HV capacitor question (Rolling my own) / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

HV capacitor question (Rolling my own)

previous 1 2 3 4 next last

Move Thread

LAN_403

Ash Small

Thu May 15 2014, 01:00PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

OK, I'm still preparing the template for marking out the polythene sheet (I want the template to be as smooth as possible, so as to prevent any risk of damage to the polythene sheet).

I'm concerned about corona forming along the edges of the aluminium foil. Even though I have a four inch overlap of polythene, I know it would make sense to put the capacitor under oil and vacuum impregnate it, but I'm reluctant to do this as I suspect it will adversely affect the resistance of the finished capacitor, so I'm considering smearing something along the edges to suppress corona.

Whatever I use I don't want to use a liquid, as if it migrates to the terminals it could adversely affect resistance, so I'm considering either something like a beeswax compound, which should be viscous enough to stay in place, or something like polyurethane varnish/enamel, which should dry and seal the area around the edges of the foil, although I'm not sure if it will actually dry if it's not exposed to the atmosphere.

I also have a coconut oil/ beeswax product handy, but only a small amount (boiled coconut oil is apparently an excellent insulator).

Any other suggestions? I know there were a few 'old fashioned' mixtures used, but I've not found the pdf I have somewhere that describes them. Any other ideas?

(The basic idea is to brush it onto the inner edges of the foil as I roll it, so that air will be expelled from these areas as I roll it up)

Newton Brawn

Thu May 15 2014, 03:52PM

Registered Member #3343 Joined: Thu Oct 21 2010, 04:06PM
Location: Toronto
Posts: 311

Hi Ash !

Im thinking leave the capacitor dry.

I this case the dielectric between the aluminum electrodes are compound of an air dielectric layer and a solid dielectric layer.

The air dielictric layer coud be any eventual air inside the insulation or between the insulation and electrode. This layer to be very thin and not less that the aluminum foil thichnes,

The solid dielectric coud be a film of polyester Mylar, polyetilene, etc with low dielectric constant.

During the coiling of capacitor, the turns to be tight enought to reduce as much as possible some ir inside,

The solid dielectric thichness to be recalculated to new situation to avoid any voltage gradient that can produce corona in such air buble.

Newton Brawn

Thu May 15 2014, 04:12PM

Registered Member #3343 Joined: Thu Oct 21 2010, 04:06PM
Location: Toronto
Posts: 311

Here the design of air-solid dielectric insulation in order to avoid corona:

]air_-_solid_dielectric_corona.pdf[/file]

As an exemple, lets consider a aluminun foil Reinolds for kitchen that has 18 um thichness,
and a eventual air bubles of 0.05mm thick that may be inside the dielectric,
and a solid dielectric material with a dielectric constant of 2, (polyetilene) with a thichness of 3mm,
the calculated corona starting voltage could be 18 kV peak.
So such capacitor may be used to 9kV peak voltage.

Recalculation for your capacitor maximum voltage has to be worked

Ash Small

Thu May 15 2014, 04:43PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Are you saying that the 100mm of air between electrodes will be sufficient to eliminate corona forming at the edges of the foil? I'm not sure that I understand your point, Newton.

At the moment I'm tempted to try polyurethane varnish, and just keep it warm for a day or two to dry it as best as possible.

EDIT: I've just noticed you've edited your post above.

Sulaiman

Thu May 15 2014, 06:16PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3141

I tried polyurethane varnish,
if you mean coat the foil then when the varnish has set roll, then it may be ok,
if you roll then impregnate with pu varnish
the outer edges set/dry but the inside remains liquid for a long time, days or months !

Ash Small

Thu May 15 2014, 06:40PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

ok, thanks for that.

Maybe I should consider epoxy?

Newton Brawn

Fri May 16 2014, 04:35AM

Registered Member #3343 Joined: Thu Oct 21 2010, 04:06PM
Location: Toronto
Posts: 311

Hi Ash !

Let me know the maximum working peak voltage to be
applied in your cap.

No, Im not sayng about the 100mm or the fringe leakage.

For a good design I think it may starting by the calculation of the dielectric thickness .

If you consider a small air pocket inside the solid dielectric sheets or between the solid dielectric and the aluminum foil and a air pocket at the margen of the aluminun foil, you will have corona in the air pockets, UNLESS the dielectric thickness is calculated to avoid excessive voltage gradient in such air pockets.

The graphic above allows you design a capacitor using a solid dielectric ( as polyetilene) with some air pockets.
.

Sulaiman

Fri May 16 2014, 05:08AM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3141

Have you noticed that with few exceptions commercially available foil/film capacitors
are not available for 10's kV ?
I suspect that there is a good reason,
even though capacitor manufacturers have access to very pure, clean materials,
and all required equipment to roll in a vacuum.

IF I was going to try roll-my-own again (I doubt that I ever will)
I think that I would build several capacitors to make an MMC or at least a series string
this would allow much lower voltage per cap so reduce the corona and surface tracking problems
I'd make more caps than required
each capacitor could be tested individually and any 'weak' ones not used in the MMC.
If any one unit failed in use then it would not be a catastrophe, just a nuisance to replace.
I'd roll a couple and test them to destruction, then build my MMC to avoid said destruction.

Despite all of my negative comments I think that you should continue
... it's a sort of 'rite-of-passage' for coilers !

Good luck and please video your progress for the benefit of those who may follow.

P.S. the 'volts per mil' of plastic dielectrics decreases with increasing dielectric thickness.

Ash Small

Fri May 16 2014, 09:04AM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Newton, I'm mainly concerned with the 'fringe' corona here. I think it will form much easier than corona on the surface of the foil.

Sulaiman, I rolled all the capacitors for my HF TIG welder in 2001 and as far as I'm aware they are still working fine (I think it was either 20 or 50nF for the primary tank cap and 4x50 for the HF filter). That had a one to one air core transformer (ten turns primary and secondary, but only needed a 1/4 inch arc under argon, so not much voltage, but it did demonstrate the usefulness of 'home rolled' capacitors.

All the points you make are valid points, and I'm aware that 'stacked, multi-plate' capacitors under oil perform better that rolled, air insulated ones, but part of this exercise is to test this polythene sheet as a dielectric, and to see what I can do with it.

The capacitor will be 'protected' by an adjustable air gap (I bought some small brass knobs the other day, although I may invest in some larger ones later), and charged with a DC flyback through a fairly high value resistor (The same as I used in my HF TIG welder). I'm planning on building a voltage divider to measure the voltage on the capacitor, and if it lasts long enough, to 'play around' with some secondaries, maybe paralleling half a dozen flybacks if necessary for more power (maybe even wind a few bobbins and knock up an H-bridge later).

I agree that putting a few in series makes a lot of sense. This is just an experimental one made from two rolls of foil and one sheet of polythene, with a 'theoretical capacitance' of ~674nF, and I know you can use a lot less capacitance for a TC (maybe two orders of magnitude less). I could roll several smaller ones using the materials I have and build an MMC, but I first need to establish the limits of this design.

I'm currently in the process of painting the template for marking out the polythene sheet, as this should reduce the likelyhood of damage to the polythene as any surface imperfection would spell disaster.

I do think that trying to suppress the fringe corona is worth the effort, but I've not yet decided on the best way to go about this. As I pointed out earlier, I want to keep oil and other insulators away from the terminals, to keep resistance to a minimum, but at the same time 'seal' the edges of the foil inside the capacitor, to suppress corona.

While epoxy may do a reasonable job here, I may be better off using something a bit more flexible. and less likely to crack.

Sulaiman wrote ...

P.S. the 'volts per mil' of plastic dielectrics decreases with increasing dielectric thickness.

This is why I'm only using one sheet

Ash Small

Fri May 16 2014, 11:39AM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

I'm currently thinking of using a silicone based dielectric grease like this one from Electrolube:

It's good for 30kV/mm, so should help suppress corona. The only downside I've found is that under arcing conditions (such as corona?) silicon carbides can be formed, but I don't think that will really be an issue here.

I considered silicone sealant earlier, but it looks like I'd need an expensive two-part type, and I can forsee other possible problems, so silicone based grease is top of the list at the moment.

previous 1 2 3 4 next last

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