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)

first previous 2 3 4 5 next

Move Thread

LAN_403

Ash Small

Wed May 28 2014, 08:07PM

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

Thanks for the explanation, Sulaiman. I thought it might be something alomg those lines, but it's good to have all the detail.

The only 'easy' way I can think of to drive it at resonance is by charging it through a (say) 330kOhm using a DC flyback, and using a variable spark gap (basically an SGTC primary circuit). Using a variable spark gap will allow me to adjust the voltage that it charges to.

However, I assume that using a spark gap will render the 'radio method' of testing completely useless, as the spark gap will cause the same effect as corona.

I don't think 'testing in the dark' will work either, as I bind my capacitors tightly with LOTS of insulating tape.

If I just drive it at resonance with a lower power supply than you suggest, won't the voltage 'ring up' by itself? (drive it as you'd drive the 'secondary' of a TC?) I could still use a variable spark gap in parallel with the capacitor to limit the voltage ring up

Sulaiman

Wed May 28 2014, 09:19PM

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

If your capacitor and inductor are 'ideal', i.e. no resistance in series or parallel,
then the ring-up could aproach infinity, or breakdown voltage or fusing current.
I'd expect a foil capacitor to have a 'Q' of 500 to >1000
but practical inductors have much lower Q, which limits ring-up.

For low power testing you can use 'loose' coupling
e.g. 45 nF in parallel with 14 uH driven via a smaller capacitor, say 1 nF in series with the 'signal generator'
this reduces the drive current by a factor of 45
but due to the Q of the inductor the voltage ring-up will be quite limited.

Probably easiest to use a simple circuit simulator to understand what happens,
the math is quite simple but it doesn't easily give a 'feel' of what goes on.
Simulate the Q of the inductor with a parallel resistance of Q x impedance
or a series resistance of impedance / Q.

From vague memory, watching the voltage across a cap aproach breakdown with a 'scope
the trace goes from a clean sinewave to 'is that something?' to OH CRAP ! very quickly :(

If you had a 156 H inductor you could use 60 Hz.

Ash Small

Fri May 30 2014, 03:41PM

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

Ok, I think I'm starting to get the hang of this.

After considering copper tube and litz wire, I started considering cheaper alternatives, looked at pancake coil theory a bit, and concluded that I can wind a 14uH 'pancake' coil with a two inch inner diameter using around 2.3 metres of aluminium foil. This solves the 'skin effect' issues and keeps costs down.

I just need to work out how much foil I need to reduce the resistance sufficiently so that the 'Q factor' is high enough.

I can always use several layers of foil separated by polythene sheet if required. Time to measure the resistance of aluminium foil, I suppose

EDIT: Not an easy thing to do with a DMM. I've cut a 25mm x 300mm strip of kitchen foil, but the meter doesn't tell me anything. Any ideas?....I guess I could just try it and see what happens.

Maybe I should try passing some current through it?

radiotech

Mon Jun 02 2014, 07:43PM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

The electro grease is used to lubricate high voltage sliding contacts, in AC gear where
air and corona are problems. We used to slather it on the 'stabs' of rail mounted breakers.

An interesting substance to consider is Dow Corning DMPSO (dimethylpolysiloxane) if
you can find some 50 centistoke stock.

Have you considered a something which changes dielectric constant with applied
field voltage?

Ash Small

Thu Jun 05 2014, 09:47PM

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

Thanks for the input, RT. I did a quick google search for the Dow Corning product, but didn't find anything yet. Found a bit on polysiloxanes, I'll read more later.

Any suggestions as to products that change dielectric constant with applied voltage?

I've been looking for cheaper sources of silicon grease, but not found any yet that are cheaper than the Electrolube HV grease. It's used for plenty of other purposes though, particularly where rubber is involved.

I'm trying to keep this to easily available products that are relatively cheap.

I've been thinking some more, changed the design slightly, and I'm currently not sure whether to buy another sheet of polythene (another Â£10) and go for ~15nF @~25kV or just use what I already have and go for ~30nF @~12.5kV.

Any advice here would be appreciated.

I've done a sketch of the dimensions of the 15nF @25kV one, and I've got some more paint on the template, but it still needs some more, and I need to fix the exhaust and suspension on my car, so progress is pretty slow at the moment, although I am making progress

The yellow bits are where I want the HV grease, I think.

radiotech

Sun Jun 08 2014, 08:24PM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

The Dow Corning stuff was available and current, sadly 55 years ago. I still may have files on it.
Looking at your capacitor build, a thought comes to mind about managing the side fringe where the plate
edges meet the air. These days, high voltage cable managements involves buying ready made
termination kits that deal with voltage stress.

This was not the way is was done in the past. Proper stress-builds were done by those who
used good techniques. (woe betide he whose job flashed)) Builds of cambric and irrathene and other
stuff was used.

You might find something in this:

Sulaiman

Sun Jun 08 2014, 09:07PM

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

sorry to be so negative but I think 16 insulating and 2 conducting layers
isn't going to be easy to roll, to put it mildly !

Good Luck ....

Ash Small

Sun Jun 08 2014, 09:49PM

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

Thanks for the pdf, RT.

I've given a lot of thought to the 'terminal areas', however, I've not finalized anything here yet. The reason I don't want to immerse everything in oil is because I don't want to increase resistance where the copper wires are wound around the terminals. For the same reasons I don't want to use the dielectric grease here either.

In the past I've just wrapped everything in insulating tape, but I'm obviously going for much higher voltages with this one.

Maybe one of the two part silicone sealants could be used here? I'm still considering the alternatives.

I've just noticed your post, Sulaiman. I agree that rolling 16 layers of dielectric won't be easy, but my reasoning is as follows: I could roll lower voltage capacitors and then connect them in series, but this would increase the amount of aluminium foil required. This is not a problem in itself, as the foil is the cheapest component used here, but it would also significantly increase the amount of dielectric grease required, and unless I can find a much cheaper supply, the dielectric grease is one of the most expensive 'components' used. Either way, the total amount of dielectric required remains pretty constant, I think.

Ash Small

Mon Aug 11 2014, 02:31PM

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

Well, It's been a couple of months, but I seem to have some free time again, so I should be able to make some more progress with this project.

I've been able to give it some thought, though, and I've decided to go with the 12.5kV, 30nF design, and test it to destruction.

The only thing I can't decide on is whether I'm better off using the thinnest foil I can find (what's sold as 'wrapping foil, unsuitable for cooking', or go for the thicker 'cooking foil'.

The thicker foil will have less resistance, although the thinner foil may be sufficient, but the thinner foil will presumably have less air at the edges, and four layers of polythene sheet may well deform sufficiently to virtually eliminate any gap.

The thinner foil will have a 'sharper' edge, so may be more conducive to corona formation, but I think the difference will probably be negligible.

So, should I use thicker foil with marginally less resistance, or thinner foil which may have less air gap at the edges?

Sulaiman

Mon Aug 11 2014, 05:24PM

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

As you will connect along the entire 'edge' of the foil
I do not think that the resistance of the Al foil is an issue.
(commercial caps use VERY thin foil, or a VERY VERY thin vapour deposited layer of aluminium)

'ragged' edges of the foil will be much more of a problem than it's thinness so no problem there

as the grease will squeze out towards the edges
I also guess that thinner foil will not really make much difference to the airgaps

so the main consideration in my opinion is
which will be easier to roll neatly/tightly .... I don't know.

first previous 2 3 4 5 next

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