Secondary: segmented/multi flanged bobbin / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Secondary: segmented/multi flanged bobbin

previous 1 2 3 next

Move Thread

LAN_403

Dr. Dark Current

Wed Jul 30 2008, 01:21PM

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

baleworker wrote ...

I dont understand how can it arc if there is no air at all? You gotta have air to have corona!
sgments are insulated by 1.5mm CPVC, that should be more than enought for 1.5-2kV difference, I am more worried about arcing inside of the segment itself.

There is no air but there is lossy dielectric. I bet it will heat up and melt, and spark over. But I would be glad if you could prove me wrong

RickR

Wed Jul 30 2008, 03:55PM

Registered Member #93 Joined: Thu Feb 09 2006, 04:11PM
Location: Houston, Texas
Posts: 64

Here's My Winding Technique. The most difficult part is doing the cuts in the UHMW rod so that there is a sufficient number of grooves to hold the necessary number of windings (without accidentally sawing off a finger). But spacing the winding areas seemed to work very well for generating high voltage.

Since I wrote the initial thread, I've had a chance to do a little debugging of coils that eventually failed. They didn't appear to have arc'ed over, rather they seemed to give out from the current - like a fuse. The wire in the failed coils seemed to be melted to form an open circuit. I also didn't find the kind of carbon tracking that I would have expected if flashover had taken place (but I'm not an expert either). The failed part always seemed to be very close to the gap in the ferrite core. Finally, I seem to remember that the only time the coils failed was while I was drawing an arc (it's been a while, though).

RickR

If you have the patience and tools, you might consider my construction technique, and try to limit the current from the supply to the coil to various values to determine what current value is best for prolonging the life of the coil.

EDIT:

It looks like you're using a pretty beefy ferrite core - it might be able to manage quite a large number of volts per turn. This could help you reduce the number of turns used for the secondary (assuming you can supply a pretty large voltage at a fairly high current to the primary). I believe with a lower secondary turn count my technique could prove very useful. But again, the down side is the construction effort required.

baleworker

Wed Jul 30 2008, 08:06PM

Registered Member #1596 Joined: Fri Jul 18 2008, 08:43PM
Location: USA
Posts: 65

RickR,

You dont have a picture, but from what I read my construction seems to be sort of similar to yours. However I use the lathe to turn the bobbin out of plastic rod.

If you look close, you can notice that sections are eccentrically machined. It effectively eliminates the problem you encountered. The wire at the top of one section is right on the same level as bottom of another section, and it does not have to cross windings in order to go from one section to another.

The notches are made on opposite sides in 1-2-1-2-1-2-1 order and they are cut at an angle to create a nice smooth path for the wire.

My coil took just a few hours to build. I will pot it with extra small viscosity potting compound under a deep vacuum once potting compound arrives.

The downside of my construction is a reduced available space in radial direction, because of eccentricity.

Other than that it is a very advanced design. It took me some time and cad modeling to come up with eccentrically placed sections and placement of notches in a way that copper does not have to go in radial direction.

RickR

Wed Jul 30 2008, 08:59PM

Registered Member #93 Joined: Thu Feb 09 2006, 04:11PM
Location: Houston, Texas
Posts: 64

Yeah, sorry about the picture. I'll see if I can take one and edit it in later. Your ferrite core looks very similar to the one Steve Ward used when he was putting together a C-W multiplier pic here. He said his core was good for about 35V/turn, so you might be able to get some pretty high voltage out. What gauge wire are you using on the secondary and about how many turns are you winding?

Rick

baleworker

Wed Jul 30 2008, 09:38PM

Registered Member #1596 Joined: Fri Jul 18 2008, 08:43PM
Location: USA
Posts: 65

RickR wrote ...

What gauge wire are you using on the secondary and about how many turns are you winding?

Its 28 gauge, 750 turns, its possible to increase excentricity and get at least 1500 out of the same size bobbin. (the windings will be closer to the outer edge). I originally intended to do so but made a machining mistake, so I will just test it from point A to point B like that and make changes/adjustments upon seeing test results

baleworker

Thu Aug 07 2008, 03:03PM

Registered Member #1596 Joined: Fri Jul 18 2008, 08:43PM
Location: USA
Posts: 65

Dr. Kilovolt wrote ...

There is no air but there is lossy dielectric. I bet it will heat up and melt, and spark over. But I would be glad if you could prove me wrong

I tried it out and there is no corona inside of the coil, but you were 100% right!!! AC corona is rediculous, it heats up 5mm thick insulation and melts it. How come is it so much hotter than DC one? I also noticed that DC corona occurs even with 5 mm insulation!!! However its very light and does not heat anything up.

What should I do to avoid a DC corona? Is it going to be there anytime load is disconnected?

Myke

Thu Aug 07 2008, 06:10PM

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

I'm assuming there is more corona with AC because it's capacitively coupled to ground. So I guess there would be more current flowing through the corona.

To avoid corona you would have have no sharp turns in the wire and the wire has to be thick enough so that the amount of voltage per unit of area is low enough so that it doesn't break down the air surrounding it. Another way of preventing corona is to vacuum pot it in epoxy (has to be vacuum potted to remove air bubbles) or to put the entire thing under oil.

Sulaiman

Thu Aug 07 2008, 07:05PM

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

Assuming that you are not prototyping a production design
I think it's better to way over-engineer devices like this,

e.g. have a much larger core than necessary to allow excessive insulation distances/thicknesses and wire cross-section etc. etc. because the loss of time and materials involved in an 'optimum' design will be much more than the extra cost of materials, in my experience anyway.
(unless you have lots of time and materials to spare)

I have a vague memory of reading that it's best not to wind over the airgap,
I can't remember where I read it or even the full reasons
but it does sort of confirm your observations.

I don't know your intended application
but if it's for general purpose use you may want to consider
how you could make either positive or negative eht.

Because the split-diode design effectively makes you choose either one or the other
because the diodes are embedded
and only one end of the windings is well insulated from the core/primary etc.

Personally I just bought 10 pieces of 35 kV TV eht flybacks for 20p each so I know my next few projects!

Dr. Dark Current

Thu Aug 07 2008, 07:41PM

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

Sulaiman wrote ...

Personally I just bought 10 pieces of 35 kV TV eht flybacks for 20p each so I know my next few projects!

I want too!

You know what would I do, connect them all in parallel for a nice 3kW HV supply

ad corona problems- DC corona is very easy to insulate (I was not able to punch a ~1mm plastic with 40kV DC) while AC corona is very hard to insulate, and it gets even harder with higher frequency.
Have you tried running your homemade transformer for longer periods of time (>5 min)?

RickR

Thu Aug 07 2008, 08:15PM

Registered Member #93 Joined: Thu Feb 09 2006, 04:11PM
Location: Houston, Texas
Posts: 64

Sulaiman, I believe you're right about placing windings over the airgap. I happened to be looking at:TI Design Doc, and at the top of page 5-4 it refers to possible melted windings resulting from placing them over the airgap in a flyback transformer. This would seem to correspond with what I remember seeing.

Rick

previous 1 2 3 next

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