Ways to stop corona..List please? / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Ways to stop corona..List please?

1 2 next

Move Thread

LAN_403

Killa-X

Sun Jan 25 2009, 05:42AM

Registered Member #1643 Joined: Mon Aug 18 2008, 06:10PM
Location:
Posts: 1039

I need a way to stop corona on parts. I was told you can use wax and stuff, but the problem for me is that my parents still limit the things I do, so it would be hard to do a lot of methods im aware of. Is there any simple easy way to prevent corona? I'm making both 50 1000V rectifiers (3 dolloars for 100), and 100 10MOHM resistors (1 dollar), and both need to be protected against corona i was told.

what are some easy ways to do so?

Arcstarter

Sun Jan 25 2009, 05:55AM

Registered Member #1225 Joined: Sat Jan 12 2008, 01:24AM
Location: Beaumont, Texas, USA
Posts: 2253

Corona dope, and for simple insulation use packing tape :). Or, melt a bit of hot glue sticks on a stove or something and dip it, but idk if this will prevent corona, but it will insulate. Corona eats through stuff easily though.

Dr. Dark Current

Sun Jan 25 2009, 11:58AM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

hmm, I don't get it, do you want to stop corona losses? If this is the case, simply encasing the parts or points in some kind of a plastic container helps.

Or do you want to protect low voltage parts from degradation by corona, in some high electric field/gradient environment? If this is the case, you can pot the parts in anything that insulates. Candle parafin wax is great for this.

DerSchwamm

Sun Jan 25 2009, 12:15PM

Registered Member #1936 Joined: Sun Jan 25 2009, 11:34AM
Location: USA
Posts: 17

I Think what he is saying is he's making a pair of 50kV rectifier modules out of 50 1N4007 diodes and a 100:1 factor HV probe for a voltmeter out of 100 10MOhm resistors and is looking for potting material suggestions. Wax is easiest, and can be removed in case of screw ups. More permanent projects can be potted with epoxy resin from an automotive or marine supply store (for use with fiberglass matting.). One should be reasonably sure your potting material is non conductive and also non-corrosive (No potting in silicone! Acetic acid == BAD!). From looking at your other posts I doubt capacitance or reactance would need to be taken in to consideration for these, but be aware in the future if you're trying to make resonant circuits that some potting materials will add capacitance to your projects.

For the diode assembly, assuming they're 50KV, you want at least 6cm from the input and output terminals. Assuming a normal DO-41 plastic case, and trimming the leads down to the smallest possible soldering length.

Case == 0.2" long.
Leads == 0.25" long each end

Total overall length of ONE diode is 0.45, we'll round it to one half.

So 50 diodes at half an inch a piece is 24 inches plus the "last" lead sticking out 24.25 technically.

So you could just assemble them into a PVC pipe 24" long with terminals at either end, or you could wrap them around a form. You could wrap them around the outside of a piece of 1/2" pvc pipe, As long as your piece of pipe is *longer* than roughly 10cm(longer than 6 because of individual turns-spacing vs peak voltage across each diode.). Then put that piece of pipe in, say 3/4" or 1" (Depending on the wall thickness of your pvc, it might not fit in a 3/4 pipe.) pipe and fill the void with wax or resin. If it is a tight fit, go a size larger so you can actually pour your potting material into the void.

Same dimensions go for your voltage divider probe, figuring on 1/4W resistors (1/8W are shorter of course) you'll end up with a string approximately 50" long.

Now a 100:1 probe can be a dangerous thing. Plugged into a multimeter you could *theoretically* read voltages up to 100KV. 100KV is a tricky beast. Things you thought were insulators before are no longer. It also arcs quite well. When I built mine (most of us eventually build a HV probe I bet.) I coiled my diodes around a perspex rod 1/2" in diameter and 36" long. and put that inside of a 1" PVC pipe with end caps drilled and banana jacks installed. I then potted the whole beast with 3 pounds of canning wax. I also made a second shorter one that was just a 10:1 divider. The added length of the pipe serves two purposes: First, it keeps *you* away from the HV. You're holding the GROUNDED end of the probe. Second, it ensures there is a low chance of flash over in the event that you actually try to measure truly high voltages. I would not recommend potting these in a folded up format. Straight out is best. Most commercial 50KV probes are 18" or more long.

If your parents whine about melting wax, tell them you _could_ use USP grade mineral oil, but its really difficult to make sure it stays inside the pipe. Meaning that wax will be less messy in the long run. Really, the boiling water is more dangerous than the melted wax, which shouldn't be much hotter than 140F worst case. Most canning wax melts easy around 120F.

Also, if you do pot with wax its probably not a good idea to leave the finished products in a closed car in July. The wax will probably find a way out.

Dr. Dark Current

Sun Jan 25 2009, 12:21PM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

DerSchwamm wrote ...

So 50 diodes at half an inch a piece is 24 inches plus the "last" lead sticking out 24.25 technically.

Alternatively you can arrange them like this:

DerSchwamm

Sun Jan 25 2009, 12:25PM

Registered Member #1936 Joined: Sun Jan 25 2009, 11:34AM
Location: USA
Posts: 17

Dr. Kilovolt wrote ...

DerSchwamm wrote ...

So 50 diodes at half an inch a piece is 24 inches plus the "last" lead sticking out 24.25 technically.

Alternatively you can arrange them like this:

1232886102 152 FT62274 Diodes

*AAAAUGH!* Why didn't I think of that. :( Its no trivial task to build a continuous chain of diodes end to end without metal fatigue. :)

I'm guessing you line them all up side by side, alternating, and put a strip of tape down them, then flip em over and brush on some super glue?

Myke

Sun Jan 25 2009, 06:37PM

Registered Member #540 Joined: Mon Feb 19 2007, 07:49PM
Location: MIT
Posts: 969

In a resistive divider, you can't measure high freq signals because of the self capacitance of the components. You would have to make a resistive capacitive divider who's freq response over the freqs that you want to measure is relatively flat (difficult to do). Also remember that if you stick your meter across one 10M resistor, it will change the resistance of that one resistor. Find out what resistance your meter has and figure out the resistance of the last resistor.

DerSchwamm

Mon Jan 26 2009, 09:53AM

Registered Member #1936 Joined: Sun Jan 25 2009, 11:34AM
Location: USA
Posts: 17

Your mention of hot glue intrigued me, and I trawled through some old usenet postings and came up with an answer that it is pretty much ok to use hot-melt glue, but only the hot (meaning high temp) stuff.

In short, the odds of you being able to identify what plastic your sticks are made out of is slim to none and the author of those posts even says candle wax wins over glue in many ways.

And really, you'd need a *LOT* of hot-melt glue to do it, and you'd also have to heat your whole rectifier assembly up to the melting point of the glue so it was able to flow in all the nooks and crannies and give air bubbles a chance to rise out. You didn't mention having a vacuum oven at home(your school/college might? can't hurt to ask if you really have to do it this way.), so I'm pretty sure your only viable material is molten paraffin, which, from experience, hurts less for a shorter period than hot-melt glue does when, in spite of the package warnings, it becomes applied topically.

YMMV.

3l3ctrici7y

Mon Jan 26 2009, 10:52AM

Registered Member #1806 Joined: Sun Nov 09 2008, 04:58AM
Location: USA
Posts: 136

DerSchwamm; I agree.. topical application of paraffin will hurt less than hot glue would :)

A while back, my brother had built a board that he wanted to seal and he covered it in paraffin wax.
You can get large blocks of paraffin wax at places like walgreens for those hot-wax-glove skin care things.

What he did was melt a large block of wax in a kettle on the stove, then dipped the board in the wax several times to build up the coating. I seem to recall that later he needed to edit the board, so he set the oven on the lowest setting, and hung the board over a catch basin. The wax melted and flowed right off the board.

An idea that just hit me would be to use plasti-dip (or some similar thing)

I don't know how well this would work, and I've not used it. The idea just occured to me; it may not actually be suitable.

Proud Mary

Mon Jan 26 2009, 12:57PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Good layout and an understanding of the conditions that give rise to corona discharges are the keys to success.

Corona discharges occur first at points and edges where the charge density is highest, so a good design will try to avoid creating these as much as possible.

High voltage dividers of very great accuracy - such as the famous Park Resistor at the National Bureau of Standards - avoid insulation altogether, since this effectively creates a leakage path in parallel with the divider. The Park Resistor consists of 100 x 1M resistors in series, arranged in a very broad open air helix to avoid the creation of points and edges that would lead to corona losses.

(I am not suggesting anyone copy the exacting methods of the Park Resistor

, but for those who don't yet know of this famous high voltage device, it may be good to know that such a simple passive component is the gold standard of high voltage measurement throughout the United States)

1 2 next

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