About making flybacks as Uzzors2k(Eirik), Steve Ward et al. describe. / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

About making flybacks as Uzzors2k(Eirik), Steve Ward et al. describe.

1 2 3 next

Move Thread

LAN_403

Patrick

Tue Mar 16 2010, 03:14AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

i will begin making a 100Kv, 1KVA flyback this week. i like uzzors method but i include past experience form steve ward and others. i plan to use tapered secondary layers insulated from each other.

my question: it will be submerged in veggie oil [low viscosity], and will be made from polyester transparency film + paper, but do you guy's use 100 micron vacuum?, to draw the oil into the small spaces of the internal windings, thus displacing and removing air? i may purchase a pump for 112$ from a source for this purpose.

Or do some just submerge the coil, from its dry construction to its "new wet" home?

Also, will corona occur just between the core+primary+secondary, or inside the secondary between layers too?

-patrick

Arcstarter

Tue Mar 16 2010, 03:22AM

Registered Member #1225 Joined: Sat Jan 12 2008, 01:24AM
Location: Beaumont, Texas, USA
Posts: 2253

Anywhere there is oxygen, corona will form. So, we use something insulating to displace the air(with something that suppresses corona, like the misc oils), while also adding more insulation, even though it is a few mils at best.

I want to know how hard of a vacuum is typically considered 'good', too. I put my transformer in a thin plastic container, and put that in a coffee can for evacuation. Of course the coffee can imploded, but i was sure it would which is why i put some wood bracing in it to protect the transformer :P.

I want to know if a pump from an air conditioning system is powerful enough to get enough air out, and i also want to know how long you should ideally keep a vacuum for. I guess once it has reached the hardest vacuum possible, that is the best it will get. I would still let it run for a minute or so more :P.

For the record, the transformer's thin insulation did not break down, despite the 40kvac or more (ignited at about 2 inches, wth?).

...

Tue Mar 16 2010, 03:47AM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

Do you mean putting it under vacuum with oil (to get all of the bubbles out), or just putting it under vacuum?

In the first case, a refrigerator pump would work fine, in the second case the vacuum from a refrigerator pump would just increase the amount of corona/breakdown inside, and make matters worse.

Patrick

Tue Mar 16 2010, 04:06AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

[Arcstarter] are you a telepath inside my head, stealing my thoughts,!!! the questions you rasie are right out of my brain.....!!!!! read this webpage....and yes the 1.5cfm vacuum pump i have my eye on is good for 50-100 microns for Air-con service+repair. it is a rotary vane vacuum pump not a refridgerant pump.

its for 60hz steel lamaninated transmission txfmr's,oh man im so buzzed on booze right now....

[edit] >>the jist of the doc, is hold a vacuum dry for 4-8 hrs @100-200microns, then and oil slowly enoguh that ur pump can still hold a high vacuum while oil is allowed in, then hold oil above the fill level for 1-4 hrs at 100-200 microns, then let it go back to atmospheric... i belieeve this would be simmilar for ferrite types for our purpose.

[...] hey i would pull the vacuum for sometime, hold it , then let the oil in , then reup the vacuum, then hold it, for minutes too hours , then let it back up to atmospheric while under oil, then im done , i would not hold the vacuum as you implied while operating the txfmr, being a vacuum tube like condition inside the txfmr would cuase fatal electron emission, as i beleve you are suggesting.

teravolt

Tue Mar 16 2010, 04:39AM

Registered Member #195 Joined: Fri Feb 17 2006, 08:27PM
Location: Berkeley, ca.
Posts: 1111

it may be easier to use a flyback with a voltage multipier if you are after HV DC

Patrick

Tue Mar 16 2010, 04:41AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

unfortunatley my modified DeSeversky ionocraft needs square wave AC , DC is not desireable in this configuration. otherwise i would use a VM. definently.

hey tera volt respect for your 8 stars i have, but you arnt a berkely communist are you?
and when do i get my 4th dam it !?

McCluskyCustom

Tue Mar 16 2010, 05:06AM

Registered Member #2538 Joined: Sat Dec 12 2009, 06:56AM
Location: San Luis Obispo, CA, USA
Posts: 10

Hi,

In building a HV transformer rated for over 20KV, I looked into this a lot. Here is what I found from a paper on Oil Impregnation in HV Transformers:

"High voltage power transformers are vapor phase dried in a
vacuum oven. While the winding structures are still hot, the
degassed, dehydrated and filtered oil is introduced into the oven
to impregnate the solid insulation at elevated temperature. The
time duration is dependent on transformer voltage ratings and it
is only a part of the total impregnation process.

Before introducing the oil into the tank, a low air pressure of
less than 1 Torr (1.33 mbar) is suggested to â€˜suckâ€™ the air out
of the pores of cellulose materials. When a minimum pressure
inside the solid insulation is obtained after a certain period of
time, the oil can be slowly introduced into the tank. The vacuum
can be released once the solid insulation is totally immersed by
oil. Thus, a great pressure difference can be formed, taking the
advantage of pressure of the atmosphere plus the oil pressure
outside the cellulose materials."

For my transformer, I built a vacuum chamber from 6" PVC, 10" long, and used 1/4" Lexan sheets on the top and bottom sealed with a rubber gasket. They definitely bent inward, but did not break. I used a 75 micron (1 Torr, suggested in the paper, is 1000 microns) 2.5CFM pump made for air conditioners (Harbor Freight cheapo, $80) connected with some misc plumbing (braided hydraulic lines) and a valve ordered from the local NAPA.

I baked my xfmr and the oil for 30min in the oven, and then put it in the vacuum chamber and pulled a vacuum for 30 min. The chamber had the container with oil inside, with the transformer suspended above it. After 30 min, I lowered the xfmr while still under vacuum (by a magnet on the outside attached to a magnet on the inside that held up the xfmr). Even then, some bubbles continued to come from the xfmr, but I was over it and released the vacuum to pressure-fill all the capillaries in the paper insulation.

In the transformer itself, I only used paper (60# Kraft paper, 3 sheets per layer) as my insulation (which is filled with oil). This was because I knew the breakdown of the Shell Diala AX oil I was using and the paper thickness. The problem with the transparency film is possibly blocking the oil from making its way throughout the secondary. If you use thick paper on both sides of the film, it will probably be okay, and having the film makes it more breakdown resistant since your oil is not characterized. Also, some people say veggie oil can go rancid after some time *stink*. When using pieces of tape to hold anything together (make sure it is non-conductive like Kapton or Mylar, as some tape adhesives, or even sharpie, are conductive at high voltages), don't overlap the tape on itself because it can trap air bubbles that the vacuum will not remove. At the HVHF, you will generate corona in the air pockets and potentially release the air, causing bad news in the xfmr (not fun to have weeks of work burn up with smoke).

This was my first oil-submerged transformer, and it worked very well, until I ran it for too long and overheated the secondary, causing a secondary short and smoke coming from the oil :( It was a "natural stopping point" in my project.

Oh, and if you are submerged in oil, you do not need to taper the secondaries. The oil has a very high breakdown (xfmr oil is ~700V/mil, ideal low-moisture conditions). Tapering the secondaries causes more secondary capacitance due to having more layers. In a flyback, this capacitance will have to be charged and discharged every cycle, (remember power=0.5*C*V^2*f) and will suck off output power and waste it, not to mention increasing secondary circulating currents and I^2*R losses. And, oil has a dielectric constant near 2.2, meaning after you submerge the xfmr the capacitance just doubled. If you want to reduce the capacitance, you can sectionalize the secondary.

I would definitely suggest submerging it in oil, and only under a vacuum. Otherwise you will trap some air in the secondary and risk arcing the secondary. Corona will occur EVERYWHERE there is high voltage. i.e. especially in the secondary between layers. Of course, I have not oil-submerged an xfmr without vacuum, so I have no experience the other way.

Good luck!

p.s. why choose a flyback rather than a topology that utilizes more of the transformer, for such high power and high voltage?

jpsmith123

Tue Mar 16 2010, 06:09AM

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

Trying to create and use 100 kv high frequency AC directly from the transformer may be very problematic. What does the load look like, electrically? Between the capacitance of a large multilayer secondary winding and a large load/stray capacitance, you will have large reactive currents flowing in the secondary circuit. You may not be able to run at the frequency you want to. You should try to estimate what your load will look like and then simulate the whole thing as a system using a circuit simulator.

Patrick

Tue Mar 16 2010, 06:21AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

yes jpsmith my output device looks resisteve/capacitive, draws little current, has high voltage drop, and causes little voltage ripple, even with high internal impedence of the secondary.

jpsmith123

Tue Mar 16 2010, 06:24AM

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

What kind of frequency are we talking about?

1 2 3 next

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