Teflon Dielectric Strength / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Teflon Dielectric Strength

1 2 next

Move Thread

LAN_403

ScotchTapeLord

Mon Sept 13 2010, 12:38AM

Registered Member #1875 Joined: Sun Dec 21 2008, 06:36PM
Location:
Posts: 635

So I was thinking about making a homemade capacitor with teflon sheets, and I found this site that seems good with plastic. However, checking their listing of properties, I found that their value for the dielectric strength of teflon is a bit lower than what other sources on the internet say. About 5x lower, actually...

Is this the right value or am I amiss on something? 285v/mil means I would need almost a quarter inch to hold off 60kV, which wouldn't shock me (well, the 60kV could), but the other figures I've read (1500v/mil) would suggest I would need a lot less thickness for such a voltage. Thanks!

site page:

Adam Munich

Mon Sept 13 2010, 12:54AM

Registered Member #2893 Joined: Tue Jun 01 2010, 09:25PM
Location: Cali-forn. i. a.
Posts: 2242

AFAIK PFTE has a dielectric strength of 60MV/Meter, or 6kV/mm.

There are 39,370 mils in a meter, so...

60,000,000 / 39,370 = 1.524kV/mil.

I'm not sure why it's lower. You should call and ask, or test yourself. Maybe they're just covering their butt.

ScotchTapeLord

Mon Sept 13 2010, 01:49AM

Registered Member #1875 Joined: Sun Dec 21 2008, 06:36PM
Location:
Posts: 635

I sent an email inquiry. The funny thing is that the density and tensile strength values they supplied match the values given on the same sites that list the dielectric strength being much higher. The material is white so I don't think there are any pigments involved. Otherwise, it seems like a great site for big DIY capacitors.

Patrick

Mon Sept 13 2010, 02:40AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Check your math grenedier. You mean 1524 V/mil, so get rid of that "k" in kV.

also, the ASTM 877 dielectric test method uses the dielectric "punch through" volts as the final point on thier "hold off" quantity. Sooo, the 1,524 V/mil number was just at or before breakdown. for a "real" usuable number in caps or wire insulation thickness you would take the 1,524 V/mil number and derate it, -10% if your feeling lucky, -30,-50% if you would like to be considered sane. But its a judgement call, based on you and your application, for a harsh application, youll need more derating. This "judgement" is why your seeing such a low number, and why others seem to vary, and why you cant find consistent standard numbers.

My advice would be too not get greedy, after you blow your stuff up, because you thought you could get a little more kV, you'll feel awfully stupid for awhile. I know I did.

when in doubt, always test it for yourself.

Adam Munich

Mon Sept 13 2010, 03:33AM

Registered Member #2893 Joined: Tue Jun 01 2010, 09:25PM
Location: Cali-forn. i. a.
Posts: 2242

Oops, that was supposed to be a decimal.

ScotchTapeLord

Mon Sept 13 2010, 03:53AM

Registered Member #1875 Joined: Sun Dec 21 2008, 06:36PM
Location:
Posts: 635

Well, I was planning on charging them to 30kV with a DC supply and using in a giant tesla coil, so I thought the 60kV figure alone would be a large enough margin of error. I want about 100nF, so I know it's going to be heavy. The idea is to fill a big plastic box with oil and then push down metal plate+plastic plate alternatively into a nice stack, wired accordingly. I'm going to need wheels on this thing, but I don't want it to be too huge!

But because I want reliability, I'm going to go with the quarter-inch thickness, and just keep stacking until I get the right capacitance...

EDIT I just ran the figures and the amount of layers I would need would be prohibitively large and expensive...

Adam Munich

Mon Sept 13 2010, 04:12AM

Registered Member #2893 Joined: Tue Jun 01 2010, 09:25PM
Location: Cali-forn. i. a.
Posts: 2242

You probably want to clamp the caps if you're using them in a tesla coil. Rolled capacitors will wander in value way too much to be of any real use.

This is the stuff you want. 5kV a mil!

Of course you could always just use transparency sheet.

klugesmith

Mon Sept 13 2010, 04:45AM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

ScotchTapeLord wrote ...
...The idea is to fill a big plastic box with oil and then push down metal plate+plastic plate alternatively into a nice stack, wired accordingly. I'm going to need wheels on this thing, but I don't want it to be too huge!

But because I want reliability, I'm going to go with the quarter-inch thickness, and just keep stacking until I get the right capacitance...

EDIT I just ran the figures and the amount of layers I would need would be prohibitively large and expensive...

1. To get maximum V with a thick dielectric sheet such as you propose, you had better use comparably thick metal plates with well-rounded edges. With thin foil, the peak electric field will be at the dielectric surface next to sharp-edged conductor, with magnitude depending more on the radius of curvature than on the dielectric thickness. Look where the disposable plates (in another thread) failed!

2. For a quick estimate of physical size requirement, figure the energy storage (in joules) of your intended capacitor, divided by the energy storage density of your chosen dielectric material (in J/m^3, it is proportional to dielectric constant and to the square of the operating electric field strength).
For a sanity check, figure the energy storage density of commercial HV capacitors. Don't expect to beat that with a homebrew solution, but it should be easy to get within a factor of 10 and if you are good, within a factor of 2.

[edit] I think PTFE is not high on the list of plastics with high dielectric energy storage density. Isn't it a relatively low-E material?

ScotchTapeLord

Mon Sept 13 2010, 04:48AM

Registered Member #1875 Joined: Sun Dec 21 2008, 06:36PM
Location:
Posts: 635

It says it drops to about 2kv/mil at 10 mils, though. Generally, if something calls for a quote, I can't afford it. I'm considering getting a thinner sheet of polypropylene and lowering my voltage expectations...

Klugesmith:

That's good advice. I've scaled my expectations down to 20kV at about 90 nF... which is only 18 joules- I didn't realize how overboard I had gone with the voltage. I am planning on using aluminum sheets with smooth sanded edges.
If a 942 is 2kv at .15uF, then that's .3 Joules, which would mean mine should be 60x bigger... Mine is going to be about 1383 inches^3 which is about 11"x11"x11" or just under a cubic foot... not including the enclosure. That's not too bad..... I just hope the 60 mil of whatever I use can stand the voltage stress.

Patrick

Mon Sept 13 2010, 04:57AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Klugesmith wrote ...

1. To get maximum V with a thick dielectric sheet such as you propose, you had better use comparably thick metal plates with well-rounded edges. With thin foil, the peak electric field will be at the dielectric surface next to sharp-edged conductor, with magnitude depending more on the radius of curvature than on the dielectric thickness. Look where the disposable plates (in another thread) failed!

Yeah this is a great point. Sharp points are always evil, in a cap.

maybe if you go to the trouble / expense of making a good cap you should use an e-field simulator. i cant figure out how to use them though.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob