.
Brian, here's the email I got from the Tiger Drylac rep:
Bill, that coating, as most are not conductive. Typically they are insulators and vary in dielectrics up to in some cases 1000v per mil. I’m not sure if that one is that high, but it is not conductive in a cured state. We do have conductive or what is more accurately called Electrostatic Dissipative coatings, but generally they are only black and in a very fine texture.
Bill, that coating, as most are not conductive. Typically they are insulators and vary in dielectrics up to in some cases 1000v per mil. I’m not sure if that one is that high, but it is not conductive in a cured state. We do have conductive or what is more accurately called Electrostatic Dissipative coatings, but generally they are only black and in a very fine texture.
Bill,
It's hard to say from what the rep said. Take a look at an earlier posting of mine:
Volume resistivities
You can see that all these materials would normally be considered as insulators, but in the high impedance game of ESLs, some are better insulators than others. Since some conductivity is required, PVC was preferred over Tefton, for example, for its 1000 times lower resisitivity. I don't know where your powder coating would fit into this spectrum, but it may well be conductive enough to work fine, no matter what the rep said. If in doubt, order the black carbon-loaded stuff like Martin Logan.
It's hard to say from what the rep said. Take a look at an earlier posting of mine:
Volume resistivities
You can see that all these materials would normally be considered as insulators, but in the high impedance game of ESLs, some are better insulators than others. Since some conductivity is required, PVC was preferred over Tefton, for example, for its 1000 times lower resisitivity. I don't know where your powder coating would fit into this spectrum, but it may well be conductive enough to work fine, no matter what the rep said. If in doubt, order the black carbon-loaded stuff like Martin Logan.
I remember reading that thread when I first started planning an ESL build. Very interesting information. I looked into doing egg crate lighting louver/wire stators using the PVC wire you mentioned. Might still try one in the future.
It should be easy to try different panels out once the audio transformers, hv power supply, and electronic crossovers are in hand.
So I wonder if insulating the stators is necessary at all? Other than providing a barrier to keep fingers from directly contacting bare metal at high voltage, what purpose does insulation serve? It would keep a steel stator from rusting, but perforated aluminum or copper wire would not need that. I'm going to try dust covers as suggested by I_F. They should keep fingers out. Just thinking out loud.
It should be easy to try different panels out once the audio transformers, hv power supply, and electronic crossovers are in hand.
So I wonder if insulating the stators is necessary at all? Other than providing a barrier to keep fingers from directly contacting bare metal at high voltage, what purpose does insulation serve? It would keep a steel stator from rusting, but perforated aluminum or copper wire would not need that. I'm going to try dust covers as suggested by I_F. They should keep fingers out. Just thinking out loud.
Besides safety, the other advantage of resistive insulation on stators is to limit current in the event that arcing to the diaphragm occurs. But I think a case can be made for uninsulated stators under certain circumstances. I've experimented with bare stainless steel mesh in a wide gap design where voltages and excursions were unlikely to provoke arcing. But do provide some kind of protection against fingers getting onto the stators. The high signal voltages driven from the relatively low source impedance of the step-up transfomer could be deadly.
I don't think you should necessarily run away from powder coating. Maybe you could do some tests as you suggested. If nothing else, coat some test surfaces with several powder coating options and measure the resistivity, if you have a mega-ohmeter available.
I don't think you should necessarily run away from powder coating. Maybe you could do some tests as you suggested. If nothing else, coat some test surfaces with several powder coating options and measure the resistivity, if you have a mega-ohmeter available.
I got out a VTVM that will measure up to 1000 Megohms and tried to check resistance of the powder coating. With one lead on the stator's electrical connection screw and the other lead touched to a nickel placed an inch away from the screw, the needle didn't move at all. I've got a very good insulator.
So I tried something else. It's been dry here and I noticed I can build up a static charge just by sliding off my computer chair's pad. One slide pulls a 1/8" spark to a grounded object and two slides gives me even more. I connected the stator screw to ground, did a chair slide and pointed to the powder coated surface. Ouch, the spark jumped from my finger to the flat surface of the stator as if the powder coat wasn't there. Three more tries in different places did the same thing.
If my finger test is any indication, the powder coating should be OK.
So I tried something else. It's been dry here and I noticed I can build up a static charge just by sliding off my computer chair's pad. One slide pulls a 1/8" spark to a grounded object and two slides gives me even more. I connected the stator screw to ground, did a chair slide and pointed to the powder coated surface. Ouch, the spark jumped from my finger to the flat surface of the stator as if the powder coat wasn't there. Three more tries in different places did the same thing.
If my finger test is any indication, the powder coating should be OK.
I've been pondering the insulation issue and, unfortunately, can't claim to fully understand it. I understand how an insulating coating that has a large resistance will diminish the DC field experienced by the diaphragm. It's less clear to me what would happen to the AC part of the field. I've been trying to think about whether you need to consider an RC time constant for the assembly, but haven't come to any useful conclusions.
In an effort to work around my ignorance I wondered whether another approach might work. Perhaps an electrically insulating but acoustically transparent thin spacer could be placed between an uninsulated stator and the diaphragm. This would prevent the diaphragm from electrically contacting the stator but might reduce the voltage divider effects of typical stator coatings. Does anyone know if there's a precedent for this approach, or perhaps whether it's just a flawed idea? I was envisioning either some very sheer material of the type used for nylon stockings, or perhaps something with a much more open weave so that there's as little dielectric as possible between the stator and the diaphragm. As long as the holes in the weave aren't so large that the diaphragm can sneak through and make contact with the stator the weave could be very open. The porous spacer would have to be very thin so you don't lose too much diaphragm/spacer gap, and it would have to be firmly attached to the stator so it doesn't flop around or resonate. Those don't seem like insurmountable problems. Does anyone see any hope for this approach?
Few
In an effort to work around my ignorance I wondered whether another approach might work. Perhaps an electrically insulating but acoustically transparent thin spacer could be placed between an uninsulated stator and the diaphragm. This would prevent the diaphragm from electrically contacting the stator but might reduce the voltage divider effects of typical stator coatings. Does anyone know if there's a precedent for this approach, or perhaps whether it's just a flawed idea? I was envisioning either some very sheer material of the type used for nylon stockings, or perhaps something with a much more open weave so that there's as little dielectric as possible between the stator and the diaphragm. As long as the holes in the weave aren't so large that the diaphragm can sneak through and make contact with the stator the weave could be very open. The porous spacer would have to be very thin so you don't lose too much diaphragm/spacer gap, and it would have to be firmly attached to the stator so it doesn't flop around or resonate. Those don't seem like insurmountable problems. Does anyone see any hope for this approach?
Few
Conductivity
Hi,
I don´t agree with Brian´s opinion about material parameters.
Conductivity:
The insulator should do what it´s supposed to do in first place -- insulate not conduct. But a slight leakage could be positive in one (!) case -- overload.
In case the diaphragm hits a stator charge will be deposited on the stators surface, thereby reducing reducing SPL at this point. The charge will leak away with a time constant RxC that in parts depends on the conductivity of the insulation (volume resistivity), thereby restoring SPL to the original level. The time constant can reach hours with highly insulative materials like PE or PTFE (10^17Ohms/cm). PVC and Nylon have a much lower resistance (10^14 Ohms/cm) so the time constant will be smaller by a factor of app. 1/1000. Recovery time is then a matter of a few seconds or less.
But as long as the panel works under normal conditions the conductivity of the insulator is of no concern.
Permitivity:
A completely different point regards permitivity (epsilon).
As it happens most highly insulative materials have low permitivity values, with some derivatives of PTFE beeing close to 1, the value of vacuum and air. For applications where extremly low dielectric losses are essential Teflon (PTFE) is a very good choice, e.g. LF-cable insulation. In case of an ESL Teflon is far from optimum because of its low permitivity.
For DC-conditions the permitivity is of no concern, but this alters under AC-conditions. The stator metal and the diaphragm form the plates of a capacitor with a dielectric consisting of stacked layers of air and of insulator. This can be regarded as two series connected capacitors. This circuit forms a voltage divider for the music signal!
To have as little signal over the insulator capacitor and as much signal over the air capacitor the capacitance of the insulation capacitor must be much higher than that of the air capacitor.
Imagine a insulation of 1mm thickness and a gap of 1mm of air between insulation and diaphragm (max. linear throw | This is not d/s, which is always used for the distance between the metal surface and the diaphragm!). Imagine further two different insulative materials one with a permitivity of 2 and one with a permitivity of 9. In the first case 1/3 of the signal voltage could be measured over the insulator and only 2/3 could be used as working voltage! This is 33% of losses.
In the second case 1/10 of the voltage is ´wasted´ and 9/10 can be ´used´. The losses are reduced to 10%.
With C proportional to the permitivity of the material and size of the surface and inverse proportional to the thickness of the material we can only reach high values by thin layers and/or high permitivity (since the surface of the insulator and the air dielectric beeing equal, the surface drops out of the equation).
But we can´t make the insulation too thin, because we need high values of safety against flashovers and sparks. Typically the insulator is between 0.3mm and 1.0mm thick (Beveridge beeing a case of its own, using very thick insulators with very high permitivity -for ESL standards). Most insulators permitivity values range below 10. Teflon beeing among the materials with the lowest values would lead to reduced SPL simply because of the high AC signal-loss!!
On the higher side reside PVC and Nylon, both beeing around a permitivity of 5. So with an thickness of 0.5mm (which is enough to withstand several kV and which is quite the thickness with ML-Panels) and an air gap of 1.0mm, we have approximately 10% of AC-signal losses. The interesting thing is that both PVC and Nylon have lower volume resistivies too, so they fulfill both of the discussed demands. I regard Nylon slightly superior to PVC but PVC is a more common and cheaper powder.
jauu
Calvin
Hi,
I don´t agree with Brian´s opinion about material parameters.
Conductivity:
The insulator should do what it´s supposed to do in first place -- insulate not conduct. But a slight leakage could be positive in one (!) case -- overload.
In case the diaphragm hits a stator charge will be deposited on the stators surface, thereby reducing reducing SPL at this point. The charge will leak away with a time constant RxC that in parts depends on the conductivity of the insulation (volume resistivity), thereby restoring SPL to the original level. The time constant can reach hours with highly insulative materials like PE or PTFE (10^17Ohms/cm). PVC and Nylon have a much lower resistance (10^14 Ohms/cm) so the time constant will be smaller by a factor of app. 1/1000. Recovery time is then a matter of a few seconds or less.
But as long as the panel works under normal conditions the conductivity of the insulator is of no concern.
Permitivity:
A completely different point regards permitivity (epsilon).
As it happens most highly insulative materials have low permitivity values, with some derivatives of PTFE beeing close to 1, the value of vacuum and air. For applications where extremly low dielectric losses are essential Teflon (PTFE) is a very good choice, e.g. LF-cable insulation. In case of an ESL Teflon is far from optimum because of its low permitivity.
For DC-conditions the permitivity is of no concern, but this alters under AC-conditions. The stator metal and the diaphragm form the plates of a capacitor with a dielectric consisting of stacked layers of air and of insulator. This can be regarded as two series connected capacitors. This circuit forms a voltage divider for the music signal!
To have as little signal over the insulator capacitor and as much signal over the air capacitor the capacitance of the insulation capacitor must be much higher than that of the air capacitor.
Imagine a insulation of 1mm thickness and a gap of 1mm of air between insulation and diaphragm (max. linear throw | This is not d/s, which is always used for the distance between the metal surface and the diaphragm!). Imagine further two different insulative materials one with a permitivity of 2 and one with a permitivity of 9. In the first case 1/3 of the signal voltage could be measured over the insulator and only 2/3 could be used as working voltage! This is 33% of losses.
In the second case 1/10 of the voltage is ´wasted´ and 9/10 can be ´used´. The losses are reduced to 10%.
With C proportional to the permitivity of the material and size of the surface and inverse proportional to the thickness of the material we can only reach high values by thin layers and/or high permitivity (since the surface of the insulator and the air dielectric beeing equal, the surface drops out of the equation).
But we can´t make the insulation too thin, because we need high values of safety against flashovers and sparks. Typically the insulator is between 0.3mm and 1.0mm thick (Beveridge beeing a case of its own, using very thick insulators with very high permitivity -for ESL standards). Most insulators permitivity values range below 10. Teflon beeing among the materials with the lowest values would lead to reduced SPL simply because of the high AC signal-loss!!
On the higher side reside PVC and Nylon, both beeing around a permitivity of 5. So with an thickness of 0.5mm (which is enough to withstand several kV and which is quite the thickness with ML-Panels) and an air gap of 1.0mm, we have approximately 10% of AC-signal losses. The interesting thing is that both PVC and Nylon have lower volume resistivies too, so they fulfill both of the discussed demands. I regard Nylon slightly superior to PVC but PVC is a more common and cheaper powder.
jauu
Calvin
Few,
Yes, in fact, there are two time constants. At low frequencies, there is a resistive voltage divider reducing the stator voltage to what appeears across the air gap. At high frequencies, there is a capacative divider, which yields another divider ratio across the gap. In between, there will a pole and a zero with a 6 dB per octave slope (asymptote) between them. If the resistance of the stator coating is many orders of magitude lower than the air gap and the capacitive reactance of the insulator it will "short out" the a parallel cap, and put all the staor voltage across the air gap, it will push the time constants and slope way higher in frequency.
I was working on a more detailed explanation for Bazukas, but didn't havr time to finish it.
Yes, in fact, there are two time constants. At low frequencies, there is a resistive voltage divider reducing the stator voltage to what appeears across the air gap. At high frequencies, there is a capacative divider, which yields another divider ratio across the gap. In between, there will a pole and a zero with a 6 dB per octave slope (asymptote) between them. If the resistance of the stator coating is many orders of magitude lower than the air gap and the capacitive reactance of the insulator it will "short out" the a parallel cap, and put all the staor voltage across the air gap, it will push the time constants and slope way higher in frequency.
I was working on a more detailed explanation for Bazukas, but didn't havr time to finish it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.