Right now I have two choices for the stator material:
1. The aluminum I have is cut 12"x40", has perforations .1875" with an open area of 51%
2. The steel material has perforations .078" with an open area of 45%.
Aluminum's conductivity is about an order magnitude better than the steel, its lighter and can be anodized. With steel I fear its own weight will cause it to warp if not properly housed in a frame. The steel was better machined less sharp edges and the perforations have a nice rounded edge. It looks like I will be working with my already cut aluminum. I will have it anodized soon, before doing so need to discuss the way I will make my contacts with it. I was just going to pass a bolt (metal or Nylon?) trough one stator (drilled through the acrylic spacers (0.08" thick) out the other stator. The wires will have clips (u shaped, smoothed out edges) attached to the bolt, and lightly tighten a bolt on the other end. Its straight-forward, but making good lasting connections is obviously a must. For the diaphragm contact I have ¼” copper tape that will be recessed into the acrylic touching the coated side of the mylar. I was just going to solder a connection to the copper before its recessed into the acrylic so I don’t melt the acrylic with my solder gun. Does this sound reasonable?
After its anodized do I have to scrap the surface so that the contacts rest on the bare aluminum surface?
What's a good cleaning agent to clean off the aluminum's grease before its anodized black?
Thank You
Bryan
1. The aluminum I have is cut 12"x40", has perforations .1875" with an open area of 51%
2. The steel material has perforations .078" with an open area of 45%.
Aluminum's conductivity is about an order magnitude better than the steel, its lighter and can be anodized. With steel I fear its own weight will cause it to warp if not properly housed in a frame. The steel was better machined less sharp edges and the perforations have a nice rounded edge. It looks like I will be working with my already cut aluminum. I will have it anodized soon, before doing so need to discuss the way I will make my contacts with it. I was just going to pass a bolt (metal or Nylon?) trough one stator (drilled through the acrylic spacers (0.08" thick) out the other stator. The wires will have clips (u shaped, smoothed out edges) attached to the bolt, and lightly tighten a bolt on the other end. Its straight-forward, but making good lasting connections is obviously a must. For the diaphragm contact I have ¼” copper tape that will be recessed into the acrylic touching the coated side of the mylar. I was just going to solder a connection to the copper before its recessed into the acrylic so I don’t melt the acrylic with my solder gun. Does this sound reasonable?
After its anodized do I have to scrap the surface so that the contacts rest on the bare aluminum surface?
What's a good cleaning agent to clean off the aluminum's grease before its anodized black?
Thank You
Bryan
Hi,
Conductivity is not important for this application.
You worry about gravity but you will have considerable electrical forces on the stators (think in the order of 20-40N) so proper housing and support is needed. The same goes for the spacer frame that will have to support the diaphragm (forces here: 100s of Newtons, and the bad news here is that if it slackens one milimeter you will have lost most of the diaphragm tension). Also you want to have a very consistent diaphragm to stator spacing all over the area. This, together with the choice of D/S spacing itself and diaphragm tension will be the most critical parameters of your design. Most other stuff you can change afterwards (electronics, HT).
Sharp edges are a show stopper, if your stators have sharp edges around the holes then you need to get those removed first.
Small nylon bolts can't be fastened good enough for making a proper stator contact. Best would be to have a nice solder lip spot welded to the stator. But any proper non-oxidating contact with enough pressure will do.
I don't see the necessity for recessing the copper tape. On the contrary, if you just put it along the inner edge of the spacers it will ensure good pressurised contact all over the area. This is another critical part of construction.
Conductivity is not important for this application.
You worry about gravity but you will have considerable electrical forces on the stators (think in the order of 20-40N) so proper housing and support is needed. The same goes for the spacer frame that will have to support the diaphragm (forces here: 100s of Newtons, and the bad news here is that if it slackens one milimeter you will have lost most of the diaphragm tension). Also you want to have a very consistent diaphragm to stator spacing all over the area. This, together with the choice of D/S spacing itself and diaphragm tension will be the most critical parameters of your design. Most other stuff you can change afterwards (electronics, HT).
Sharp edges are a show stopper, if your stators have sharp edges around the holes then you need to get those removed first.
Small nylon bolts can't be fastened good enough for making a proper stator contact. Best would be to have a nice solder lip spot welded to the stator. But any proper non-oxidating contact with enough pressure will do.
I don't see the necessity for recessing the copper tape. On the contrary, if you just put it along the inner edge of the spacers it will ensure good pressurised contact all over the area. This is another critical part of construction.
Hi Bryan,
ESLs are high impedance devices. As such the conductivity of the stators only plays a role above values of several kOhms (think of the resistor values needed to segment a panel). Regarding conductivity there will be no difference between aluminium and steel worth mentioning. BTW: make sure that the sheet´s holes are rounded on both sides of the sheet. Sanding and/or etching is strongly optional.
I wouldn´t recommend the usage of aluminium sheets for any other than very small tweeter-panels. Besides beeing too soft for larger panels a serious problem will be to create a long lasting stable electrical contact point. As you know aluminium oxidizes fast. A simple single bolt forming a contact by pressure on a pair of contact clips won´t last long and would ´compress´ the complete panel. A contact failure is preprogrammed.
Rather a 2-bolts solution, one bolt for each stator sheet could be feasable, since You could tighten each bolt very hard without exerting pressure on the panel itself. Additionally the 2-bolt solution must be manufactured before the spacers and membrane are glued to the stator while the single bolt solution will be manufatured as one of the last building steps. This creates the problem of how to test within the building process. This means: You´d like to test e.g. the membrane coating on function. Therefore You take just the one stator to which the membrane is glued and contact the membrane and stator to the HV-supply. When energized the membrane should bow towards the stator sheet. If not, of if its not doing so completely and evenly, you can correct for by doing a recoat.
So, I recommend using always 2 independent contact points, one for each stator.
With aluminium it could be a solution to use press-in bolts, since those deform the material and cog with the sheet material. Nylon is out of question here since the forces needed for a good contact are way too low.
What´s Your intention to anodize the aluminium? You have to insulate the sheets properly. Just anodizing -though it forms a isolation barrier- is way too less. You need to coat the sheets with a thicker and proper insulating material.
With steel sheets as stators on the other hand you can use either the bolt solution very well or You can solder a wire or contact point directly on to the sheet (ML and I do it this way). Another solution could be to weld a solderless express crimp terminal (3/16" or 1/4") to the sheet.
Copper tape is fine as diaphragm contact. If its pure copper than oxidizing can create problems over time. When putting the copper onto the diaphragm You should wet the tape with the coating fluid so that the fluid ´glues´ the tape to the membrane thereby creating a good solid large area contact on one hand and forms a anti-oxidizing barrier (top coat, like a laquer) on the other hand.
Why do You have just the 2 material-choices? Both don´t look optimal to me, because of their hole sizes. 0.1875 (+5mm) is slightly larger than You´d like to have for a mid-high-panel with a d/s around 1-1.5mm and 0.078" (~2mm) is very small. While smaller holes are basically a plus, they can create problems with the insulating coating. Besides the possiblity of completely sealed holes the reduction in open-area-percentage is higher than with larger holes. It seems that 1/8" (~3mm) holes on a 1/6" grid is the best compromise with regard to strength, openness, efficiency, coatabilzy and even optics. If You want to build something serious, I recommend to ´rethink´ certain points of Your design. If You just want to do a ´test´, than go ahead.
jauu
Calvin
ESLs are high impedance devices. As such the conductivity of the stators only plays a role above values of several kOhms (think of the resistor values needed to segment a panel). Regarding conductivity there will be no difference between aluminium and steel worth mentioning. BTW: make sure that the sheet´s holes are rounded on both sides of the sheet. Sanding and/or etching is strongly optional.
I wouldn´t recommend the usage of aluminium sheets for any other than very small tweeter-panels. Besides beeing too soft for larger panels a serious problem will be to create a long lasting stable electrical contact point. As you know aluminium oxidizes fast. A simple single bolt forming a contact by pressure on a pair of contact clips won´t last long and would ´compress´ the complete panel. A contact failure is preprogrammed.
Rather a 2-bolts solution, one bolt for each stator sheet could be feasable, since You could tighten each bolt very hard without exerting pressure on the panel itself. Additionally the 2-bolt solution must be manufactured before the spacers and membrane are glued to the stator while the single bolt solution will be manufatured as one of the last building steps. This creates the problem of how to test within the building process. This means: You´d like to test e.g. the membrane coating on function. Therefore You take just the one stator to which the membrane is glued and contact the membrane and stator to the HV-supply. When energized the membrane should bow towards the stator sheet. If not, of if its not doing so completely and evenly, you can correct for by doing a recoat.
So, I recommend using always 2 independent contact points, one for each stator.
With aluminium it could be a solution to use press-in bolts, since those deform the material and cog with the sheet material. Nylon is out of question here since the forces needed for a good contact are way too low.
What´s Your intention to anodize the aluminium? You have to insulate the sheets properly. Just anodizing -though it forms a isolation barrier- is way too less. You need to coat the sheets with a thicker and proper insulating material.
With steel sheets as stators on the other hand you can use either the bolt solution very well or You can solder a wire or contact point directly on to the sheet (ML and I do it this way). Another solution could be to weld a solderless express crimp terminal (3/16" or 1/4") to the sheet.
Copper tape is fine as diaphragm contact. If its pure copper than oxidizing can create problems over time. When putting the copper onto the diaphragm You should wet the tape with the coating fluid so that the fluid ´glues´ the tape to the membrane thereby creating a good solid large area contact on one hand and forms a anti-oxidizing barrier (top coat, like a laquer) on the other hand.
Why do You have just the 2 material-choices? Both don´t look optimal to me, because of their hole sizes. 0.1875 (+5mm) is slightly larger than You´d like to have for a mid-high-panel with a d/s around 1-1.5mm and 0.078" (~2mm) is very small. While smaller holes are basically a plus, they can create problems with the insulating coating. Besides the possiblity of completely sealed holes the reduction in open-area-percentage is higher than with larger holes. It seems that 1/8" (~3mm) holes on a 1/6" grid is the best compromise with regard to strength, openness, efficiency, coatabilzy and even optics. If You want to build something serious, I recommend to ´rethink´ certain points of Your design. If You just want to do a ´test´, than go ahead.
jauu
Calvin
arend-jan
Yes electrical forces are much stronger than gravitational forces, my frame to support the panels is made of oak that will be treated to avoid warping they measure 1.9cmX3.8cm *length (6feet).
Calvin
I was looking at steel perforated panels with specs
hole diameter 1/8"
% of open area 40%
Center to center spacing 3/26"
From McMaster-Carr Product number 9255T661. Your absolutely right one bolt would compress and compromise my D/S spacing. I will be using the 0.08" acrylic, would this be ok to use with the above metal?
IF I follow Sanders guidelines he says to satisfy:
2*stator thickness<hole diameter<D/S spacing
The steel I currently have (0.078" hole diameter 0.03" thick with a D/S spacing of 90 mil (acrylic plus glue layer) satisfies the inequality nicely, I understand that the lower bound is required so that the perforations look like holes not tunnels (Helmholtz’s resonators) the upper bound I thick is a result of not having the film be damped by reflections off the non-perforated part.
You are more experienced in building ESL's So I ask you, should I make one more trip to McMaster Carr, get the above metal (product number 9255T661) to be used with the 0.08" acrylic or stick with the steel I have that satisfies the Roger's inequality?
I will probably just solder a good contact directly to the steel.
You can have a look at the metal at:
www.mcmaster.com
and place product number,9255T661 in search box
Thank You Very Much,
Bryan
Yes electrical forces are much stronger than gravitational forces, my frame to support the panels is made of oak that will be treated to avoid warping they measure 1.9cmX3.8cm *length (6feet).
Calvin
I was looking at steel perforated panels with specs
hole diameter 1/8"
% of open area 40%
Center to center spacing 3/26"
From McMaster-Carr Product number 9255T661. Your absolutely right one bolt would compress and compromise my D/S spacing. I will be using the 0.08" acrylic, would this be ok to use with the above metal?
IF I follow Sanders guidelines he says to satisfy:
2*stator thickness<hole diameter<D/S spacing
The steel I currently have (0.078" hole diameter 0.03" thick with a D/S spacing of 90 mil (acrylic plus glue layer) satisfies the inequality nicely, I understand that the lower bound is required so that the perforations look like holes not tunnels (Helmholtz’s resonators) the upper bound I thick is a result of not having the film be damped by reflections off the non-perforated part.
You are more experienced in building ESL's So I ask you, should I make one more trip to McMaster Carr, get the above metal (product number 9255T661) to be used with the 0.08" acrylic or stick with the steel I have that satisfies the Roger's inequality?
I will probably just solder a good contact directly to the steel.
You can have a look at the metal at:
www.mcmaster.com
and place product number,9255T661 in search box
Thank You Very Much,
Bryan
Calvin,
The system is a hybrid, with transmission line woofer crossed over (passive) around 400Hz. The panels will not be full range just operating mids and highs. Phase cancellation will be reduced with a baffle design. The panel will measure 12"X40". It will be played in a small room (10'*12'), I usually don't listen loud, max SPL around 85dB a meter away. I will probably measure the frequency response and build a simple equalizer if needed, the baffle should be enough. I’m sure I left something out, just ask can provide.
Bryan
The system is a hybrid, with transmission line woofer crossed over (passive) around 400Hz. The panels will not be full range just operating mids and highs. Phase cancellation will be reduced with a baffle design. The panel will measure 12"X40". It will be played in a small room (10'*12'), I usually don't listen loud, max SPL around 85dB a meter away. I will probably measure the frequency response and build a simple equalizer if needed, the baffle should be enough. I’m sure I left something out, just ask can provide.
Bryan
Hi,
ok, thats what we needed. ;-)
Well, playing from 400Hz on up, the spacers should not exceed 1mm (1/25") thickness. 2mm (0.078") is too much already.
Your panel size and width is large enough to omit a baffle completely, which is always preferable in sonic terms. Use a thin but strong frame instead (besides: such a thin frame assists the acoustics witth optics, as it adds to the ´light and airy´ impression)
Keep in mind that even in a small room the panels need space around them. The min distance to the back wall should not be less than 40".
If You can manufacture, try to curve the panel. This will increase the stability of the panel very much --> less rattle, less resonance. This could be of elevated relevance if You use those thin MasterCarr sheets.
Since the d/s will be small and the panel area large, the capacitance will be high (between 1nF and 2nF, but so will be the efficiency. You will need a transformer or a pair of with a rather low transformer ratio of 1:50 to 1:75. The polarizing voltage can be quite low too (1-1.5kV). Lower voltages are very desirable because it means less problems with insulating and less losses, longer lasting materials and improved dynamics. You won´t need a very powerful amp either, if the bass doesn´t ask for.
Have You spent some thoughts on how to insulte the stator sheets? If not, then start doing and keep us informed!
jauu
Calvin
ok, thats what we needed. ;-)
Well, playing from 400Hz on up, the spacers should not exceed 1mm (1/25") thickness. 2mm (0.078") is too much already.
Your panel size and width is large enough to omit a baffle completely, which is always preferable in sonic terms. Use a thin but strong frame instead (besides: such a thin frame assists the acoustics witth optics, as it adds to the ´light and airy´ impression)
Keep in mind that even in a small room the panels need space around them. The min distance to the back wall should not be less than 40".
If You can manufacture, try to curve the panel. This will increase the stability of the panel very much --> less rattle, less resonance. This could be of elevated relevance if You use those thin MasterCarr sheets.
Since the d/s will be small and the panel area large, the capacitance will be high (between 1nF and 2nF, but so will be the efficiency. You will need a transformer or a pair of with a rather low transformer ratio of 1:50 to 1:75. The polarizing voltage can be quite low too (1-1.5kV). Lower voltages are very desirable because it means less problems with insulating and less losses, longer lasting materials and improved dynamics. You won´t need a very powerful amp either, if the bass doesn´t ask for.
Have You spent some thoughts on how to insulte the stator sheets? If not, then start doing and keep us informed!
jauu
Calvin
I agree with Calvin, about 1mm diaphragm-stator distance should be good and give a nice sensitivity (aprox. 95db/2.83V/1m with a stepup of 1:75, 1500V, depends on equalisation). This is good. But with 1mm, construction will be more difficult as you will need more diaphragm supports to get a stable diaphragm. The advantage of more supports is increased structural integrity, which you will need to get good enough tolerance (diaphragm - stator distance). A disadvantage is increased dead capacitance.
If you want to go ahead with the 2mm spacers then you can get aproximately the same sensitivity with a stepup of 1:100 and 5000V polarisation. As Calvin pointed out, this might give you electrical problems, but it can be done. And the higher the step-up, the more difficult it is to make a good transformer.
But I see two other problems with this design.
First, without segmentation a panel this size will beam like a laser! Unless you want to listen with your head in a vice, you need some form of segmentation. A popular way to achieve this (with perforated metal stators) is to split the stator into two pieces. One normal piece that covers most of the panel which handles the mid range, and a second piece in the form of a 1" wide vertical strip, running down the length of the panel on one side to handle the high frequencies.
Second, crossing over as high as 400Hz will make integration between the woofer and the panel much more difficult. At least this is my experience. I prefer to cross over below 100Hz.
If you want to go ahead with the 2mm spacers then you can get aproximately the same sensitivity with a stepup of 1:100 and 5000V polarisation. As Calvin pointed out, this might give you electrical problems, but it can be done. And the higher the step-up, the more difficult it is to make a good transformer.
But I see two other problems with this design.
First, without segmentation a panel this size will beam like a laser! Unless you want to listen with your head in a vice, you need some form of segmentation. A popular way to achieve this (with perforated metal stators) is to split the stator into two pieces. One normal piece that covers most of the panel which handles the mid range, and a second piece in the form of a 1" wide vertical strip, running down the length of the panel on one side to handle the high frequencies.
Second, crossing over as high as 400Hz will make integration between the woofer and the panel much more difficult. At least this is my experience. I prefer to cross over below 100Hz.
Thanks for getting back. I like the idea of using closer spacing so that I don’t need as much potential and less possible for electrical insulation problems to spoil the show. I already have this acrylic cut, they are double the size you suggest 80mil. I know I can still use it, just need greater voltages. SO to avoid the electrical problems my stator prep is as follow:
Step 1
Using a dremil (low speed) with cone shaped sander attached that just fits outside the hole diameter. This will round off the holes. This is tedious and time consuming.
Step 2. Use dremil to remove any sharp edges around corners from shearing.
Step 3. Sand both sides of the steel
Step 4. I have experimented with pc-board etching solution. This stuff is used to make circuit boards out of sheet copper. Agitate for an hour. This really does a nice job of cleaning the steel, dulls it out.
Step 5. Some kind of coating, maybe epoxy, went to home-depot they had some type of spray called plastic dip, works on steel, and gives insulation coating. Need to research its dielectric breakdown voltage per thickness of coating. Will get back.
Step 6. Testing, hook up stators separated by 2*D/S to high voltage supply, repeat steps if problems occur.
Or do you thick it would be wiser to just use 3M foam tape at 1mm thickness? It seems less work is involved. But I have Sanders book too much on my mind. I know it is not a good thing to stick hard to instructions, where’s the fun.
I don't know if you guys have RadioShacks near you, but I went to several near me and RadioShack Just isn't the same. I couldn't find any pots, or hook-up wire. Just batteries and cell-phones. Its a sad thing.
Some people, I thick older a-philes, like laser imaging. You have a tiny sweet spot but the imaging is fantastic, give and take. Curving the cell tries to widen that sweet spot, but too much dispersion you will lose imaging, because now the room is more involved. I opted for a slight curved cell. I have access to metal shop, so curving the steel is no problem. The toughest part will be tensioning the diaphragm just right. Maybe mech. jig.
So keep the acrylic or just use 3M?
Sincerely,
Bryan
Step 1
Using a dremil (low speed) with cone shaped sander attached that just fits outside the hole diameter. This will round off the holes. This is tedious and time consuming.
Step 2. Use dremil to remove any sharp edges around corners from shearing.
Step 3. Sand both sides of the steel
Step 4. I have experimented with pc-board etching solution. This stuff is used to make circuit boards out of sheet copper. Agitate for an hour. This really does a nice job of cleaning the steel, dulls it out.
Step 5. Some kind of coating, maybe epoxy, went to home-depot they had some type of spray called plastic dip, works on steel, and gives insulation coating. Need to research its dielectric breakdown voltage per thickness of coating. Will get back.
Step 6. Testing, hook up stators separated by 2*D/S to high voltage supply, repeat steps if problems occur.
Or do you thick it would be wiser to just use 3M foam tape at 1mm thickness? It seems less work is involved. But I have Sanders book too much on my mind. I know it is not a good thing to stick hard to instructions, where’s the fun.
I don't know if you guys have RadioShacks near you, but I went to several near me and RadioShack Just isn't the same. I couldn't find any pots, or hook-up wire. Just batteries and cell-phones. Its a sad thing.
Some people, I thick older a-philes, like laser imaging. You have a tiny sweet spot but the imaging is fantastic, give and take. Curving the cell tries to widen that sweet spot, but too much dispersion you will lose imaging, because now the room is more involved. I opted for a slight curved cell. I have access to metal shop, so curving the steel is no problem. The toughest part will be tensioning the diaphragm just right. Maybe mech. jig.
So keep the acrylic or just use 3M?
Sincerely,
Bryan
Jan
Forgot to ask you, the sensitivity you mentioned 95db/2.83V at a meter away is this something you measured? That sounds really nice.
About, Segmenting… The geometry of the spacer pattern (length and width of each cell) along with the films tension should define the films normal modes of vibrations, i.e. what frequencies they like to play. The way I understand it is that you design your spacer pattern, you suggested 1" wide tweeters down the whole length of the cell. The film is glued down along the spacer thus defining the segmented cell, is this what you are referring? If you look at ML's Spire or vantage you may notice that the width of each cell is something like, from top to bottom, wide, thinner, wide, thinner, etc... If you wanted to vary the tension of each cell, then were going to have to use several separate films, which means more contacts. Or do you just set the tension say a 100N below the tensile limit?
Sincerely,
Bryan
Forgot to ask you, the sensitivity you mentioned 95db/2.83V at a meter away is this something you measured? That sounds really nice.
About, Segmenting… The geometry of the spacer pattern (length and width of each cell) along with the films tension should define the films normal modes of vibrations, i.e. what frequencies they like to play. The way I understand it is that you design your spacer pattern, you suggested 1" wide tweeters down the whole length of the cell. The film is glued down along the spacer thus defining the segmented cell, is this what you are referring? If you look at ML's Spire or vantage you may notice that the width of each cell is something like, from top to bottom, wide, thinner, wide, thinner, etc... If you wanted to vary the tension of each cell, then were going to have to use several separate films, which means more contacts. Or do you just set the tension say a 100N below the tensile limit?
Sincerely,
Bryan
Sensitivity will depend on equalisation. The figure of 95db is a rough calculation at 1kHz and asuming voltage driver rather than current drive. In practice you are likely to get a little less.
The spacers will provide mechanical segmentation and provide enough support for the film not to collapse into the stators (stability). For the tweeter part you'd also need electrical segmentation, e.g. that small strip of the stator must be electrically isolated from the larger stator. With a series resistor in the audio drive you can now alter the frequency response (introduce roll off). To do this the film can be just one piece, no need for different tensions.
For the spacers I would prefer acrylic strips of some sort. Foam tape has a limited lifespan and might not give you the structural integrity you're looking for.
Some manufacturers vary the dimensions of the cells to spread the resonance over a larger frequency range. For this purpose you can also work with dots of silicone to provide mechanical segmentation (e.g. break up resonance modes).
About film tension, your primary concern would be to get a stable diaphragm. This depends on polarisation voltage and stator spacing. Normally you would also have to take into account the low frequency limit, but this is not a problem as your panel will only be playing from 400Hz upward.
I could not make up from your post what exactly it is that you were asking, but I hope this answers your questions.
The spacers will provide mechanical segmentation and provide enough support for the film not to collapse into the stators (stability). For the tweeter part you'd also need electrical segmentation, e.g. that small strip of the stator must be electrically isolated from the larger stator. With a series resistor in the audio drive you can now alter the frequency response (introduce roll off). To do this the film can be just one piece, no need for different tensions.
For the spacers I would prefer acrylic strips of some sort. Foam tape has a limited lifespan and might not give you the structural integrity you're looking for.
Some manufacturers vary the dimensions of the cells to spread the resonance over a larger frequency range. For this purpose you can also work with dots of silicone to provide mechanical segmentation (e.g. break up resonance modes).
About film tension, your primary concern would be to get a stable diaphragm. This depends on polarisation voltage and stator spacing. Normally you would also have to take into account the low frequency limit, but this is not a problem as your panel will only be playing from 400Hz upward.
I could not make up from your post what exactly it is that you were asking, but I hope this answers your questions.
The tape seems like a good idea in that it offers more flexibility in expermenting with D/S spacing, just peal the tape off. However, I worry about the longevity of it as oppose to acrylic. So I decided to use the tape as a means to experment, once could response has been acheived I will replace it with acrylic of same size, what do you guys think of my above stator prep?
Bryan
Bryan
Hi,
95dB@2.83V@1m is way out of the physical capabilities! 85dB is a already good value! But probabely AJ meant 95dB@2.83V@4m. When executed as strip source (line source) the SPL increases on the first 3-4m distance with increasing distance. You can reach the figure of 95dB, but only with a very elaborated metal sheet stator panel at a distance of 3-4m.
A good panel nowadays takes care of the distribution character. Imaging will still be excellent, but You´re not restricted to sit still like a mummy with Your ears fixed to a tiny micro spot! Changes of up to 20dB in amplitude within 1° of horizontal deviation are not a sign of good design!
A flat panel wo segmentation (i.e. flat metal stators) must be considered as inferior. Besides the very tight restrictions to the listening position the mechanical stability is low. You´d have to stabilize the panel somehow, but with the sheet´s advantage of thinness gone, there are superior alternatives like wire.
So if You want to stay with a flat panel, I recommend to change to wire stators with electrical segmentation. Easy to build, quite flexible with regard to eq´ing and easier to design ´safe´.
When using metal sheet stators the panel should be curved. Besides easing the sweetspot problem this considerably stabilizes the panel so that no additional bracing is needed. This gives You an easy and the thinnest possible solution with fine optics.
You only have to waste a thought or two about how to tension the membrane and which membrane material to use.
To take for building tolerances into account, the d/s of the ´backside´ should be slightly larger than the one to the ´frontside´. 1/10-2/10mm are fine (when the sheets are precisely curved and the insulative laquering is done 1st class). 3M offers the right tapes for this.
They are tightly tolerated in thickness. Acrylic spacers are not necessarily tolerated tighter in thickness, not to mention the tolerances of the ´handmade´ glue joints.
Ageing is no problem with the 3M brand and the ´softness´ helps in dampening membrane resonances and reduces ´rattle´. Actually the tapes completely miss out on the danger of ´cutting´ the membrane which could happen with stiff spacers and hard drying glues. The tapes are a clean and quick to work with solution and give a instantaneous strong bond, but ask for a calm hand and aren´t cheap.
You don´t neccessarily need to dremil each and every of the ~40.000 holes Sanding the rough side of the sheet (without pressure!!) and etching will do the job.
Its more important which laquer You use to insulate the sheets.
PVC, Nylon, PU and Acrylic are fine with regard to electrical parameters. Ask for a laquer that gives a nice even coating at roundings and edges (which rules out most water-based and one-component laquers). Be very sensitive with coloured laquers, especially grey and black ones, since those are often conductive under HV-conditions, eg. Krylon). If You use a conductive laquer, use it as a thin first or second layer and overcoat with several thick clear coats. A solvent based PU-Laquer used for coating boats or staircases is an excellent choice with regard to surface quality (optics), electrical parameters longevity and safety. When wet spray-coating expect to perform at least 8 runs on every side of the sheet. The best solution is to have the sheets thickly powder coated with a white Nylon powder, one thin layer of colour and a thick wet high gloss PU-coating. This yields ~0.4-0.5mm thick insulation that allows for voltage levels very close to field saturation (highest efficiency) and perfectly safe stators (up to 3-4 times the highest working voltages).
You use only one tension value for the complete panel. The spacer distance may vary slightly, though the claimed effect of ´resonance distribution´ can hardly be confirmed in praxis unless the distances vary widely and this may introduce other problems.
jauu
Calvin
95dB@2.83V@1m is way out of the physical capabilities! 85dB is a already good value! But probabely AJ meant 95dB@2.83V@4m.
When executed as strip source (line source) the SPL increases on the first 3-4m distance with increasing distance. You can reach the figure of 95dB, but only with a very elaborated metal sheet stator panel at a distance of 3-4m.A good panel nowadays takes care of the distribution character. Imaging will still be excellent, but You´re not restricted to sit still like a mummy with Your ears fixed to a tiny micro spot! Changes of up to 20dB in amplitude within 1° of horizontal deviation are not a sign of good design!
A flat panel wo segmentation (i.e. flat metal stators) must be considered as inferior. Besides the very tight restrictions to the listening position the mechanical stability is low. You´d have to stabilize the panel somehow, but with the sheet´s advantage of thinness gone, there are superior alternatives like wire.
So if You want to stay with a flat panel, I recommend to change to wire stators with electrical segmentation. Easy to build, quite flexible with regard to eq´ing and easier to design ´safe´.
When using metal sheet stators the panel should be curved. Besides easing the sweetspot problem this considerably stabilizes the panel so that no additional bracing is needed. This gives You an easy and the thinnest possible solution with fine optics.
You only have to waste a thought or two about how to tension the membrane and which membrane material to use.
To take for building tolerances into account, the d/s of the ´backside´ should be slightly larger than the one to the ´frontside´. 1/10-2/10mm are fine (when the sheets are precisely curved and the insulative laquering is done 1st class). 3M offers the right tapes for this.
They are tightly tolerated in thickness. Acrylic spacers are not necessarily tolerated tighter in thickness, not to mention the tolerances of the ´handmade´ glue joints.
Ageing is no problem with the 3M brand and the ´softness´ helps in dampening membrane resonances and reduces ´rattle´. Actually the tapes completely miss out on the danger of ´cutting´ the membrane which could happen with stiff spacers and hard drying glues. The tapes are a clean and quick to work with solution and give a instantaneous strong bond, but ask for a calm hand and aren´t cheap.
You don´t neccessarily need to dremil each and every of the ~40.000 holes
Sanding the rough side of the sheet (without pressure!!) and etching will do the job.Its more important which laquer You use to insulate the sheets.
PVC, Nylon, PU and Acrylic are fine with regard to electrical parameters. Ask for a laquer that gives a nice even coating at roundings and edges (which rules out most water-based and one-component laquers). Be very sensitive with coloured laquers, especially grey and black ones, since those are often conductive under HV-conditions, eg. Krylon). If You use a conductive laquer, use it as a thin first or second layer and overcoat with several thick clear coats. A solvent based PU-Laquer used for coating boats or staircases is an excellent choice with regard to surface quality (optics), electrical parameters longevity and safety. When wet spray-coating expect to perform at least 8 runs on every side of the sheet. The best solution is to have the sheets thickly powder coated with a white Nylon powder, one thin layer of colour and a thick wet high gloss PU-coating. This yields ~0.4-0.5mm thick insulation that allows for voltage levels very close to field saturation (highest efficiency) and perfectly safe stators (up to 3-4 times the highest working voltages).
You use only one tension value for the complete panel. The spacer distance may vary slightly, though the claimed effect of ´resonance distribution´ can hardly be confirmed in praxis unless the distances vary widely and this may introduce other problems.
jauu
Calvin
Hi Calvin,
I did a numerical solution of the generalised (for near field and off axis) Walker integral to calculate the theoretic SPL. Like I said, in pratice it's probably going to be a bit lower.
When executed as strip source (line source) the SPL increases on the first 3-4m distance with increasing distance.
I heard about this once before on a forum (maybe by you?) but never could find an explanation for this effect in the 'official' theoretic works. Nor did I ever measure (or notice) this effect on a real panel. Could you clarify the conditions that are required for this effect to take place? Do you know of a theory that explains it?
regards,
Arend-Jan
I did a numerical solution of the generalised (for near field and off axis) Walker integral to calculate the theoretic SPL. Like I said, in pratice it's probably going to be a bit lower.
When executed as strip source (line source) the SPL increases on the first 3-4m distance with increasing distance.
I heard about this once before on a forum (maybe by you?) but never could find an explanation for this effect in the 'official' theoretic works. Nor did I ever measure (or notice) this effect on a real panel. Could you clarify the conditions that are required for this effect to take place? Do you know of a theory that explains it?
regards,
Arend-Jan
Hi,
the effect of having a SPLmax around 3-4m (it depends on the length of the panel, but with usual sizes of 1m-1.5m the max. is at ~3-4m)
can be easily verified not only by listening but by measurement too.
The difference between 1m and 4m can reach up to 10dB.
I measured this effect myself and it was also measured at Aachen Unversity in their anechoic room.
jauu
Calvin
the effect of having a SPLmax around 3-4m (it depends on the length of the panel, but with usual sizes of 1m-1.5m the max. is at ~3-4m)
can be easily verified not only by listening but by measurement too.
The difference between 1m and 4m can reach up to 10dB.
I measured this effect myself and it was also measured at Aachen Unversity in their anechoic room.
jauu
Calvin
Calvin,
I do like that I don’t glue using the tape bonds, I’m giving it serious consideration. Plus using tape is simpler for building curved cells, since the tape complies well to any shape, with the acrylic heat strips would have to be used to get the desired curve, and this could be hit or miss. If that weren't bad enough keeping the glue layer as uniform as possible will be impossible. I think ‘Ockham’s Razor’ applies here well.
Thanks for that nice recipe for prepping the stators. Forget about drilling the holes, I practiced on scraps I find with motor tool you do more damage than you had to begin with. I will get the stators coated by someone professional.
Bryan
I do like that I don’t glue using the tape bonds, I’m giving it serious consideration. Plus using tape is simpler for building curved cells, since the tape complies well to any shape, with the acrylic heat strips would have to be used to get the desired curve, and this could be hit or miss. If that weren't bad enough keeping the glue layer as uniform as possible will be impossible. I think ‘Ockham’s Razor’ applies here well.
Thanks for that nice recipe for prepping the stators. Forget about drilling the holes, I practiced on scraps I find with motor tool you do more damage than you had to begin with. I will get the stators coated by someone professional.
Bryan
Hi
An effect that big (10dB) can not be overlooked in measurements.
So there must be a logical explanation why you are seeing this and I'm not. Perhaps it has to do with the curvature of your panels? Mine (Quad ESL panels) are slightly curved vertically (radius of about 3 meters), but otherwise flat.
regards,
Arend-Jan
An effect that big (10dB) can not be overlooked in measurements.
So there must be a logical explanation why you are seeing this and I'm not. Perhaps it has to do with the curvature of your panels? Mine (Quad ESL panels) are slightly curved vertically (radius of about 3 meters), but otherwise flat.
regards,
Arend-Jan
That is strange one would think that as you move further away from the sound source the levels would decrease. A lot of topic, reminds me of the corona heating problem. The surface of the sun is at a temperature of about 5800K as we move away from the surface the temperature increases to about 1.5 million kelvins. This is counter-intuitive as you move away from the heat source you would think that the temperature should decrease. I will do some thinking about this. Essentially the corona heating problem is one of the famous unsolved problems in solar-physics.
Jan, that is slight curvature. I imagine a cylinder of that radius, what angle did you curve through? I take it that with such a slight curvature you had little troubles tensioning the film as well as shaping the metal. Again thanks for all your input, maybe I will have pictures to share later. Also the etching solution cause the my sampled steel to rust slightly.
bryan
Jan, that is slight curvature. I imagine a cylinder of that radius, what angle did you curve through? I take it that with such a slight curvature you had little troubles tensioning the film as well as shaping the metal. Again thanks for all your input, maybe I will have pictures to share later. Also the etching solution cause the my sampled steel to rust slightly.
bryan
Hi,
AJ, are You speaking of the 57? Well, I guess You should be able to measure the quoted effect, but since the Quad is rather low in panel height, the effect should only be measurable up to a smaller distance, maybe 1m and at elevated frquencies. But since the Quad has a great building width and the bass panels work as baffles for the tweeter panel, it isn´t a real strip source(!) and the effect would be countered by the low freq-lift because of the nearfield-proximity effect. I guess the Quad might not exhibit this effect as clearly as real strip sources like Audiostatics, bigger MLs or Finals with panels much longer than the Quad´s (the panle size has to be larger than the wavelength to produce the effect).
jauu
Calvin
AJ, are You speaking of the 57? Well, I guess You should be able to measure the quoted effect, but since the Quad is rather low in panel height, the effect should only be measurable up to a smaller distance, maybe 1m and at elevated frquencies. But since the Quad has a great building width and the bass panels work as baffles for the tweeter panel, it isn´t a real strip source(!) and the effect would be countered by the low freq-lift because of the nearfield-proximity effect. I guess the Quad might not exhibit this effect as clearly as real strip sources like Audiostatics, bigger MLs or Finals with panels much longer than the Quad´s (the panle size has to be larger than the wavelength to produce the effect).
jauu
Calvin
Hi,
I've got audiostatics myself.
Never heard of any increase in soundpressure at a greater distance. Certainly not in mono.
My book of physics (which explains theories of Huygens and Fresnel) doesn't describe this phenomenum.
The only thing what came in m y mind is a kind of interference pattern which amplifies sound upon distance.
Calvin, did you measure some frequency anomalies as well?
Are there minima as well?
I've got audiostatics myself.
Never heard of any increase in soundpressure at a greater distance. Certainly not in mono.
My book of physics (which explains theories of Huygens and Fresnel) doesn't describe this phenomenum.
The only thing what came in m y mind is a kind of interference pattern which amplifies sound upon distance.
Calvin, did you measure some frequency anomalies as well?
Are there minima as well?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.