50KV 500ma HFHV Transformer Design / Want to Build 50ma Version / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

50KV 500ma HFHV Transformer Design / Want to Build 50ma Version

1 2 next

Move Thread

LAN_403

LutzH

Wed Jul 08 2009, 07:54PM

Registered Member #1721 Joined: Sat Sept 27 2008, 08:44PM
Location:
Posts: 136

Hello to All:

I found a design for a reliable 50KV, 500ma HF transformer which is simply the best design that I have ever seen for such a transformer. It addresses and it solves all of the usual HFHV transformer problems, which folks are having when building these transformers, all at an efficiency level of 98.5%! This transformer gives you a full 50KV out at an incredible 25KW power level. I am looking for others who would be seriously interested in building a scaled down up to 50ma version to team up with.

This transformer is the answer to my problem in how to drive a planned million volt VM, in a reliable manner. For those who do not need to drive a big VM, this same transformer could be used for many other high power HV applications like X-ray, Tesla Coils, TEA Lasers, Plasma Projects, etc

Here is a section of the text describing a part of this unique and innovative design, I think many folks will find it very interesting!

Firstly, in order to obviate the divergent electric stress that
would otherwise occur at the ends of the windings, stress relief
rings are used. These rings, which are split to prevent shorted
turns around the transformer core, are connected to the ends
of the windings; indeed they form the connections. The usual
way to relieve stress is to alter the winding spacing at the ends
but this can be very wasteful of space. The stress relief rings
decrease the divergent field due to the high voltage, ,
by increasing the effective radius of the wire, . In this way the
maximum field, on the inside of the stress relief ring, is limited
to 5 kV.mm .
The other unusual feature is a slotted bobbin that allows
layered windings. In this way it was possible to increase the
number of turns considerably without placing large demands on
the insulation of the winding wire; indeed this was only rated at
600 V. Furthermore the parasitic capacitance between the ends
of the windings is reduced. A single cylindrical winding along
the surface of the bobbin, which only had 300 turns, was found
to have a parasitic capacitance of 130 nF at 1 kHz. However the
arrangement shown here, which had more windings (700 turns),
had a reduced capacitance of 64 nF. Such an arrangement does
give rise to locally high divergent fields at the corners at the
bottom of the slots and an electrostatic analysis was made to
ensure that a field of 5 kV.mm was not exceeded. The results
of this analysis are shown in Fig. 7.

This HFHV transformer design can take full advantage of Steve's, and others, excellent H-Bridge IGBT drivers, while avoiding most of the common problems, and unknowns which prevent people from attempting to build one.

I am firmly commited to building a slightly scaled down version for: 50KV, 50ma max., which will run at about 20-26KHz. I have already ordered 2 Feroxcube U126/91/20 cores for the project. These U cores will give a 123mm long section for the secondary, which is more than enough.

If anyone else is interested in building one, then please let me know. Maybe we can realize some savings in having the slotted bobbins, and end rings made. Sadly I do not own, or have access to a lathe so I will have to have the bobbins custom made. I had even considered using a square winding, and cutting the slots using a table saw, but I am not sure if there would be corona problems etc. with a square design? I have a big spool of 29G Polyimide premium magnet wire on hand which I would be willing to share for this project also.

The only thing that I have not been able to figure out is: How exactly do they wind the secondary? For example: Does the outer winding of one section, connect to the inner winding of the next? Maybe someone reading this in the forum, can better explain this? Any help with the best winding structure would be greatly appreciated :)

Now here is the good part, the actual design, Aloha.....Lutz :)

]hvhf_transformer.pdf[/file]

Finn Hammer

Thu Jul 09 2009, 06:30AM

Registered Member #205 Joined: Sat Feb 18 2006, 11:59AM
Location: SkÃ¸rping, Denmark
Posts: 741

Lutz,

If you search for the CCPS thread, you can learn from the mistakes we made......

Cheers, Finn Hammer

LutzH

Thu Jul 09 2009, 07:37PM

Registered Member #1721 Joined: Sat Sept 27 2008, 08:44PM
Location:
Posts: 136

Thank You.... I appreciate your suggestion, will do. I did not realize that I could search just one thread like this :) Maybe I can answer my sectional winding question here.

jpsmith123

Thu Jul 09 2009, 07:55PM

Registered Member #1321 Joined: Sat Feb 16 2008, 03:22AM
Location:
Posts: 843

Hello Lutz,

I want to make a HFHV transformer too (although I'm only looking to get 20 to 30 kv or so, w/midpoint grounded), but the difficulty of finding the materials and hardware I wanted to use is turning into a big PITA.

For example I wanted to use polypropylene film for layer insulation, but I'm finding out that you can't just go out and easily buy a small quantity of, say, 10 mil PP film anywhere.

So instead of that, I just ordered some "Scotch 69" glass-cloth insulating tape, in 7 mil thickness, which the manufacturer suggests is suitable for layer insulation in coils.

This stuff has a higher dielectric constant and a higher dissipation factor than I would like (3.87 and 0.0047 respectively @ 100kHz) but at least it is readily available in a small quantity.

Anyway, if I don't like the way it turns out, I think I am probably going to try to do something with ignition coils.

One other option (although it will cost some money) might be to go to a custom coil winding shop that will do small jobs...a shop with HV experience that can do vacuum encapsulation.

LutzH

Thu Jul 09 2009, 09:50PM

Registered Member #1721 Joined: Sat Sept 27 2008, 08:44PM
Location:
Posts: 136

Hello:

I have found Kapton, and other Polyimide insulation tapes regularly on ebay if I take the time to look, they insulate very well and these will go to 400 deg to boot. You may find some of these in short lengths, available there.

Also a great insulating fluid to use instead of messy oil is the Galden line, they are fluorocarbon fluids which will handle 40KV/mm. I like Galden HT135 the best it boils at 135 deg C. These fluids are totaly inert, but they are not messy like oil because they will evaporate overnight leaving a clean dry surface. Also these fluids are compatable with magnet wire insulation, plastics, rubbers etc. The only compatability problem is Viton, and PTFE which they will swell somewhat.

New these fluids cost $500 gal! but on ebay I have picked it up for as low as $50 for a Gallon. It is used in the semiconductor industry, and by the govenment to cool lasers and electronics. Some of these fluids will carry oxygen, and they were used to do the now famous video clip of a rat breathing a liquid. In humans they are used to allow the lungs of inhalation burn victims to heal, with the person breathing the fluid.

I would not recommend trying to breath it, but it does support the claim that they are not toxic, and completely inert. For smaller HV projects this stuff is great! It insulates better than oil, makes no mess, leaves no residue, and conducts heat much better than oil. Is viscosity is much lower than oil so it will penetrate the windings very well, for a few layers, you can get away with no vacuum impregnation.

Sorry to ramble on as you can tell I like this fluid very well for a few specialized applications like HV, and laser cooling.

Aloha....Lutz

jpsmith123

Thu Jul 09 2009, 11:58PM

Registered Member #1321 Joined: Sat Feb 16 2008, 03:22AM
Location:
Posts: 843

Thanks for the info on those Galden heat transfer fluids. I hadn't heard of those before.

I've previously looked into Kapton tape, and I think it probably isn't what I need.

Most of the readily available Kapton tape I've seen is only one or two mils thick, and the dielectric constant and loss factor don't seem that much better than the glass-cloth tape I ordered.

While one or two mils would apparently provide enough dielectric strength, I'm actually depending on the layer insulation more to provide separation between layers (thereby reducing the field stress and the capacitance), than to "insulate" per se; thus I'm more interested in the thickness of the material and the loss factor than other properties.

Steve Conner

Fri Jul 10 2009, 12:40PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Hi all

The only improvement in this design, over designs I've seen on 4hv in the past, is the corona rings at the ends of the secondary. Finn has done the slotted former, and Steve Ward has made some good transformers with the single-layer winding.

Nevertheless I think the corona rings should be well worthwhile.

When winding these things, yes, you fill one slot at a time and the wire goes from the top of one slot to the bottom of the next. There's no other way to do it. This winding technique gives the lowest self-capacitance possible, far better than a straight full-length layer winding.

To get the lowest self-capacitance and inter-layer stress with layer winding, you need to pile lots of narrow layers up into a flat disc as seen on old AC flybacks, called a "pie" on account of its shape and brown colour. And a transformer like the one under discussion amounts to a pile of these pies connected in series.

A square former would lead to considerably more field stress at its corners than a round one. The circular corona rings might help to relieve it if they were made big enough. Extra grading rings down the length of the secondary, connected to taps on the winding, would help too, but I think it really is worthwhile making a round former.

Finally remember that big HV transformers are easier to build and more efficient than small ones.

GeordieBoy

Fri Jul 10 2009, 02:35PM

Registered Member #1232 Joined: Wed Jan 16 2008, 10:53PM
Location: Doon tha Toon!
Posts: 881

I agree with everything Steve C said.

I might also add that seperately rectifying each of the pie windings might have some advantages over connecting all the pies in series and then rectifying the whole output.

Self resonance of the secondary winding as a whole requires current to slow back and forth over time. If you connect diodes between all of the pies it prevents this back-and-forth current flowing between the pies. Essentially, self resonance still exists, but it is killed after every half-cycle when the current tries to reverse.

So for example a 50kVDC supply, might have 10 pies that are each rectified to produce 5kVDC. These ten DC outputs are then connected in series to provide the required 50kVDC. The same theory can be extended to series connecting several complete DC power supplies that use seperate transformers. The AC properties like leakage inductance and stray capacitance of interconnecting wires become much less important when the majority of the voltage or current present is DC.

-Richie,

Dr. Slack

Fri Jul 10 2009, 03:36PM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

When winding these things, yes, you fill one slot at a time and the wire goes from the top of one slot to the bottom of the next. There's no other way to do it.

I've often wondered about that. Dropping the wire to the bottom of the next slot applies the full slot voltage across two wire insulation thicknesses, which limits the height of any one slot, though a shim of insulation over the wire before winding the next slot would improve things dramatically. From the point of view of wire to wire insulation, winding up one slot and down the next would be ideal, were it not impossible topologically. I belive it would have higher total secondary capacitance than all discs wound upwards, though still a lot better than layer upon layer. Allowing much taller slots and fewer of them might improve the self capacitance again.

It would be possible to do by passing a wire between two slots at the bottom, winding one clockwise, the other anti-clockwise as they built up, doing the same to every pair of slots, and then connecting the wires of adjacent slots at the top. The joint may be a corona hotspot, but some care should aleviate that. If you are considering corona rings or rectivifcation every slot or two, then they have to connect somewhere

RickR

Fri Jul 10 2009, 04:00PM

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

I've developed and documented a technique for stacking the windings without having to create and stack individual pies. It also gets around having the wire pass windings in a pie to get to the bottom of a pie: (http://4hv.org/e107_plugins/forum/forum_viewtopic.php?22460.0#post_22523). There is some wasted space with this technique, but if you've got approximately 4 inches to play with, this might not be a big deal. The big deal will be fabricating the bobbin. You will need a rather large cylinder of UHMW (or other) to form the bobbin. But I'm guessing that if you're in for a dime, you're in for a dollar. Plus, if enough people are interested, maybe you can work a group discount on the material and machining (if you choose not to do it yourself).

I've also modified this technique a little bit so that now there are two notches cut perpendicular to the winding grooves and separated from each other by 180 degrees: one notch allows you to bring the wire out at the top of a winding without having to go over the lip of the winding groove (better flashover protection) and the other notch allows you to bring in the wire at the bottom of a winding groove.

I've used this technique successfully on several flybacks and I don't see why it couldn't be scaled to this size of transformer, although YMMV.

If anyone does see some problems with this technique, please sound off as I'm thinking of starting to experiment some more with this way of winding.

Rick

1 2 next

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