ESL powder coating?

I've finally taken the plunge...
Parts have been ordered and soon I'll start tinkering with my first set of ESL's.

I've read some previous posts and I'm trying to figure out the powdercoating bit.
300-500um seem like a good place to start and I found a local shop that can do 250-450um without much difficulties.

Is there anyting specific I should look for?
What thickness is "best"?

The stators are often recommended to have 40-50% open area.
Using Rv3-4 stators (51%) and a 300um coating will give you roughly 32.65% open area.
Rv5-6 (63%) is special order = VERY expensive -> 48.8% open area after coating. (300um)
Rv4-6 (43%) -> 29% open area after coating. (300um)

keeping with recommendations I should opt for the Rv5-6 perf metal but it would be insainly expensive and I'm curious how much the pattern would affect the efficiency of the panels?

Rv3-4 with 170um coating would give me 40% open area and be much much cheaper.

What would you do?
 
Last edited:
Hi,

between what one claims and what one can actually do there are sometimes whole worlds. :rolleyes: I have way so often heard from coaters in before, that it wouldn´t be any difficult....but in the end none of them could do the job properly.
250µ-400µm powder coating on punched metal sheet with an opening of app. 50% is not at all trivial. In fact it requires at least double- maybe even triple-coating, because a single layer of electrostatic spray coating is good only for 100-150µm based on Nylon/Rilsan, which is asumably the best powder material for our application. Since the powder layer needs to completely melt to achieve a pinhole free surface, a second or even third layer is difficult to apply.
Fluidized bed coating theoretically allows for sufficient layer thickness in one go, but the metal sheet cools down too quick for the powder to melt fully. It probabely requires some additional heat treatment process.
I doubt that a coater would be so lucky to find the right process parameters with the first try.
The way of least problems is possibly to have one good and thick layer spray coated with pure Nylon and add multiple layers of wet coating, with the first one thin and in the desired color and the following layers adding thick clear top-coats.
Rv3/4 is a good one. Punched metal sheets are best for stators with small d/s values, which in turn asks for small holes.

jauu
Calvin
 
Last edited:
He offered to measure the thickness if I needed to be sure.
It's a big industrial coater with very large capacity.
They are used to coating perf metal.
Multiple layers was doable without much trouble.
All acording to the person with whom I spoke.

I like Rv3-4 as well. (Cheap and easy to get + good open area ratio)
If the powder coater holds true to his word and sucessfully adds a 300um coating it'll result in 32.65% open area, is that ok?
Or should I go for a 170um coating to get the recommended 40% open area?
All assuming he can do it.
If he can't do it I think you are very much correct it's a good alterative way of doing it. :)
 
...Nylon/Rilsan, which is asumably the best powder material for our application.

Why would that be the case? A thermoplastic will never have the low viscosity of an epoxy or polyester, so will not flow or level as well to give a thin and very even coating.

FWIW, I worked with a derivative of Kynar which had astounding adhesion to metal as a powder coating. This material was also made by Arkema. Kynar has much lower moisture regain than nylon, as well as superior insulating properties. It is, sadly, quite expensive.
 
Hi,

Kynar, as You suggested SY, would indeed be a alternative by it´s electrical properties. One can test this by using Kynar insulated wire-wrap wire for stator construction (and its silver plated copper wire-----so it must sound well *lol*).
I see two pros for Nylon/Rilsan. One is the rather lowish surface- and bulk-resistivity which reduces the tendency of charge buildup on the stator, reduces overload recovery time and which -in combination with it´s epsilon value- allows for a good efficiency.
The second is the low value of radius for equal coating thickness which allows for a slightly thicker layer at the whole edges.

jauu
Calvin

btw. pure Nylon is white in color
 
Last edited:
Hi,

Kynar, as You suggested SY, would indeed be a alternative by it´s electrical properties. One can test this by using Kynar insulated wire-wrap wire for stator construction (and its silver plated copper wire-----so it must sound well *lol*).
I see two pros for Nylon/Rilsan. One is the rather lowish surface- and bulk-resistivity which reduces the tendency of charge buildup on the stator, reduces overload recovery time and which -in combination with it´s epsilon value- allows for a good efficiency.

Hi Calvin,

Can you share some specifics for bulk resistivity and dielectric constant for the Nylon/Rilsan mix you feel has advantages?


For comparison, here are values I found on the web for Kynar and PVC.

Kynar electrical properties:
Kynar bulk resistivity: 1.0e14 - 2.0e14 (ohms-cm) (dependent on Kynar type)
Kynar dielectric constant: 7.3 - 9.5 (depending on signal frequency and Kynar type)

PVC electrical properties:
PVC bulk resistivity: 1.0e13 - 1.0e15 (ohms-cm) (dependent on PVC type)
PVC dielectric constant: 3.5 - 8.0 (dependent on signal frequency and PVC type)
 
Last edited:
Please explain the mechanics at work here?
Resistivity being the inverse of conductivity should be as high as possible in an insulator?
Why would a low value be positive?

The dielectric constant should be low though to avoid charge build-up as much as possible? (If I'm remembering my physics right?)
Since it's also frequency dependant I'm curious as to how this will affect things in the audio band?

A low radius value sounds like a good thing for coating holes with small radius such as ours.

Thinking one step further...
Double the resistivity would allow for half the thickness of the coating with preserved insulation. With less dielectric it's also possible that this would lessen the charge buildup as well?
What am I missing here?

Since we're getting all scienfic here...
What's our target number?
Resistivity being thickness dependant, what's the minimum acceptable resistivity in the coating?

To make things even more technical.
What is lethal current?
I have a vague memory of hearing 15mA?
Let's assume 1mA?
And while we're at it, let's assume 3200V voltage.
R=U/I -> To avoid electricution the resistance should be 3.2e6.
Assuming the body resistance is negligeble compared to the dielectric.
Working with PVC @ 1.0e13 ohm/cm.
3.2e-9m would give us enough resistance for it not to be lethal.
That is 0.0032mm.
Please tell me if I'm missing something?
 
Last edited:
Please explain the mechanics at work here?
Resistivity being the inverse of conductivity should be as high as possible in an insulator?
Why would a low value be positive?

The dielectric constant should be low though to avoid charge build-up as much as possible? (If I'm remembering my physics right?)
Since it's also frequency dependant I'm curious as to how this will affect things in the audio band?

To maximize efficiency, you want an insulation with lower resistivity and higher dielectric constant.
Lower insulation resistance maximizes the DC bias voltage in the air gap.
Higher dielectric constant maximizes the AC signal voltage in the air gap.

See this post for more details, including a pic of the equivalent AC & DC circuit.
http://www.diyaudio.com/forums/plan...truct-cube-louver-acoustat-7.html#post2154621

Teflon would be an example of a poor choice for stator insulation.
It is a very good insulator with bulk resistivity > 1e18 ohm-cm.
It has a low dielectric constant of about 2.1.

So, compared to PVC, the use of Teflon would result in more AC signal voltage dropped going thru the lower capacitance of the insulation. And, more bias voltage dropped going thru the higher insulation resistance.
 
Pardon my previous rant.

bolserst> Ok, I get it. That's what you want to maximize efficiency. The insulaion isn't there to increase anthing, it's there to protect you from getting shocked.
How will this combine?

I've got a feeling I'm missing something important here?
Should we be looking at dielectric breakdown instead?
Regular conduction wouldn't result in anthing hazardous?
 
Last edited:
I think I'm starting to understand the concept.
Maybe people reading this thread are curious as well and maybe you people who have a better understanding can correct any misstakes on my part?

Dilectric breakdown occurs when the electric field is greater than the dielectric strenght.
I'm having difficulties finding good figures but I found a referece for Nylon 11 coatings.
Supposedly natural coating has a dielectric strength of 990V/mil and black 288V/mil.
A 3mil (75um) natural coating would almost take care of the dielectric strength and a black top coat would make it fool proof.
3x990V + 3x288V = 3834V before dielectric breakdown. Two 75um coatings -> 150um total thickness.
If everything adds upp a Rv3-4 stator with 150um coating will have >40% open area, be safe to touch and be fairly efficient. :)
 
The dielectric constant should be low though to avoid charge build-up as much as possible? (If I'm remembering my physics right?)

Hello MarkusG,

I just realized that lower bulk resistivity would also help avoid charge buildup on the outside of the insulation by providing a drain path to the stator. The lower the resistance, the shorter the time constant for draining off the charge.

Also, in case you haven't read it already, Few had started a thread on the purpose of insulation on the stators late last year.
http://www.diyaudio.com/forums/planars-exotics/154203-role-esl-stator-insulation-2.html#post1967738
 
Hi,

the problem is that those breakdownvoltages are tested and measured on plain straight massive metal sheet and not on punched metal with shap edges. It´s not important how thick the material build up will be on the front and back surface, because it always will be thicker than the buildup at edges. Too, edges are the areas of highest field strength.
In other words. A insulation thickness of 400-500µm on the flat surface may generate a much thinner insulation thickness over at the edges which is at the same the most stressed arey electrical field wise.
Stressing a material close to its limits increases the aging rate considerably.
Failing of insulation is only a matter of time then.

jauu
Calvin
 
Yeah, I'm with you Calvin. It's always good engineering practise to overdimension a bit where possible to account for problem areas and material defects.

Are my calculations correct on the dielectric breakdown?
If so my example showed it was safe for 3800V. To get that voltage in the stators you'd have to play som pretty loud music, right? It might not be an issue with people touching the panels when playing that loud?

Avoiding arcing on the inside however would be an issue. Are the same rules applied when doing the numers for the d/s? Or is the need greater/smaller on the inside?
One would have to take into account the dielectric effects of air as well since the diaphragm isn't supposed to touch the stator.

I do take your advice serious but being a techie I like to understand what I'm doing. :)
 
FWIW, I worked with a derivative of Kynar which had astounding adhesion to metal as a powder coating. This material was also made by Arkema. Kynar has much lower moisture regain than nylon, as well as superior insulating properties. It is, sadly, quite expensive.

How does PPG's Megaflon® compare? The reason I ask is that there is a local powder coating outfit that uses it rather than Kynar 500 P® for architectural work.

John
 

svenny

Member
2009-10-07 10:43 pm
Adding to Calvin's comments:
The holes in perforated sheet metal are made by punching. The face that is punched will have holes with smoother edges than the other side which is very sharp (which you can feel with your fingers).
The radius on the sharp edge will be very small, and with only 1 to 3 coatings and will cause arcing ( at high enough voltages) especially if they face each other (on the inside) since the electric field will be very strong on the very small radius of the edge.
You can face the stators outward, but to protect the curious and inquisitive you may wish to enclose the panels in some kind of cabinet e.g., like the Quad 57's.
Another possibility would be to have the sharp edges sand blasted.