Calculating self-resonant frequency of HV Transformer? / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Calculating self-resonant frequency of HV Transformer?

first previous 2 3 4 5 next last

Move Thread

LAN_403

Ash Small

Sat Feb 02 2013, 02:30PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Well, I think I can wind the primaries from foil 'discs', and place them between, and adjacent to, the four pancake secondaries. The only issue will be insulation.

It seems to me that the idea is to contain all of the 'flux' within the volume enclosed be the windings themselves. Also, if you have the lower voltage windings on the outside, it reduces EMI.

Apparently, this will increase capacitance between primary and secondary, but I think it's worth trying if it significantly reduces leakage inductance. I may wind a few alternative designs, and measure self-resonant frequencies of each. Thanks for the help.

Newton Brawn

Tue Feb 05 2013, 12:27PM

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

Hi Ash !

I just made some calculation to verify the working inductiion of your core:

Your design parameters are:
Core cross section is 30mm x 60mm,
50 turns to produce 1250Vpeak,
Square wave, @ 25kHz:

The core maximun induction will be
Bmax = V/(4 x N x F x A)
Bmax = Max Induction in teslas, [T]
V = Peak voltage proced by a coil on N turns, [V]
N = number of turns in the coil,
F = frequency, [Hz]
A = Core cross section in SQUARE METERS, [m^2]

The core cross section is: 0.03m x 0.06m = 0.0018m^2

So
Bmax = 1250/(4 x 50 x 0.0018 x 25000) = 0.1388T

I believe that Bmax can be more, resulting in an easier transformer construction.

Also I have more comments, BUT first let me know the working cycle of your transformer.

May I assume that:
1- The transformer will work 2 minutes and rest 10 minutes ?
or
2- The transformer will run continuously during one hour?

Regards

Ash Small

Tue Feb 05 2013, 12:45PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Am I correct in assuming that Bmax can be pushed to ~0.2T?

Ideally, I want it to run continuously without overheating. It will have 'active cooling'.

Newton Brawn

Wed Feb 06 2013, 01:15AM

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

Ash:

Considering 0.2T as Bmax, F=25000Hz, A=0.0018m^2

The number of turns in a layer to produce 1250V could be:

Bmax = Vmax / (4xNxFxA) >> N= Vmax / (4xBmax FxA)

N= 1250/4x0.2x25000x0.0018 = 35turns

The insulation between two adjacent layer shall be designd to effort 1250 X2 =2500V plus some safety factor ....

Assuming that the insulation to be 0.5mm, each layer (wire +insulation) high will be 0,7mm. ..So... each pancaque will take 7mm. I think yous core window has room enough, The form high has to be calculated.

Regarding the "ative cooliong", is it a fan ? Or ONAF or OFAF?
I am using since 1970 poiyester (Mylar) film, as insulation for small transformers, workiing temperature, is class F .

When winding very thinner wire, 0.4mm and less, use polyerster 0.1mm. If nescessry se 2, 3,4, or more 0.1mm sheets between the layers You cam buy 0.9m x1.2m in the shops tha rewind motors and generators . Last time I have payd $4.50 for one 0.1mm sheet.

Transformer oil, from Shell, Esso. Mobil you can get from contractors that make maintenance in industrial transformers.

Regards

Ash Small

Wed Feb 06 2013, 04:05PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Thanks for the reply, Newton.

I assume the cores will run cooler if Bmax = 0.14T ?

I was planning on having 1mm between layers to reduce inter-layer capacitance. I may even leave more than 1mm between layers.

I was planning to use paper as insulation, and putting it under vacuum under oil.

I'm currently working on using 'pie windings, with foil primaries, but haven't yet finalised the primary design.

Newton Brawn

Thu Feb 07 2013, 02:51AM

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

Info from the Epcos catalog regarding core U93/30:

One U core working at 0.15T and 25kHz the losses are 16 watts .

One U core working at 0.20T and 25kHz the losses are 30 watts,

One U core working at 0.25T and 25kHz the losses are 46 watts

I am gessing that you are going to use 4 U cores ?? Please confirm .
So, one fan are required to cool down.
Where you are going to get insulating paper ? I think it is a Kraft insulating pH neutral, 0.1mm ?
I have used copper roof sheet for primaries.,...sometimes aluminum roof sheet 0.3mm thick
from Home Depot is experimental coils.

I think is tme you start to produce and post some drawings of your project.

Ash Small

Thu Feb 07 2013, 11:35AM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Thanks for pointing out the figures. I do remember seeing a figure of 16W before, when I was considering what frequency to run.

I am using four cores.

If I use polystyrene bobbins I will need to keep the temperature as low as possible, as it gets soft well below the glass transition temperature of 100C. 16W per core is still over 60W total, so I plan to circulate oil through a fan cooled radiator.

There are thousands of listings for kraft paper on Ebay.

I will produce more drawings once I've considerel alternative primary designs.

Patrick

Fri Feb 08 2013, 07:28PM

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

Ash Small wrote ...

The working temp. of most epoxies isn't that much greater than polystyrene. I'm currently looking at PE, PP, PVC and ABS, etc. Interestingly, PP is above it's glass transition temp. already at room temp. I suspect that PE and PVC are as well, but I've not checked.

What are the properties of polyester and epacrin resins?

well if your whole thing is under oil, and your using a transmission radiator like i did, then your hottest point should be way below
200F / 100C...

pics here from 2009ish

Ash Small

Fri Feb 08 2013, 08:05PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Patrick wrote ...

well if your whole thing is under oil, and your using a transmission radiator like i did, then your hottest poit should be way below 200F / 100C...

I am planning on using a transmission radiator or two, and if the cores are only losing ~60W, I should be OK, but there are also copper losses, and I may end up pushing the Bmax higher than 0.15T, if I need to reduce inter-layer capacitance.

I need to wind a secondary, and do some measurements.

BTW, What thickness of epoxy do you think I'll need for 50kV@25kHz?

Steve Conner wrote ...

What thickness of foil is needed? Rod Elliott recommends a current density of 2.5 amps per square millimeter for transformer windings. Copper has a skin depth of 0.4mm at 25kHz. So, the foil needs a cross-section of 8 sq mm, and a thickness no more than about 2 skin depths.

So 0.8mm thick X 10mm wide should be OK for ~20A. Would more layers of thinner foil be better? (ie 8 layers 01.mm thick, insulated with paper (The primary windings will be mounted vertically, maybe I should draw what I mean)

Patrick

Fri Feb 08 2013, 08:27PM

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

Ash Small wrote ...

Patrick wrote ...

well if your whole thing is under oil, and your using a transmission radiator like i did, then your hottest poit should be way below 200F / 100C...

BTW, What thickness of epoxy do you think I'll need for 50kV@25kHz?

well thats why i suggested S2 or E glass, since youd only need ~40% epoxy for ~60% glass volume, since the glass has better E and M field behavior than epoxy... but if your asking for v/mil, generally 500v/mil, is the breakdown, so for less than 0.1 inches id say 300v/mil operation for a good quality epoxy or PMMA.

first previous 2 3 4 5 next last

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