High Power HV ZVS Inverter W/Ferrite HF Transformer Design, 25KV 100ma / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

High Power HV ZVS Inverter W/Ferrite HF Transformer Design, 25KV 100ma

1 2 next

Move Thread

LAN_403

LutzH

Thu Oct 22 2009, 01:55AM

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

Hello Everyone :)

I just stumbled onto something on the net which got me excited! I deemed this worthy of posting on the forum ASAP since several of us are working on similar projects.

This ZVS SMPS HVPS (Wow did I abbreviate enough on that one?) design is fairly complete, and best of all it addresses the subject of at least three previous threads 1. (IGBT Inverter Design with ZVS) 2. (HFHV Transformer Design, and 3. (Ideal Bobbin Material for HFHV Transformers).

Finding this article now has made me somewhat happy that I had to delay my 50KV 50ma HFHV transformer project. The many thank U's still however stand, to all those who contributed to the previous threads.This design is far from perfect and I can already think of a few improvements, like a larger core like the U126 which would give much more comfortable margins, etc. Also a corona shield at each end would not hurt either.

For now I am pretty much set on: A PTFE bobbin, with segmented windings, but maybe trying some self supporting adhesive coated wire, so in affect a bunch of mini pancake coils in series, with the larger U-126 core. I did find some real thick wall PTFE pipe available which has a 1.5 inch ID, with a 3 inch OD, the bobbin could be cut from this easy. Then I can also drop it all in oil for extra insurance. As far as the driving electronics go this is not my strong point, here I would love to hear some input from more skilled members in this area, lets see if we can improve on this one :)

So now no more writing I will get straight to the "Meat":

]zvs25kv4kwinvrt.pdf[/file]

Herr Zapp

Thu Oct 22 2009, 07:29AM

Registered Member #480 Joined: Thu Jul 06 2006, 07:08PM
Location: North America
Posts: 644

LutzH -

I especially liked the first sentence of the introduction, where it is stated that "The high voltage spower supply includes positive voltage source and negative one, which is wildly used in such fields as scientific research, industry and medical treatment."

This sounds ideal for 4HV members, who tend to put HV power supplies to some pretty "wild" uses.

Note the details of the HV transformer secondary shown in Section 3.2, illustration b. This looks amazingly similar to the construction described by Finn Hammer recently for his HV supply.

Regards,
Herr Zapp

LutzH

Thu Oct 22 2009, 07:43PM

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

Herr Zapp:

Hello...

I think this was a "typo" but I love the humor in part because the irony is that "Wildly" is more correct than the intended wording as you pointed out. At least they did translate better than the now famous Japanese Ginsu knives, which came with a warning to: "Keep Knives Out of Children"

On the subject of "self adhesive wire" do you recomend any sources? I figure about a pound of 28g wire should be plenty, but who will sell this amount? My ideal wire would be 28g Polyimide double or triple build, with an adhesive coating. If you do not know of any vendors I will just start on my own to try to find this wire, and post the results, thanks.

rp181

Thu Oct 22 2009, 10:53PM

Registered Member #1062 Joined: Tue Oct 16 2007, 02:01AM
Location:
Posts: 1529

check out parawire, ask for a quote. They have both heat activated and chemical activated.
EDIT: wrong vendor. I meant MWS wire

Herr Zapp

Fri Oct 23 2009, 12:22AM

Registered Member #480 Joined: Thu Jul 06 2006, 07:08PM
Location: North America
Posts: 644

LutzH -

You'll have to do some searching to find small quantities of "bondable" magnet wire. 1, 2, and 5lb spools may be available from a large magnet wire distributor.

I don't know where you are located, but for smaller quantities you can see if there are any local manufacturer of inductors, transformers, or other electrical items that utilize magnet wire. I was fortunate enough to find a manufacturer nearby who was a general "magnetics" supplier, and wound coils and transformars of all types. They had stocks of magnet wire from #12AWG to #42AWG, and were willing to provide small quantities of wire (a pound or so) for free.

The next place to look is a motor rewinding shop that rebuilds electric motors. In the US, these guys typically purchase magnet wire in 75lb plastic "pails", and usually have lots of remnants laying around that you can purchase for the price of scrap copper.

If you have a large local magnet wire distributor, ask them if they have any "broken" spools. I once found a distributor that had accumulated a whole pallet full of 5, 8 and 12lb spools with broken or cracked flanges, missing labels, etc that they were going to sell as scrap. They allowed me to pick out as many spools of whatever sizes I wanted, and just pay whatever they would have gotten from the scrap metal house, which wasn't much at the time (before the '06-'08 copper price spike).

Regards,
Herr Zapp

jpsmith123

Fri Oct 23 2009, 01:33AM

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

I wonder what kind of adhesives they used to hold that teflon bobbin together, and to secure the HV lead to the bobbin?

Seems to me that this issue would need to be addressed before beginning to construct anything out of "slippery" plastics.

Anyway, I guess one advantage of a conventional multiple-layer-wound coil is that you would put the shield/primary winding on first, inside the bobbin, seperated from the first secondary layer by a low loss film of PP or PE; that way, the bobbin could be made out of CPVC, for example, without having to worry about dielectric loss/overheating problems.

Anyway, with regard to their design, I wonder if they really needed the layer insulation in each segment? It seems to me that it would be a real PITA to have to handle tiny strips of plastic like that.

Herr Zapp

Fri Oct 23 2009, 03:29AM

Registered Member #480 Joined: Thu Jul 06 2006, 07:08PM
Location: North America
Posts: 644

jpsmith123 -

When a design requires that "Teflon" need to be assembled with an adhesive, the parts are usually etched with a very active solution containing metallic sodium, then "primed" with a adhesion promoter. All this adds cost, so an alternative is to design the assembly so it uses external clamps or some other sort of mechanical structure to hold it together.

I agree that the interlayer insulation in each secondary "pie" would be a real production nightmare; better to use fewer turns per segment, and more segments.

Regards,
Herr Zapp

jpsmith123

Fri Oct 23 2009, 04:39AM

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

I consider the mechanical design of the bobbin a very important aspect of the overall design; thus I don't see the choice of particular adhesives, potting compounds, etc., as trivial details. It would've been better if the author went into more detail about what he used, IMO.

In any case, the larger point I was getting at was that the decision to use teflon has to be carefully considered. A small teflon bonding kit (contains etchant and epoxy) from McMaster costs about $77.00; IMO, yes, that's expensive. But putzing around with different mechanical arrangements can be expensive and time consuming too, especially if you're paying a machine shop to make things rather than doing it yourself.

LutzH

Fri Oct 23 2009, 06:49AM

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

That is where the potential distributing copper end rings come in, I can drill and carefully tap the teflon bobbin and screw these copper end rings directly to the teflon bobbin with nylon screws. Now with self adhesive magnet wire it will stay put once bonded, and the coil wire ends are soldered, or attached to these rings. It should work, at least it does do far in my head :)

jpsmith123

Fri Oct 23 2009, 11:35AM

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

If you are going to vacuum encapsulate the windings in something, like the author did, then I think you don't need any grading rings.

Speaking of that, one thing I don't understand about this author's design is, if you are going to go through the trouble of using layer insulation in your segmented windings, and then vacuum encapsulate the coil, as he did, then why not just go with a straight multiple layer type coil in the first place?

IOW, I wonder what advantage did the author get from the added hassle of segmented windings in this case?

1 2 next

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