What? You drinkin' Bloody Marys tonight or what?
I'm havin' a few 'Beveridges' myself tonight...
(Actually, I have no idea what you're talking about...haha)
I was digging around in the McNichols catalog a few days ago and found some 63% open perf sheets (18 gauge) that were $155 for a sheet 36" by 120". I asked for a quote to ship it to me cut in 4, and they gave me a price of $230...WTF? That's not including the $20 shipping and handling... Really? You guys want that much to set it up? Yikes!
I looked up gauges in inches on Wiki and it says that 18 gauge is .05 inches, then I looked up 20 gauge and that stuff is .0375 inches, but at a cheaper price ($112), so I imagine that 20 gauge would be about $190 or so, much better.
That is if 20 gauge is going to be OK for ESL construction.
If I plan on using a fair amount of paint/powder/insulation, do you guys think 20 gauge would be OK?
It looks a lot like MLs X-stat perf, only not quite as open, but real close.
I attaches a couple of pics for visual reference below:
Just another question in a long line of other questions...
cheers
-wreck
Attachments
Sorry for hijacking the thread a bit, but I find the discussion about ESL panel size and SPL interesting. I have a pair of Acoustat Spectra 11's as mains right now and I feel like the SPL is enough at about 8 feet listening distance (my room is about 180 sqf). For the lower frequencies I'm building a OB bass tower similar to Calvin's designs (thanks for the help Calvin ).
I've considered replacing the Acoustat with a smaller and more modern design like the ER Audio Mini Panels (they are currently redesigning them to be a bit slimmer and taller, and add vertical segmentation for better HF dispersion). What would be compromises in terms of sound quality and SPL be? To calculate how much air a dynamic driver is moving you can simply multiple Xmax with the surface area. Is it the same for an ESL panel and what is typically "Xmax" for a panel? Does the transformers limit the SPL of the panel?
Thanks.
).I've considered replacing the Acoustat with a smaller and more modern design like the ER Audio Mini Panels (they are currently redesigning them to be a bit slimmer and taller, and add vertical segmentation for better HF dispersion). What would be compromises in terms of sound quality and SPL be? To calculate how much air a dynamic driver is moving you can simply multiple Xmax with the surface area. Is it the same for an ESL panel and what is typically "Xmax" for a panel? Does the transformers limit the SPL of the panel?
Thanks.
phazer99, at $410 a panel, I think you're better off staying with what you have.
Otoh, you can build your own ESLs anyway you like. Looks like the same outfit sells the important and difficult to find "stuff" for building ESL cells.
In ESLs, all things being equal, more surface area = more SPL, and usually more bass output (not necessarily LOWER bass though, but probably so since baffle width plays a big role.
Otoh, you can build your own ESLs anyway you like. Looks like the same outfit sells the important and difficult to find "stuff" for building ESL cells.
In ESLs, all things being equal, more surface area = more SPL, and usually more bass output (not necessarily LOWER bass though, but probably so since baffle width plays a big role.
Reason why I wouldn't is in my first post in this thread, the wall vs window argument.
I didn't even know Beveridge existed until this morning, looks like the curve I put on this tweeter way back in '99, my first DIY on a pair of Scan-Speak drivers. (see attachment)
@ Phaser, Not a big derail, no worries.
Forgot to mention that the McNichols perf I posted last night has 5/32" holes on 3/16" centers, are those considered too big? Enough to negatively affect field density? If anyone knows or has an opinion, please respond, thanks.
Off topic: I saw this thread while I was at work today but couldn't respond because I was on my Android phone, does anyone know if there is an app for doing this?
thanks
-wreck
Attachments
Ok, either you or I are rather confused?
Sounds like you want horizontal dispersion? No??
You said you wanted "curved" panels, right? The purpose of the curve is dispersion, yes?
Flat panels are not going to have good HF horizontal dispersion unless either exceptionally good segmentation is employed (frequency as in Acoustat Spectra, or time as in ESL63) or an acoustic lens is used (Beveridge). There are not too many other choices.
Oh, wait, I just re-read you post and your saying that the curve on the tweeter you built is like the Beveridge. No, it's not in
any way close. Best read up a bit more on the Beveridge design. While it is *not* precisely what you are going for, it does
represent a means by which "curvature" can be produced by a flat diaphragm speaker. It's *not* a curved box that is important to the Beveridge design.
Sounds like you want horizontal dispersion? No??
You said you wanted "curved" panels, right? The purpose of the curve is dispersion, yes?
Flat panels are not going to have good HF horizontal dispersion unless either exceptionally good segmentation is employed (frequency as in Acoustat Spectra, or time as in ESL63) or an acoustic lens is used (Beveridge). There are not too many other choices.
Oh, wait, I just re-read you post and your saying that the curve on the tweeter you built is like the Beveridge. No, it's not in
any way close. Best read up a bit more on the Beveridge design. While it is *not* precisely what you are going for, it does
represent a means by which "curvature" can be produced by a flat diaphragm speaker. It's *not* a curved box that is important to the Beveridge design.
Last edited:
Leaving this alone, I stated I'm a 'beginner' earlier in this thread.
Yes, I want horizontal dispersion, preferably with an ESL panel. I read in Sanders cookbook that if had a 30 degree curve on my ESLs, at 10 feet away, I'd have 5'7" horizontal dispersion, then I extrapolated that figuring (into) using 2 panels, and that intrigued me as well. That was the main point of me starting this thread.
Only if it works as advertized, even if it doesn't, and say it only provides half of that performance, I'm still intrigued.
My flats certainly don't give good HF-HD, I was under no illusion that flats did, I own them, I know.. You guys keep mentioning segmentation, got links? (or explanations etc.)
I merely took a quick look at some Beveridge images, and if you re-read my post, I said it looks like my old DIY project, nothing more.
Upon re-observation of this thread, seems that many have thought my proposed panels are too wide. Am I using incorrect equations when it comes to this? I figured 200 Hz quarter wavelength was around 16 (point something )inches... and that if I want to avoid dipole phase cancellation, I'd pick a width nothing short.
The last thing I want to do is artificially dope my ESLs with a FR it can't handle, it's artificial.
Hi,
hass it been veryfied, that such a lens performs as claimed?
Recently a guy showed his built in the german Elektrostaten-Forum.
It was similar to the Beveridge lens, made from thin plywood, with a lower number count of slots.
Unfortunately he couldn't measure to verify the claims for such a lens, resp. simplified lens.
I assume that the effect will be lower than expected, the same as the degree of a curved panel doesn't equal the degree of dispersion.
Can one be sure that the lens itself doesn't effect the sound with its own acoustic fingerprint through reflections, resonating et al?
jauu
Calvin
hass it been veryfied, that such a lens performs as claimed?
Recently a guy showed his built in the german Elektrostaten-Forum.
It was similar to the Beveridge lens, made from thin plywood, with a lower number count of slots.
Unfortunately he couldn't measure to verify the claims for such a lens, resp. simplified lens.
I assume that the effect will be lower than expected, the same as the degree of a curved panel doesn't equal the degree of dispersion.
Can one be sure that the lens itself doesn't effect the sound with its own acoustic fingerprint through reflections, resonating et al?
jauu
Calvin
Last edited:
Hi,
So I won´t spoil the fun of finding out who´s the murderer. Sure You´ll soon find out the answer.
But I could pluck out some of my spare gray hair and mail it to You
It seems since #8 to be an upcoming idea, that to increase SPL one could just crank up signal and biasing voltages and make the panel larger.
While this works at first glance on paper there are limits and restrictions in practise.
Biasing and Signal voltage regard the achievable force on the diaphragm.
This is eventually limited by the flashover voltage of air, typically quoted to be ~2kV/mm. 1.5/mm to 1.7kV/mm is a more practical value to provide for a truly quiet panel under different air conditions.
Now, as Gerald said in #8 You do the math for 0.075", 6kV bias and 20-25kVpp signal.
Under perfect conditions a 0.075" airgap would allow for 3.8kV Bias before the flashover treshold is reached.
This Bias value allows for 7.6kVpp signal, before signal frequency doubling occurs. Obviously much lower values.
Since I trust him even with those bold claims, the only explanation I have is, that between the measurement points and diaphragm a great deal of the voltages must have ´disappeared´.
This could be in form of too high valued resistors in the biasing path, massive leakage, or due to a very lossy insulative coating of the stators. In the latter case, it might have been of interest to measure, if a temperature rise occured
Since most panels electronics will be designed already close to the voltage tresholds the panel allows for, the most promising way to increase SPL is to size up the panel.
The problem here beeing that a larger panel also requires larger listening distances, proportional to the increasing nearfield-to-farfield transistion distance.
As for the amount of eq against acoustic phasecancellation, wider panels would require less dBs at a certain low xover frequency.
But again one has to watch for the relationships. Changing just one parameter, changes several other related parameters as well.
Typically the larger/wider panel will be designed to achieve a lower bandwidth limit.
This regards the distances of spacers, the thickness of spacers (d/s), mechanical diaphragm tension, bias and signal voltages, transformation factor, etc. etc.
For highest efficiency and dynamic stability the diaphragm needs to be pulled as taut as possible.
Depending on membrane type and thickness this will result in a base resonance 120Hz<Fs<250Hz. If You keep off of Fs by at least a factor of 1.5 You´ll end up with Xover-freqs of 200<Fxo<400Hz. The difference in equing between the panels is not so big any more that it plays a deciding role.
You can always fudge a bit and add wings to the panel sides, thereby ´passively´ increasing baffle width.
The high mechanical tension requires a solid fixation of the diaphragm in the vertical direction. I´d choose at least 1" of non-perforated rim here.
The vertical rims may be thinner as there´s nearly no mechanical tension in the horizontal. 1/2" is a good choice as it is wider than a charging ring, which is positioned around the perimeter also for high flashover treshold and nice optics.
Regarding different spacer thicknesses front/back stator, it is a con in theory, but can be a pro in practise.
It doesn´t necessarily increase asymmetry, but actually helps in increasing symmetry (low THD) and dynamic range.
It counters to a degree the unavoidable tendency of the diaphragm towards a hour-glass shape due to manufacturing tolerances.
A difference of ~10% is perfectly practical.
The 5/32" to 3/16" pattern is a good choice. If the perimeter is perforated also, You have to treat the rims, such that no sharp edges remain, which are prone to flashovers.
jauu
Calvin
I meant, that 3/8" will make a noteable difference to 3/16" and that either this or that is not debatable, as design requirements for an optimum design will immideately dictate the right choice.
Well, I guess You are eagerly reading papers, threads and books at the time.
So I won´t spoil the fun of finding out who´s the murderer. Sure You´ll soon find out the answer.
But I could pluck out some of my spare gray hair and mail it to You
It seems since #8 to be an upcoming idea, that to increase SPL one could just crank up signal and biasing voltages and make the panel larger.
While this works at first glance on paper there are limits and restrictions in practise.
Biasing and Signal voltage regard the achievable force on the diaphragm.
This is eventually limited by the flashover voltage of air, typically quoted to be ~2kV/mm. 1.5/mm to 1.7kV/mm is a more practical value to provide for a truly quiet panel under different air conditions.
Now, as Gerald said in #8 You do the math for 0.075", 6kV bias and 20-25kVpp signal.
Under perfect conditions a 0.075" airgap would allow for 3.8kV Bias before the flashover treshold is reached.
This Bias value allows for 7.6kVpp signal, before signal frequency doubling occurs. Obviously much lower values.
Since I trust him even with those bold claims, the only explanation I have is, that between the measurement points and diaphragm a great deal of the voltages must have ´disappeared´.
This could be in form of too high valued resistors in the biasing path, massive leakage, or due to a very lossy insulative coating of the stators. In the latter case, it might have been of interest to measure, if a temperature rise occured
Since most panels electronics will be designed already close to the voltage tresholds the panel allows for, the most promising way to increase SPL is to size up the panel.
The problem here beeing that a larger panel also requires larger listening distances, proportional to the increasing nearfield-to-farfield transistion distance.
As for the amount of eq against acoustic phasecancellation, wider panels would require less dBs at a certain low xover frequency.
But again one has to watch for the relationships. Changing just one parameter, changes several other related parameters as well.
Typically the larger/wider panel will be designed to achieve a lower bandwidth limit.
This regards the distances of spacers, the thickness of spacers (d/s), mechanical diaphragm tension, bias and signal voltages, transformation factor, etc. etc.
For highest efficiency and dynamic stability the diaphragm needs to be pulled as taut as possible.
Depending on membrane type and thickness this will result in a base resonance 120Hz<Fs<250Hz. If You keep off of Fs by at least a factor of 1.5 You´ll end up with Xover-freqs of 200<Fxo<400Hz. The difference in equing between the panels is not so big any more that it plays a deciding role.
You can always fudge a bit and add wings to the panel sides, thereby ´passively´ increasing baffle width.
The high mechanical tension requires a solid fixation of the diaphragm in the vertical direction. I´d choose at least 1" of non-perforated rim here.
The vertical rims may be thinner as there´s nearly no mechanical tension in the horizontal. 1/2" is a good choice as it is wider than a charging ring, which is positioned around the perimeter also for high flashover treshold and nice optics.
Regarding different spacer thicknesses front/back stator, it is a con in theory, but can be a pro in practise.
It doesn´t necessarily increase asymmetry, but actually helps in increasing symmetry (low THD) and dynamic range.
It counters to a degree the unavoidable tendency of the diaphragm towards a hour-glass shape due to manufacturing tolerances.
A difference of ~10% is perfectly practical.
The 5/32" to 3/16" pattern is a good choice.
If the perimeter is perforated also, You have to treat the rims, such that no sharp edges remain, which are prone to flashovers.jauu
Calvin
Very Good Point's!
As raising the bias and/or stator voltages does increase efficiency, This is often confused with overall peak output of the panel and should not be as they are two very different parameters.
The amount of surface area and diaphragm displacement (area * D/S) for the most part determines on what the peak output will be for a certain frequency.
For your average (monopole) driver it takes 4 times the excursion to maintain the same SPL at the next octave lower and then you have dipole cancellations working against you as well.
In practice for a Planar Dipole your lowest frequency of interest will set your overall peak SPL for the panel if it is to have a flat response.
Before my stator coating had failed the very first time in those tests, I was at one point or another ionizing the air within the gap.
I pushed it to that point several times for long peroids before failure finally occurred.
At that time strangely enough I had not burned any holes in the diaphragm as it was not arcing through from stator to stator, it was just ionizing the air.
I was quite fascinated by this and I tried my best at the time to determine if it was actually getting any louder or not as I pushed the amp more, as it had plenty of more headroom before I had to worry about clipping.
Also note that I was working with test tones and not music, so it was a real stress test.
Sadly, I didn't have a working SPL meter at the time and eventually the panel failed completely.
I spent more time repairing it a few times rather than just building a new one.
I still had the other second panel but by the time I got back to testing again, one of my smaller speaker cabinets had fell on it and cracked one of the stators ripping the screen.
So I knew right then and there, that it would be a lost cause to even try and use it for what I was trying to achieve.
In Peter walker's paper he states that the maximum force that can be applied to the Diaphragm is limited by the ionization of the air and is about 50 n/sm( or something as I was just reading this again) and he stated a figure of about 105db SPL.
But, I forget what the variables were as far as panel size and displacement that he was referring to ( maybe it was for 1 sq. meter and I will have to read it again).
It seems that I had exceeded that but he doesn't state what his testing frequency was or maybe I missed something.
I was in the middle of verifying all of that when havoc struck!
Now as I stated it was quite loud for such a little panel and I could only imagine what a large panel would be like under such conditions.
I have found that about 7kv bias is a reachable maximum limit at least for my small panel designs and the peak stator voltage levels makes for earsplitting SPL's.
It is more than enough especially for a larger panel, most rarely exceed half of those voltage levels before their woofer system can't keep up anymore!
Yes, I got an extra Free 6db every time I doubled the voltage from 2.5Kv to 5Kv and then to 10Kv.
But that does not say what my Peak SPL capability was.
It did however take a lot less amplifier power to get to 105db though, Much Less!!
Once I got to 10Kv then coating material breakdown became a big issue at that point.
As Calvin stated (and was discussed in another thread) that, Even if I was at a 10Kv bias, The voltage in the gap was still only maybe about 5.75Kv to 7Kv or so in the gap at a D/S of .072" and 80V to 100V per mil (average breakdown of air on a good dry day), and, the rest was dissipated in the resistance factor of the coating itself.
I need to build an ammeter for my supply in order to determine this.
This stuff is what I have been studying slowly and was why I built my Variable regulated HV supply.
But it is slow going and time consuming, not to mention the cost of making samples to test and get reliable information consistently in order to confirm the results.
In short I spent more time burning the things up than listening to them!!! He,he,he,he
I just wanted to give some justification to my statements about the voltage levels before anyone thinks that it is easy to start raising the voltages and expect it all to be good.
It is possible to an extent, but one must understand the basic science behind it First, as there are limitations as well.
jer
As raising the bias and/or stator voltages does increase efficiency, This is often confused with overall peak output of the panel and should not be as they are two very different parameters.
The amount of surface area and diaphragm displacement (area * D/S) for the most part determines on what the peak output will be for a certain frequency.
For your average (monopole) driver it takes 4 times the excursion to maintain the same SPL at the next octave lower and then you have dipole cancellations working against you as well.
In practice for a Planar Dipole your lowest frequency of interest will set your overall peak SPL for the panel if it is to have a flat response.
Before my stator coating had failed the very first time in those tests, I was at one point or another ionizing the air within the gap.
I pushed it to that point several times for long peroids before failure finally occurred.
At that time strangely enough I had not burned any holes in the diaphragm as it was not arcing through from stator to stator, it was just ionizing the air.
I was quite fascinated by this and I tried my best at the time to determine if it was actually getting any louder or not as I pushed the amp more, as it had plenty of more headroom before I had to worry about clipping.
Also note that I was working with test tones and not music, so it was a real stress test.
Sadly, I didn't have a working SPL meter at the time and eventually the panel failed completely.
I spent more time repairing it a few times rather than just building a new one.
I still had the other second panel but by the time I got back to testing again, one of my smaller speaker cabinets had fell on it and cracked one of the stators ripping the screen.
So I knew right then and there, that it would be a lost cause to even try and use it for what I was trying to achieve.
In Peter walker's paper he states that the maximum force that can be applied to the Diaphragm is limited by the ionization of the air and is about 50 n/sm( or something as I was just reading this again) and he stated a figure of about 105db SPL.
But, I forget what the variables were as far as panel size and displacement that he was referring to ( maybe it was for 1 sq. meter and I will have to read it again).
It seems that I had exceeded that but he doesn't state what his testing frequency was or maybe I missed something.
I was in the middle of verifying all of that when havoc struck!
Now as I stated it was quite loud for such a little panel and I could only imagine what a large panel would be like under such conditions.
I have found that about 7kv bias is a reachable maximum limit at least for my small panel designs and the peak stator voltage levels makes for earsplitting SPL's.
It is more than enough especially for a larger panel, most rarely exceed half of those voltage levels before their woofer system can't keep up anymore!
Yes, I got an extra Free 6db every time I doubled the voltage from 2.5Kv to 5Kv and then to 10Kv.
But that does not say what my Peak SPL capability was.
It did however take a lot less amplifier power to get to 105db though, Much Less!!
Once I got to 10Kv then coating material breakdown became a big issue at that point.
As Calvin stated (and was discussed in another thread) that, Even if I was at a 10Kv bias, The voltage in the gap was still only maybe about 5.75Kv to 7Kv or so in the gap at a D/S of .072" and 80V to 100V per mil (average breakdown of air on a good dry day), and, the rest was dissipated in the resistance factor of the coating itself.
I need to build an ammeter for my supply in order to determine this.
This stuff is what I have been studying slowly and was why I built my Variable regulated HV supply.
But it is slow going and time consuming, not to mention the cost of making samples to test and get reliable information consistently in order to confirm the results.
In short I spent more time burning the things up than listening to them!!! He,he,he,he
I just wanted to give some justification to my statements about the voltage levels before anyone thinks that it is easy to start raising the voltages and expect it all to be good.
It is possible to an extent, but one must understand the basic science behind it First, as there are limitations as well.
jer
Last edited:
Calvin,
My take on the Beveridge design is that it is a combination of simple "waveguide" (its shape bends) and a diffraction lens. I think it could be improved upon today. Undoubtedly it has some audible effects, it would be very difficult for it not to.
Can't say what the fellow who tried to make a copy did or how well it succeeded. The son of the inventor still makes the speakers, or did as of a few years back, so it ought to be possible to hear/see/measure the real thing somehow.
The idea of the Beveridge (think there is a thread here on them) was that the dispersion was nearly 180deg, so the speakers were to be aimed *AT* each other from the left and right! Unusual for sure.
wreckingball, I understand that you are a beginner, that's no problem. It's that you seemed to have stated something contrary to your original post, then referenced the original post to back that up. So, I just was trying to clear it up.
As far as what you said, again, I was trying to make it clear that there is an *idea* to consider, even if you don't attempt to build it or use it. Merely investigating the design of the Beveridge speakers will serve to make clear some aspects of acoustics and speaker design.
To recap what I was trying to convey:
- curved stators may provide some increased HF dispersion, but I'm unclear that they are a real benefit compared to side-by-side thin width vertical cells.
- The other problem with curved stators AND horizontal segmentation is that the centers of each segment are spaced rather widely, so you may get a comb filter effect, which is undesirable.
- you can do a curved stator and *vertical* segmentation, that may buy some benefit in terms of cell-to-cell distance compared to discrete cells placed side by side.
- something else, but I can't think of it.
Oh, another thought to confound the entire idea. The actual height above ground will dramatically effect the response below ~250Hz (for example with the speakers you have now) due to "floor bounce".
Some of the rules that apply to line arrays apply to ESLs as well. That too is something to consider.
Also, you may wish to read up on older threads here and information elsewhere on the Acoustat design. It managed to go low into the bass with panels that were 8" or 9" wide.
And fwiw, jer and Calvin have much information and expertise WRT ESLs...
Searching the older threads will get you a *boatload* of fantastic information and ideas on ESLs... ought to keep you busy for a while and answer many questions, but raise new ones too.
Enjoy the quest!
My take on the Beveridge design is that it is a combination of simple "waveguide" (its shape bends) and a diffraction lens. I think it could be improved upon today. Undoubtedly it has some audible effects, it would be very difficult for it not to.
Can't say what the fellow who tried to make a copy did or how well it succeeded. The son of the inventor still makes the speakers, or did as of a few years back, so it ought to be possible to hear/see/measure the real thing somehow.
The idea of the Beveridge (think there is a thread here on them) was that the dispersion was nearly 180deg, so the speakers were to be aimed *AT* each other from the left and right! Unusual for sure.
wreckingball, I understand that you are a beginner, that's no problem. It's that you seemed to have stated something contrary to your original post, then referenced the original post to back that up. So, I just was trying to clear it up.
As far as what you said, again, I was trying to make it clear that there is an *idea* to consider, even if you don't attempt to build it or use it. Merely investigating the design of the Beveridge speakers will serve to make clear some aspects of acoustics and speaker design.
To recap what I was trying to convey:
- curved stators may provide some increased HF dispersion, but I'm unclear that they are a real benefit compared to side-by-side thin width vertical cells.
- The other problem with curved stators AND horizontal segmentation is that the centers of each segment are spaced rather widely, so you may get a comb filter effect, which is undesirable.
- you can do a curved stator and *vertical* segmentation, that may buy some benefit in terms of cell-to-cell distance compared to discrete cells placed side by side.
- something else, but I can't think of it.
Oh, another thought to confound the entire idea. The actual height above ground will dramatically effect the response below ~250Hz (for example with the speakers you have now) due to "floor bounce".
Some of the rules that apply to line arrays apply to ESLs as well. That too is something to consider.
Also, you may wish to read up on older threads here and information elsewhere on the Acoustat design. It managed to go low into the bass with panels that were 8" or 9" wide.
And fwiw, jer and Calvin have much information and expertise WRT ESLs...
Searching the older threads will get you a *boatload* of fantastic information and ideas on ESLs... ought to keep you busy for a while and answer many questions, but raise new ones too.
Enjoy the quest!
Lighter Side
OK..as usual, a huge amount of stuff is being bandied about here. On lighter subjects:
Charlie has re-activated the DIY phase of building ESLs, in my opinion. RIP Lucas and others. In recognition of that, I propose that anyone who uses the design from his blog should be required to refer to the finished ESL as a "Jazzman." (I already do.)
Wreck: is that your wife and/or GF in the pic? Standing next to an speaker, no potted plants in the way,WITH A SMILE ON HER FACE? You are a lucky man.How did you achieve the high WAF?
Keep on building, everyone. I'm currently rebuilding my Jazzmans because I screwed up the stator insulation process the first time...had some arcing problems that I hope have been resolved with Jer's advice on insulation painting.
OK..as usual, a huge amount of stuff is being bandied about here. On lighter subjects:
Charlie has re-activated the DIY phase of building ESLs, in my opinion. RIP Lucas and others. In recognition of that, I propose that anyone who uses the design from his blog should be required to refer to the finished ESL as a "Jazzman." (I already do.)
Wreck: is that your wife and/or GF in the pic? Standing next to an speaker, no potted plants in the way,WITH A SMILE ON HER FACE? You are a lucky man.How did you achieve the high WAF?
Keep on building, everyone. I'm currently rebuilding my Jazzmans because I screwed up the stator insulation process the first time...had some arcing problems that I hope have been resolved with Jer's advice on insulation painting.
Thanks, but you're much too generous. Almost everything learned and shared on my blog was gleaned from other builders on this forum. And we know who they are
Agreed! on both points...
Charlie's a helluva good guy, and he's done (whether he knows it or not) a great service to all of us just dipping our toes into the deep-end water of this crazy ESL building world. Been lurking since he assisted Mavric on his little quest...
HAHA! I wish ma brutha... No no no, that's just some Google image, but it was the only one that showed, in any kind of detail, ML's new perforated panels... I used to have a better picture that showed them to scale, but alas, it has disappeared into the aether, oh well...
Tell you what, you find a feline like that for me?? ...Lifetime of beers provided, on the house ..
@ Jer and Calvin, thank you very much for the info you have provided. ...matter of fact, I have started printing your posts and putting them in a little folder for future reference, Great stuff!
Hey bear, no hard feelings, eh? Might have come across weird last night.
Floor bounce, vented alignments (always preferred QB3), sealed, standing waves, reflections etc.etc. are all in my speaker-building background.
Also, I used ti use LAUD and Lsp CAD, but my buddy (who has the time) now runs Praxis and another modeling program...Short: should be able to provide measurements for anything I build, which should be a plus for anyone following.
But these ESLs? Until I know more, I'm gonna have to 'stand on the shoulders of giants' in order to get my first grasp, and thank goodness there are people like that here at DIYaudio to provide.
I see Charlie has entered, hey Jazzman! (good to see you).
I just got some 'C' Mylar in today, stay tuned for a test on shrink-ability tomorrow, (I'll take pics)
cheers
-wreck
Last edited:
Tried to test heat-shrink-ability today and the hair dryer I tried to use pushes too much air, doesn't get hot enough.
So I lit up a cig, and hovered it over the surface about 1/4".. that worked, as you can see the 2 small strips within the wrinkle smoothed out.
See attached pics:
This is obviously impractical, and I'll try again tomorrow with a 2400W heat gun at work.
For the record, this is standard "C" Mylar, as opposed to the "heat-shrinkable" Mylar. Note about that stuff: they claim 50% shrinkage while dipping into boiling water, I think that would send the diaphragm right into the back stator on a curved ESL, not good, and I think this standard stuff is gonna be just fine, but we'll see...
So I lit up a cig, and hovered it over the surface about 1/4".. that worked, as you can see the 2 small strips within the wrinkle smoothed out.
See attached pics:
This is obviously impractical, and I'll try again tomorrow with a 2400W heat gun at work.
For the record, this is standard "C" Mylar, as opposed to the "heat-shrinkable" Mylar. Note about that stuff: they claim 50% shrinkage while dipping into boiling water, I think that would send the diaphragm right into the back stator on a curved ESL, not good, and I think this standard stuff is gonna be just fine, but we'll see...
Attachments
Yes, the First panel I made I tried to use a hair dryer and it didn't work at all and it was a high wattage type.
The thermostat that is built in to them keeps them from getting hot enough so I had to go out and get a real Heat Gun and that did the trick!!!!
Just don't take your eyes off of what you are doing even for a split second as it is very easy to burn hole in your freshly applied diaphragm!!!
I know this because I have done it a few times!!! He,he,he,he
So, Keep the gun Moving!!!
The type of mylar you should be using should be of a tensilized type.
Tensilized Mylars are pre-stretched and will only shrink so much like about 2% maybe up to 4% or so and no more than that no matter how many times you re-heat it.
You can find the exact info on the DuPont website.
I have had a few on occasion lose their tension some what for whatever reason such as taking the panels apart and putting them back together again.
All I had to do was hit them with some heat again and all was good again with very consistent results even after I had taken them apart and put them back again with out heating them.
My typical diaphragm resonance target was about 70Hz to 90Hz and rarely ever got much above 120Hz to 150Hz for my panel dimensions when I really heated them up to see how far they would go.
Once I had got some practice under my belt I aimed for the lowest resonate frequency (tension) that I could get that still allowed the diaphragm to be stable and not get stuck to one of the stator's.
This was more difficult to do with my wider 8" panels using the thinner .25 mil Mylar.
On my very first panels I was using some very thin model airplane covering's as they are a type of Mylar or Polyester too.
One time I kept heating one up while it was not sandwiched together to see how far it would go and it kept shrinking until it snap my support frame.
In that case it was not a tensilized material.
I still have some of that stuff!!
Once I had my design working properly and the all the coating's and shrink methods down, I started using my hard to find .25mil Mylar for the more permanent applications.
jer
The thermostat that is built in to them keeps them from getting hot enough so I had to go out and get a real Heat Gun and that did the trick!!!!
Just don't take your eyes off of what you are doing even for a split second as it is very easy to burn hole in your freshly applied diaphragm!!!
I know this because I have done it a few times!!! He,he,he,he
So, Keep the gun Moving!!!
The type of mylar you should be using should be of a tensilized type.
Tensilized Mylars are pre-stretched and will only shrink so much like about 2% maybe up to 4% or so and no more than that no matter how many times you re-heat it.
You can find the exact info on the DuPont website.
I have had a few on occasion lose their tension some what for whatever reason such as taking the panels apart and putting them back together again.
All I had to do was hit them with some heat again and all was good again with very consistent results even after I had taken them apart and put them back again with out heating them.
My typical diaphragm resonance target was about 70Hz to 90Hz and rarely ever got much above 120Hz to 150Hz for my panel dimensions when I really heated them up to see how far they would go.
Once I had got some practice under my belt I aimed for the lowest resonate frequency (tension) that I could get that still allowed the diaphragm to be stable and not get stuck to one of the stator's.
This was more difficult to do with my wider 8" panels using the thinner .25 mil Mylar.
On my very first panels I was using some very thin model airplane covering's as they are a type of Mylar or Polyester too.
One time I kept heating one up while it was not sandwiched together to see how far it would go and it kept shrinking until it snap my support frame.
In that case it was not a tensilized material.
I still have some of that stuff!!
Once I had my design working properly and the all the coating's and shrink methods down, I started using my hard to find .25mil Mylar for the more permanent applications.
jer
Last edited:
Wreck: assuming you are following the Jazzman ESL procedures (stretch Mylar, apply stator, flip it over, apply other stator) check for wrinkles resulting from the tape pulling up off the stator. I had this problem on my recent rebuild. I tried heating the Mylar which removed the wrinkles but I just wasn't sure I had a good stretch. So after a suitable amount of obsessing, I decided to go another route: I cut out a hole in my stretch table the size of a stator so that I can apply the stators to each side of the stretched Mylar while it is still in tension. I haven't finished the second panel yet, but I think I will get good results this way. Good luck...
Hi,
Mylar C can only be used with mechanical tensioning for curved ESLs.
For heat treatment only flat panel design are feasable.
As I mentioned earlier, one needs a special film that allows for heat treatment of curved panels, but this film is afaik long obsolete.
@bear
I agree that the Beveridge lens may be viewed as a combination of waveguide and diffraction lens. The inner slots seem to function like a waveguide, and the outer slots like delay-lines.
The mentioned simpler copy in the german forum used only 5-7 slots, with a rather wide central slots and a pair of 2-3 slimm outer slots.
After my understanding the inner slot was chosen too wide and the outer slots introduced too low a delay to function as intended.
They used a old Audiostatic panel, which beeing a wire-stator panel wouldn´t have required the "help" of a diffracting device anyway.
jauu
Calvin
Mylar C can only be used with mechanical tensioning for curved ESLs.
For heat treatment only flat panel design are feasable.
As I mentioned earlier, one needs a special film that allows for heat treatment of curved panels, but this film is afaik long obsolete.
@bear
I agree that the Beveridge lens may be viewed as a combination of waveguide and diffraction lens. The inner slots seem to function like a waveguide, and the outer slots like delay-lines.
The mentioned simpler copy in the german forum used only 5-7 slots, with a rather wide central slots and a pair of 2-3 slimm outer slots.
After my understanding the inner slot was chosen too wide and the outer slots introduced too low a delay to function as intended.
They used a old Audiostatic panel, which beeing a wire-stator panel wouldn´t have required the "help" of a diffracting device anyway.
jauu
Calvin
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.