonably priced side, sans crossover, $250-$350 per side (including wood) should make a decent open baffle speaker

two 18 inch GW1858 per side would move a fair amount of air without much excursion - I've used on in an H-baffle - its qts is a bit high
but a narrow baffle would offset that factor. Moving mass is pretty low and its suspension seemed made to protect it
against over powering. It supported fairly level to 35Hz. Any ills in the fullrange driver could probably be notched out passively

Tangband W8-2145 https://www.parts-express.com/tang-band-w8-2145-8-paper-cone-full-range-driver-8-ohm--264-960
Goldwood GW1858 https://www.parts-express.com/goldwood-gw-1858-18-pro-woofer--290-386


such a system with H-baffle and one 8 inch fullrange would be like this one

http://www.quarter-wave.com/Gallery/AN_and_H_Frame.jpg

 

[Thomas savage]

Open waffle..