after how many turn on a HV transfomer you start insulation / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

after how many turn on a HV transfomer you start insulation

first previous 4 5 6 7 next

Move Thread

LAN_403

Dr. ISOTOP

Mon Oct 15 2012, 07:57PM

Registered Member #2919 Joined: Fri Jun 11 2010, 06:30PM
Location: Cambridge, MA
Posts: 652

Andre wrote ...

so if I reverse enginering this transformer, this would be my calculation
N = V / (4*f*B*Ae)
V=110
f=60hz
B= about 1T for nickel iron core

Ae = I'm guesing .0004m2

N = 110 /(4*60*1*.0004)
N = 1146 on primary
from the pictures I think it runs 2 secondaries at 35KV
(35000/110)*1146 = 364363 Turns on each secondary. since it peaks at 70KV it can be 5% less turns

would this be about right?

XRT's often have internal multipliers to further reduce the turn count.
If you run your inverter at 300KHz (doable with modern IGBT's), you can push maybe 120V/turn on the largest cores that can be reasonably sourced. This means you can run a 2-turn primary off line voltage or an intermediate step-up converter, and put 300 turns on the secondary using fine wire, which is doable.
Keep in mind you'll probably still need to immerse everything in oil to prevent corona.

GrantX

Mon Oct 15 2012, 11:31PM

Registered Member #4074 Joined: Mon Aug 29 2011, 06:58AM
Location: Australia
Posts: 335

Andre wrote ...

so if I reverse enginering this transformer, this would be my calculation
N = V / (4*f*B*Ae)
V=110
f=60hz
B= about 1T for nickel iron core

The transformer has less turns than that. If I recall, someone on the Fusor.net forums took some measurements on this transformer and discovered it ran at about 1.6V per turn and was well into saturation at full input. But yes, you are correct, it has 2 secondaries capable of 35kVp each, each one has its own plastic bobbin. Strangely enough, the secondaries have very minimal insulation, relying entirely on oil to be squeezed into the windings, which is partly why this transformer is so fragile (I wouldnt ever draw arcs from it at full voltage, I think someone on Fusor.net killed theirs by drawing arcs from it)

Andre

Mon Oct 15 2012, 11:49PM

Registered Member #6921 Joined: Wed Sept 26 2012, 07:47PM
Location:
Posts: 109

Thanks XravenorX I found the transformer

if it has 174 on the primary then it has 55363 on each secondary,
some people said it would need over a million turns for something low frequency like this,

I think I should start a new thread of reverse engineering this transformer, I will really like to know the calculations of it. cause this is getting of topics of this thread.

Thanks guys for giving different options, but I'm more interested of how it works and how can I build a transformer from scratch rather than just buy one, and the calculation of it.

Pinky's Brain

Tue Oct 16 2012, 12:28AM

Registered Member #2901 Joined: Thu Jun 03 2010, 01:25PM
Location:
Posts: 837

First find a double E-core or E/I-core 10 inch square ... that's what about what you would need for a continuously run 30 Hz version of that thing.

So uhm ... we are about at that stage where you are trying to build a pole pig (which I think is the most realistic source for a core BTW). Small ... :P

Andre

Tue Oct 16 2012, 04:27AM

Registered Member #6921 Joined: Wed Sept 26 2012, 07:47PM
Location:
Posts: 109

Pinky's Brain I'm looking to do all the calculation necessaries and send it to a coil company to be build, they are a friend of a friend, so it would be cheap, but they want all the data, if they have to designed is more expensive. I just need help with the formulas, and to confirm the calculation I did, but most people ignores that, and talk about something else, I don't understand why

Pinky's Brain

Tue Oct 16 2012, 07:40AM

Registered Member #2901 Joined: Thu Jun 03 2010, 01:25PM
Location:
Posts: 837

You don't understand why?

First off you said you wanted to do it with a small ferrite transformer, so we talked about why it's impossible.Then you talked about cloning a x-ray transformer which is running so close to saturation that theoretical calculations using simple static permeability formulas is null and void. Personally I'm simply assuming you still want to build a low current continuous use 30 Hz transformer ... because why the hell clone a transformer which doesn't do what you want?

There is the fact that both the winding and reliability of impregnation for production use of these things is going to be so far out of amateur considerations that giving concrete recommendations based on first order approximations is seen as a bad idea ... apart from the guy who does it professionally I doubt there are very many members in this thread who have ever tried to wind polepig sized 100K turn range secondaries with oil impregnated insulation between layers. So we just point out errors in your thinking which are obvious enough, without giving concrete recommendations.

Finally the actual requirements for your transformer are incomplete, hidden in the middle of the thread and come with a completely unrealistic core. Lets try to get it straight. You want a 40 kVpp 0.3 mA split secondary transformer (20 kVpp per secondary), it needs to run at 30 Hz, it needs to be able to stand continuous use ... and it doesn't need to be small :) ... is that correct?

Try plugging in the numbers in the Excell sheets here (with 28 kV secondary voltage, he uses RMS and obviously doesn't assume a split secondary) and experiment a bit with the parameters :

PS. you're approaching this as an engineering exercise, while it is an experiment ... for it to be the former you first need experience. Be careful about how much money you spend (or how many favours from your friend).

Wolfram

Tue Oct 16 2012, 02:52PM

Registered Member #33 Joined: Sat Feb 04 2006, 01:31PM
Location: Norway
Posts: 971

Andre wrote ...

well Wolfram it been 48 hours of silence, since you havent reponded I'm going to tell you the answer,
if is an 100 layer coil it does not matter if you start insulation from a layer 25 or 50 or 75, it would be the same as not putting insulation at all, because at 40000 KV you would have about 400V in each layer, so as __=|(:3)-|--{__ said best way to separate in 2 diferent transformer or put the insulation in every layers, if you think I'm wrong you can prove me with science(formulas) not philosophy.

I don't visit the forum every day.

What are you talking about? I was not talking about insulation, I was talking about the self resonant frequency of the transformer, which will limit your maximum drive frequency to well below the 40kHz you said you wanted to run it at earlier. I am not talking philosophy but experience, and while a calculation would prove my point, it would probably be quicker to actually wind a test coil.

As several others have pointed out, try to get an actual idea of what you need before you start designing the transformer. You mentioned driving it with a 30Hz square wave, what sort of rise time do you need? Do you need the actual 30Hz fundamental on the output or do you just need short pulses with a 30Hz repetition rate? What is it being used for? These things have a lot to say for the design, and you're just wasting everybody's time when you come up with these vague and incomplete design specifications that you keep changing.

Andre

Tue Oct 16 2012, 03:50PM

Registered Member #6921 Joined: Wed Sept 26 2012, 07:47PM
Location:
Posts: 109

sorry guys for the confusion, yes at the bigining I wanted to get to 40Kv to run an x ray and do some hovering experiments

but this got more interesting that the frequency can make a big difference,
and the reason I wanted to do all the calculation I wanted to do the Tesla way :) he use to do all the calculation first before he build something to make sure it would work, while others would just go on trial and error,
I been looking so many transformer design notes, and is a bit confusing about the formulas so I was looking for help here not to do the calculations for me but to tell me if the calculations I did are correct, in the past week I had learn so much in this forum, and I will like to know more about how the transformer works,
things like how you find out when it saturates?
whats the maximum voltage you can have between layers?
was the calculation I did correct?
They are so many varienble, and I know I'm trying to learn all in one thread :)

Steve Conner

Tue Oct 16 2012, 04:16PM

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

The lifter guys normally use TV flyback transformers to generate a HV DC output. For a pulsating output, they gate the drive to the flyback on and off. For example (excuse the Blaze labs link :) )

This is much easier than trying to wind a transformer to generate 40kV at the pulsation frequency directly.

Andre

Tue Oct 16 2012, 04:23PM

Registered Member #6921 Joined: Wed Sept 26 2012, 07:47PM
Location:
Posts: 109

LOL no matter how much I try learn about how a tranformer work, and the calculation to build one

, the best option I keep getting is buy a trasnformer from ebay.
come on guys :)

first previous 4 5 6 7 next

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