Disc flyback / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Disc flyback

1 2 3 next

Move Thread

LAN_403

nOg

Thu Jul 21 2011, 07:02AM

Registered Member #4011 Joined: Mon Jul 18 2011, 08:14AM
Location:
Posts: 12

I got some problem with my winding flyback.
In disc shape as radhoo, It drive with ZVS and GDT. I could not get any spark from it.
The circuits still work. I also run with DC flyback, and got spark.

As in pictures, the disc flyback. I used 30AWG wire. Put in oil.

I had wind multilayers and donut flybacks and got black spot. (Not put in oil)

Pls give me advice.

radhoo

Thu Jul 21 2011, 02:18PM

Registered Member #1938 Joined: Sun Jan 25 2009, 12:44PM
Location: Romania
Posts: 701

now that I see the pictures, I can tell you you didn't build it correctly.

The wire's insulator doesn't hold the high voltage generated, so you get dark discharges, corona, and probably internal arcing, all with the same result: short circuit inside.

If you look at my design, you will see that I used a lot of epoxy while winding the coil, so in this way I get an easy inter-layer insulation, like in the case of standard flybacks.

Look again at my pictures here:

After each few turns you need to add epoxy, that will be put in place by the next few turns. You don't need any oil at all.

radhoo

Thu Jul 21 2011, 02:19PM

Registered Member #1938 Joined: Sun Jan 25 2009, 12:44PM
Location: Romania
Posts: 701

Relevant picture:

Also very important, the disk shape must be thin, 2-3 mm ! Else this insulation method will fail rapidly.

nOg

Fri Jul 22 2011, 01:46PM

Registered Member #4011 Joined: Mon Jul 18 2011, 08:14AM
Location:
Posts: 12

Thanks to finger me out.

Still not success winding yet.

What kind of epoxy did you use, general purpose?

radhoo

Fri Jul 22 2011, 03:20PM

Registered Member #1938 Joined: Sun Jan 25 2009, 12:44PM
Location: Romania
Posts: 701

transparent, general purpose.

Newton Brawn

Fri Jul 22 2011, 07:33PM

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

Hi nOg !

Now I understand what happens with the winding... The commercial copper wire insulation has a strenght of 200-250V at low industrial frequency. (60 Hz)
If the switching frequency is high, or transients are present this strenght is reduced.

ZVS and GDT or other switching devices may induce 20 or more volts per turn, in a higher frequecy than 60Hz.

So, the voltage between two adjacent turns may be 20 volts. (and the wire insulation will be very happy).

Supose that the coil is 3mm wide and the wire gauge is 30AWG (0.27mm diameter). The slot 3mm wide can house 3/0.27= 11turns.
The voltage between the fist turn and the 11th turn will be 11x20 = 220 volts.

As this layer is in contact with the previous layer ( that also has 11 turns), the voltage induced BETWEEN the two adjacent layers can go to 440V. (the wire insulation is in trouble and will fail)

The solution is provide a film insulation between EACH layer. The film insulation may be Mylar, Polyester, Nomex, 0.1mm thicknes. These insulation films may be purchased at motor rebuilding shops. Or at arts supply store. Engineer design office use Polyester paper for drawings.
Milar, Polyester, or Nomex paper 0.1mm thichness has a strenght of 800V considering the air inside the winding, that results the corona effect due two dielectics media betwen the wires.

How to do it:

1- Verify the voltage between each turn. Show the schematic of your ZVS, GDT, disclose the power source voltage, number of primary turns. Let me know, I can tell you the volts per turn.
2 - Calculate the the voltage generated by each layer, voltage between two adjacent layers. If you are using Mylar, Poly or Nomex, 0.1mm, the voltage between 2 adjacent layers shall be limited to 600V or less.
3- Verfy the number of layers, and the space taken by the winding, it shoud fit in the core window. if you need more turns, a second windind can be done, as a pancake winding. Sevral pankakes may be done.
4 - Verify the air gap between the pancakes and air gap between the pancake and core.... The air gap strengt may be limited to 2-3 kV/mm.
5- Epoxi glue may be used for mechanical purposes. No effect on insulation betwen turns or between layers .

Enjoy your winding

Regards

EDIT : see the thread ARC STARTER AND STABILIZER - COMPLETED
showing "Device to wind the pulse transformer - a turn counter and cranckcase"

nOg

Sat Jul 23 2011, 02:23PM

Registered Member #4011 Joined: Mon Jul 18 2011, 08:14AM
Location:
Posts: 12

@radhoo
I used this one, bought from resin shop (right), on the left, I just buy new type of epoxy.
The right one, I got failed. Left, new one to try.

@ Newton Brawn
Thanks a lot for your comment.

For my 3mm slot, it fit for 4-5 turns. So I was starting the test with 10 turns.
My simple winding device is likely as Radhoo.

My coil :
1. I use famous Mozzila ZVS, 2 ATX cores with 30 winds per each and two of 1MF MKP caps in series.
The primary is 6+6 turns.
2. I calculated my core when I input 50v for:
1 turn got 11-12v
10 turns got 260v with 1 layer, and 2 layers got 750v
3. When I wound 20 turns with 1 layer, I got 1kv.

And I used polyester insluted film.

Newton Brawn

Sat Jul 23 2011, 05:40PM

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

Hi !

I coud not find the Mozzila Schematic... But I think that the schematic that you refer is shown bellow. Please confirm.

]vladmiro_mazzilli_schematic.pdf[/file]

Do you mind to make a sketch of the ferrite core, showing the dimentions of the core, internal and external dimentions, core cross section dimention ? So I can figure out how many secondary turns (and voltage) you can get .

If the wire that you are using is the # 30AWG the diameter is 0.27mm, so in 3mm slot you can wind 10 ~ 11 turns. Please verify.

I agree that you can get 11 volts per turn,
so 11 turns you can get 121 volts,
so one one layer 11 turns + 1 layer of 11 turns you can get 242V,
so the insulation betwen the layers has to have a strenght of 242V X a safety factor (2-3X)

Please see the next post with the trafo drawings

Regards

cduma

Sat Jul 23 2011, 07:34PM

Registered Member #1822 Joined: Fri Nov 21 2008, 08:04PM
Location:
Posts: 300

It seems that you would need to buy a special purpose HV wire to use.

Newton Brawn

Sat Jul 23 2011, 10:21PM

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

Here is the design of the transformer. The 2 attached drawings give more details.
The secondary coil is wound dry, no glue during the winding operation. After the coil is finished it should dry for 1 hour in a oven at 70 degrees celsius and imersed in a clear varnish. I have used Alkid base varnishe, as exterior paint varnishe - spar. The Alkide base varnish is used for more than 40 years for electrical insulation. After the imersion, leave to dry for 2 - 4 hours at 50 degrees celsius.

]noqtransformer_1-2_model_1.pdf[/file]
]noqtransformer_2-2__model_1.pdf[/file]

1 2 3 next

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