Does Dielectric Strength Depend On Thickness?

Dielectric strength would depend on the material wouldn't it ? The thickness x this dielectric strength would give you the breakdown voltage, would it not ?

So a 10mm column of air would need 30 kV to break down.

Why the doubt?

Maybe i haven't understood your question correctly.

Yes Sir , Dielectric strength depends on the material. If it is multiplied by thickness , It will give the BDV.

So It should be the property of material and independent of thickness of it.

This excerpt is taken from Wikipedia

Factors affecting dielectric strength

• it increases with the increase in thickness of the specimen. (Directly proportional)
• it decreases with the increase in operating temperature. (Inversely proportional)
• it decreases with the increase in frequency. (Inversely proportional)
• it decreases with the increase in humidity. (Inversely proportional).

Do you agree with the first point?

Classy Lass

Ah, i had not seen that one...

Well, it doesn't sound logical to me, but one is hesitant to question Wikipedia. Let us hope some expert will help us both.

Maybe it has something to do with the first sentence of the previous paragraph...

^{The theoretical dielectric strength of a material is an intrinsic property of the bulk material and is dependent on the configuration of the material or the electrodes with which the field is applied.}

(_{Btw, i think it is against the CR4 code of conduct to have more than one profile ... please check and if yes, delete one} )

Sir that account is of my sister. When I checked for this thread , she was logged in and I replied from her account only.

With Warm regards

Thank you 67model. Your posts always add something to my knowledge.

But Do you agree to fact that Dielectric strength depends on thickness ideally ?

I think you are right. There is a break in the layering of material that separates flaws in the layers. Whereas the solid thickness has homogeneous flaws as well. Layering dielectrics results in better performance / unit thickness.

There is only one area where the general rules about dielectric strength don't matter, and that is a resonant coil. Tesla discovered you could have a theoretically extreme potential on such a coil without breaking down the dielectric. But if the coil collapses it's resonant state.....watch out! Instant discharge and a vaporized winding!

The book "Reference Data for Radio Engineers" page 41-1 has curves which give the following Spark-gap breakdown voltages.

Sphere 0.04" Gap length in inches, 4.1 KV peak
Sphere 0.40" Gap length in inches, 35 KV peak
Needle Gap 0.04" Gap length in inches, 1.6 KV peak
Needle Gap 0.40" Gap length in inches, 11 KV peak

The preceding indicates that the volts/mil (or volts/mm) for breakdown depends on the gap spacing and also on the gap type (sphere or needle gap).

To answer your original question:

Since the unit is kV/mm, this value will theoretically be a constant for a particular material under specific conditions defined when the value was checked and tabulated for different materials (i.e. temperature, humidity , pressure etc).

therefore, your question is answered: The value of dV/dr does not depend on thickness! This is the Constant. But the dielectric strength of a particular isolator used in a specific place, with specific dimensions will depend on the thickness as well as the prevailing conditions if they drastically differe from the ones used as standards for evaluating the dielectric constant (dV/dr).

I hope this addresses the question.

The breakdown voltage behavior and its deviation from linear was well described at small distances.

A similar effect is at spiel at multi-meter distances. Let's say for arguments sake, dry air breaks down at 20kV/cm = 2MV/m. All high voltage insulators both for power transmission and in radio station's antennas are 2.5 - 5 timers larger than that to function right in real life. When you look at any of those, the corrugated surface is another 3-4 times longer than a simple smooth surface would be, to satisfy creeping current (caused by dirt) requirement, and corona discharges in wet weather.

Then there is lightning. 20kV = 2MV/m = 2GigaV/km. Now, lightning is normally many km long. Nobody in his right mind thinks, we have multiGigaVolt discharges in lightning. Simply, because it may be started by cosmic rays as initiator, then the charges start moving, and the air breaks down in steps.

So, I would summarize, that air would be linear from a few millimeters to a few meters, with the condition that no sharp corner or tip causes the locally permissible field strength to be exceeded. So, yes indeed there is more, than meets the eye.

Dielectric strength is already in units of [Volts/meter] and should be a constant for most dielectric materials. As you surmised, the breakdown voltage [Volts] of a particular case needs a distance [meters].

Dielectric strength [Volts/meter] x distance [meters] = breakdown voltage [volts]

If you use the proper units there should be no confusion.

As mentioned above, a measured dielectric strength "may" drop with thicker dielectrics. This is due to flaws (contamination, density variation, inclusions, voids, etc.) in the material. Using many thin layers of high quality dielectric is usually much better than only one thick layer and the dielectric strength should be very close to predicted theory.

I'm fairly certain this is an ERROR on the wiki Dielectric_strength page.

Factors affecting dielectric strength
* it increases with the increase in thickness of the specimen. (Directly proportional)

I disagree. The breakdown voltage [volts] DOES increase with dielectric thickness. However, the "dielectric strength" [Volts/meter] of a material is independent of thickness by definition. Unless someone can provide a reliable reference to support the wiki statement, it should be considered an error.

GA

This answers the question.

Yes it's that simple, dielectric strength is a property of a material, for fixed conditions of temperature, pressure and maybe other physical factors, and it's independent of thickness.

The breakdown voltage of that material in use depends on thickness and a few other factors such as shape.

For solid insulation, break down takes place due the heat generated in side the insulation. Each insulating material has finite resistance associated with it. So the I²R loss generated due to V²/R would result in temperature rise of the material. Now, if the resultant final temperature is within its withstand capability, no break down takes place. This temperature rise may be higher in case of any local flaw situated deep inside a thicker insulation (where heat cannot be dissipated to out side) and may cause local charring, there by making some local conducting path and increasing the current passing, which in turn results in further charring at that location and resulting in more charring of material. This process continues till break down takes place. Same flaw when situated in a thinner material or near any of the surfaces, may not result in same process due to better heat transfer (less thickness of insulation to cross) and comparatively lower final temperature.

Since no insulation is perfect, the chances of flaw remaining deep in side a thicker insulation are more (due to imperfect curing, imperfect pressing etc.). So in practical situation break down voltage strength (volts/mm) will not be in increasing in direct proportion to the thickness. In thinner insulation it is more and in thicker insulation it is less. Theoretically, yes V/mm does not depend on thickness but practically it does.

Regarding the Thermal Theory, you have not mentioned AC operation, which must account for the majority of insulation service. "Electrical Measurements and Measuring Instruments" by Golding and Widdus ISBN 0273 41058 in 1973 reprint makes the points :-

• Dielectric losses are proportional to frequency.
• and to field strength squared.

Additionally, for solid materials, we are talking about a destructive test. Practically, each specimen will fail at a different voltage. So there is a statistical curve involved, with scatter, a standard deviation and a mean value. The standard deviation is seldom given in the data table, nor is the test duration (as time passes the insulation will get hotter, increasing probability of failure).

So tables of "dielectric strength" are useful as an indication of whether one material might be better than another, but no more a guide to success, in service, than the breaking strength of materials in mechanical tension.

Right on 67. In the defense industry it is common to see meters with stickers on their faces that say "Indication Only". This is because the values calibrated for are secret.

But keeping in mind that all the variables involved in dielectric strength, and breakdown voltage are, well, variable, it's a good idea to keep the words "indication Only" in mind at all times.

Dielectrics are not perfect barriers one moment and broken the next. They are inhomogeneous materials between two differing charges. Much that happens is dependent upon what is happening at the atomic scale between those two charges and what the changes are in the charges themselves.

A slight change in temperature, frequency, voltage, pressure (physical), current, can tip the balance and "break" the barrier.

There is always current across a dielectric if there is potential. It's just not always measurable.

Back to the OP. The thickness separates the charges and lowers the potential for current no matter the material. Dielectric strength determines how thin a material can be and still maintain a nominal current. A thin piece of Mica is as good as a much thicker piece of wax paper, until they are heated....A thick piece of high density fluoroplastic is almost as good as a thick piece of glass, until they are heated....

The dielectric strength value that you can find in tables for various materials is considered a constant, but this is only if the other factors, temp, frequency, etc are also constant. I think that what you may want to know is, does having a thicker insulation allow for higher voltages? YES! When the insulation material at your disposal is the one you have...making it thicker will let you move up to higher voltages.

Tesla used gutta-percha and boiled-out mineral oil, paper, glass, mica, ceramic, in short, what he had available. He made them as thick as he needed to reach the voltages he was looking to reach.

This person asked does maximum dv/dr depend on thickness. Does any dv/dr depend on thickness?

Let wikipedia know this...

i think Wiki is correct. Please see the Dupont paper given by BertHickman

Extracts here for quick reference....

AC voltage causes dielectric heating which melts the dielectric and causes breakdown at lower field strengths. NO disagreement here and I believe wiki is correct on the quoted statements I've labeled below as (2),(3),(4).

"Factors affecting dielectric strength

(1)* it increases with the increase in thickness of the specimen. (Directly proportional)
(2)* it decreases with the increase in operating temperature. (Inversely proportional)
(3)* it decreases with the increase in frequency. (Inversely proportional)
(4)* it decreases with the increase in humidity. (Inversely proportional)"

However, statement (1) IS disputed and I'm claiming (in post #15) that (1) is false.

Hey buddy , good chunk of information!

Thanks!

I was taught about the linear relationship between BDV and thickness.

So I was doubtful about the variation of dielectric strength with thickness.

More Dielectric Strength Information from the MATWEB.COM Material Information Website.

Please excuse presentation - CR4 does not like tables! Also the mixed inch/metric units - that is the way they give it.

Dupont:

T = Thickness mil [1 mil = 0.001 inch]

DS = Dielectric Strength kV/mm

Kapton Polyimide Film

T, DS

0.5, 339 [8610 kV/in]

1, 276

2, 217

3, 197

5, 154

NOMEX 410 "Paper" (a plastic polyamide film, like "paper"!)

T, DS

5, 27

7, 33

10, 32

15, 33

30, 27

Test Method (Kapton):-Flat sheets in air placed between 1/4 in diameter brass electrodes with 0.8 mm (1/32 in) edge radius subjected to 60 cycles AC voltage at 600V/s rate of rise to the breakdown voltage.; ASTM D149-81.

Notice that with the thick NOMEX films, the dielectric strength has reached a kind of "uniformity of mediocrity" compared to the much thinner films, which have greater dielectric strength as they get thinner.

And I suggest ksvridar made a mistake when he wrote [under a chart showing kV/mm falling with increasing thickness!] the data confirmed the Wikipedia "rule" that dielectric strength increases proportionally with insulation thickness.

And thanks and praise to Dupont for giving data (unlike many sources) which clearly gives the units and test method and making it clear that results depend on the test electrode diameter.

Super 67model

Classy lass, you may be an electrical/electronics student, but all tangible products require some mechanical knowledge. And certainly you will wow your professors and classmates with data from MATWEB.

Register,search and download to excel, pdf,whatever.

http://www.matweb.com/search/DataSheet.aspx?MatGUID=1a046308aeb94ba39259a481dc7f3dc1&ckck=1

Thanks Kvsridhar,

The quote from Wikipedia in post #15 was.....

Factors affecting dielectric strength

* it increases with the increase in thickness of the specimen. (Directly proportional)

.... the evidence seems to be from tests that dielectric strength is not directly proportional to increase in thickness - rather it it is a curve which falls further below a straight line as thickness increases. If one takes "dielectric Strength" as MV/m (actually numerically the same as kV/mm), as given in the following section 1 of the Wiki article referenced in post #15, the test evidence indicates an initial inverse, but not proportional, relation.

WOW! Beautiful data there pal! How interesting. Totally counter intuitive. This does not happen when the dielectric is a vacuum or a ceramic....

Very interesting. Back to the thinker for me!

Who brought in dv/dt, brings in all kinds of complications.

This person wanted dv/dr voltage gradient with distance, if you start playing with other variables, particularly changes in volts as above, then of course it complicates matters.