Yep, the sparks are very evident with alumylar
. I gave the weak spots generous coats of 2kV/mil conformal coating yesterday and after that I applied two more layers of urethane-alkyd all over.
The panels are these: Building an Electrostatic Speaker
. I gave the weak spots generous coats of 2kV/mil conformal coating yesterday and after that I applied two more layers of urethane-alkyd all over.The panels are these: Building an Electrostatic Speaker
The more I read about coating mesh, the more I feel that Kynar wire staters may be an easier option to give good results. I like the look of mesh staters better than wire ones, but it is the sound quality that matters most.
What type of coating is on the MartinLogan staters? If they are selling them to the general public at the price they are, they surely are happy that their stator coating will be good for years, as they don't want people returning them all the time.
What type of coating is on the MartinLogan staters? If they are selling them to the general public at the price they are, they surely are happy that their stator coating will be good for years, as they don't want people returning them all the time.
Yes, I would like to give it a try as well!!!
A few have used it and say it works well.
I had tested the wire back when I built my first panels and it withstood at least 10Kv to 12Kv or so!!
At least it never broke down with the supply I was using.
I didn't use it because it was at the time when the price of wire just went through the roof and Radio shack was charging (and still do) a ridiculous price for a little roll of it.
In a search I have been able to locate a 1000' spool of it for about $30 to $50 and that is the best deal I could find so far.
The trickiest part about using mesh is properly sealing the sharp cut edges.
They just need to be buried deep in the silicone is all.
I will have to start a new thread sometime on how to use this method.
I would like to take it a step farther and do a electrically segmented version using the mesh method as well.
I am not positive, but, I think ML uses Powder coating on their stators.
Here is an excerpt from this post,
http://www.diyaudio.com/forums/inst...peaker-reproduce-piano-sound.html#post2547936
As a testament to how well the little black mesh ones sounded before I had a woofer system matched up to them,
"Before I built my first panels I had listened to some ML's and my very first panels were not any different as far as midrange and high frequency's are concerned and the sound quality is second to none IMHO.
They sounded so accurate that one time I was listening to a cassette recording of some rain and forest sounds and my dog would go to the door to watch the rain.
But she would not go outside for fear of getting wet when there was NO RAIN (ha,ha,ha).
She is now dubbed the Infamous Sadie in these threads."
Seriously it was not raining but she thought it was.
You see she is a Florida dog and it rained every day their and she used to love to sit and look out the screen door and watch it pour!!!
This was before I had a real amp and transformer's.
I was using a cheapy ole LXI 35 watt amp and a pair of 6V6 p-p output transformers.
They finally shorted out and then I didn't fire them back up until 2010.
My little black ones were double powder coated and that was it.
Until 2010 when I refurbished them and gave them quite a few coats of clear acrylic enamel.
And those are the infamous ones that took all of the extreme voltages.
jer
A few have used it and say it works well.
I had tested the wire back when I built my first panels and it withstood at least 10Kv to 12Kv or so!!
At least it never broke down with the supply I was using.
I didn't use it because it was at the time when the price of wire just went through the roof and Radio shack was charging (and still do) a ridiculous price for a little roll of it.
In a search I have been able to locate a 1000' spool of it for about $30 to $50 and that is the best deal I could find so far.
The trickiest part about using mesh is properly sealing the sharp cut edges.
They just need to be buried deep in the silicone is all.
I will have to start a new thread sometime on how to use this method.
I would like to take it a step farther and do a electrically segmented version using the mesh method as well.
I am not positive, but, I think ML uses Powder coating on their stators.
Here is an excerpt from this post,
http://www.diyaudio.com/forums/inst...peaker-reproduce-piano-sound.html#post2547936
As a testament to how well the little black mesh ones sounded before I had a woofer system matched up to them,
"Before I built my first panels I had listened to some ML's and my very first panels were not any different as far as midrange and high frequency's are concerned and the sound quality is second to none IMHO.
They sounded so accurate that one time I was listening to a cassette recording of some rain and forest sounds and my dog would go to the door to watch the rain.
But she would not go outside for fear of getting wet when there was NO RAIN (ha,ha,ha).
She is now dubbed the Infamous Sadie in these threads."
Seriously it was not raining but she thought it was.
You see she is a Florida dog and it rained every day their and she used to love to sit and look out the screen door and watch it pour!!!
This was before I had a real amp and transformer's.
I was using a cheapy ole LXI 35 watt amp and a pair of 6V6 p-p output transformers.
They finally shorted out and then I didn't fire them back up until 2010.
My little black ones were double powder coated and that was it.
Until 2010 when I refurbished them and gave them quite a few coats of clear acrylic enamel.
And those are the infamous ones that took all of the extreme voltages.
jer
Last edited:
Yep, That will work good.
My little one got to about 105db at 6Kv peak and 1 meter.
A larger panel will be much louder than this.
Figure about 6db for every doubling of the surface area.
This calculator will help you to get a better idea,
Electrostatic Loudspeaker (ESL) Simulator
jer
My little one got to about 105db at 6Kv peak and 1 meter.
A larger panel will be much louder than this.
Figure about 6db for every doubling of the surface area.
This calculator will help you to get a better idea,
Electrostatic Loudspeaker (ESL) Simulator
jer
I seem to recall last time looking into Kynar wire insulation and the bulk resistivity was around 1e14 (ohm-cm), which is pretty much in the range of most PVC insulation compounds isn't it? Nothing like Teflon with bulk resistivity in the 1e18 to 1e20 (ohm-cm) range.
Interesting about the piezoelectric effect...
I am curious where your 10^12 number came from for PVC, did you measure it?
I am still working on a reliable way to measure bulk resisitivities in this range.
Internet searches for bulk resistivity values of PVC wire insulation yields range of 10^12 to 10^15 depend on the exact compound. Does this seem reasonable to you?
I can borrow nanoammeter but what I could of actually measure is a surface resistivity. To measure bulk one need a thin film I guess plus pretty good controlled environment to exclude surface water and such and that's rather hard to get your hands on.
EDIT
On the other end you can submerge a long spool into a salted water and measure it for yourself but, and I do not like that word, you need to make sure that isolation is defect free...
EDIT
On the other end you can submerge a long spool into a salted water and measure it for yourself but, and I do not like that word, you need to make sure that isolation is defect free...
Last edited:
Hi,
I wondered the same as bolserst. Different sources, different formulations/dopings, different measurement setups, different numbers.
The only reliable thing I could find is that PVC and PU are rather on the lower side of volume and surface resistivity and PE, PP and various FEPs and PTFEs are rather on the high side.
jauu
Calvin
I wondered the same as bolserst. Different sources, different formulations/dopings, different measurement setups, different numbers.
The only reliable thing I could find is that PVC and PU are rather on the lower side of volume and surface resistivity and PE, PP and various FEPs and PTFEs are rather on the high side.
jauu
Calvin
I can get .5mm^2 or .6mm^2 PVC coated wire pretty cheap. The PVC insulation is 0.3mm, which is about 12mills. PVC is said to withstand 500v per mill, so hopefully this should work well for a first prototype build
Considering you want some of the electrical charge to escape the insulation and be in the air gap, can your sprayed on insulation, or wire insulation be too thick.
I know you can just up the voltage on the membrane to compensate, but wouldn't it be just as good or better to have a lower voltage, and a thinner coating/PVC.
For example I see 600 V PVC wire (thicker PVC jacket) and 300 V PVC wire (thinner PVC jacket). If the 300 V. wire has enough thickness to not arc at my high voltage working point it will be preferred to the 600 V. product right?
Thanks,
Paul
Hi,
the issue is explained quite well, I think, in the 73s patent of Beveridge.
In that a stator is described, which is claimed to be alot more flashover secure and efficient at the same.
In short does it view at the behaviour of the electrical field and the flashover behaviour independently.
For a high flashover treshold the thickness of the stator counts, as this reduces the voltage stress gradient [V/mm or V/in] within the stator crossection.
This also regards the Q of the resistive value of the stator material and the airgap.
In a schematic this would be represented by two series connected
This implies at first glance that the volume resistivity of the stator isn't an issue at all, as a too low voltage gradient in the airgap just required to raise the Bias voltage.
On second thought, one will realize that the volume resistance should be prefferrably low to keep the voltage gradient within the stator material sufficiently low and to reduce overlad recovery time.
AC-wise -or in terms of the electrical field- the equivalent circuit is that of two series connected caps.
They form a AC voltage divider where the voltage distribution is inverse proportional to the cap value.
To aquire as much voltage over the airgap, and to loose as few Volts over the stator, the dielectric constant must be high, preferrably >10 times the airgap capacitance.
Beveridge suggested to mix a blend of conductive and high-K dielectrics with a (fluid) base material of high resistivity and low K, that could be cast into the desired shape to form a material with the improved R and K.
In thecequivalent schematic the resistor chain and the cap chain were paralleled.
In other words, this forms a frequency dependent resistor with a -not too high- DC resistance and low AC resistance.
A matter which J.Strickland also has written a note about.
jauu
Calvin
the issue is explained quite well, I think, in the 73s patent of Beveridge.
In that a stator is described, which is claimed to be alot more flashover secure and efficient at the same.
In short does it view at the behaviour of the electrical field and the flashover behaviour independently.
For a high flashover treshold the thickness of the stator counts, as this reduces the voltage stress gradient [V/mm or V/in] within the stator crossection.
This also regards the Q of the resistive value of the stator material and the airgap.
In a schematic this would be represented by two series connected
This implies at first glance that the volume resistivity of the stator isn't an issue at all, as a too low voltage gradient in the airgap just required to raise the Bias voltage.
On second thought, one will realize that the volume resistance should be prefferrably low to keep the voltage gradient within the stator material sufficiently low and to reduce overlad recovery time.
AC-wise -or in terms of the electrical field- the equivalent circuit is that of two series connected caps.
They form a AC voltage divider where the voltage distribution is inverse proportional to the cap value.
To aquire as much voltage over the airgap, and to loose as few Volts over the stator, the dielectric constant must be high, preferrably >10 times the airgap capacitance.
Beveridge suggested to mix a blend of conductive and high-K dielectrics with a (fluid) base material of high resistivity and low K, that could be cast into the desired shape to form a material with the improved R and K.
In thecequivalent schematic the resistor chain and the cap chain were paralleled.
In other words, this forms a frequency dependent resistor with a -not too high- DC resistance and low AC resistance.
A matter which J.Strickland also has written a note about.
jauu
Calvin
Thanks Calvin,
I (usually) always enjoy your answers as they show me how little I know.
If we could boil it down just a little...
If I'm planning on HV running at ~5000V.
PVC~ 550 V / mil
300 V PVC is 16 mils thick = ~8800V.
This should be more than adequate.
600V PVC wire at 32 mils thick = ~ 17000V
I Re read Jim Stricklands white paper, and he describes the charge being held like a ballast in the thickness of the PVC. Not sure if thats good or bad, so ...
If 300 V wire is thick enough to prevent arc overs, is it better than 600 V rated wire? (or visa versa).
Thanks,
Paul
I (usually) always enjoy your answers as they show me how little I know.
If we could boil it down just a little...
If I'm planning on HV running at ~5000V.
PVC~ 550 V / mil
300 V PVC is 16 mils thick = ~8800V.
This should be more than adequate.
600V PVC wire at 32 mils thick = ~ 17000V
I Re read Jim Stricklands white paper, and he describes the charge being held like a ballast in the thickness of the PVC. Not sure if thats good or bad, so ...
If 300 V wire is thick enough to prevent arc overs, is it better than 600 V rated wire? (or visa versa).
Thanks,
Paul
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.