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 2 3 4 5 next last

Move Thread

LAN_403

Bored Chemist

Sat Oct 13 2012, 08:12PM

Registered Member #193 Joined: Fri Feb 17 2006, 07:04AM
Location: sheffield
Posts: 1022

ben123324 wrote ...

and for those of you out there not so mathematically inclined: the core(if circular in cross section) must have a radius of ~8.31102 inches! 211cm in all important metric, but the imperial value inspires more fear don't you think

Err well, no not really because 211 cm would barely fit in a small room.
Which value is right?

genious 7

Sat Oct 13 2012, 09:34PM

Registered Member #2887 Joined: Sat May 29 2010, 11:10PM
Location: Panama City, Panama
Posts: 107

I'm sorry to be honest, but making such a transformer is extremely hard. What are you trying to accomplish. Would a smaller transformer followed by a voltage multiplier work?

Besides, what do you want. I am confused. Do you want a transformer that works with 30hz AC (which is very hard on a reasonable budget), or do you want a transformer that works with a high frequency signal that is pulsed at 30 hz.

A very important thing you haven't mentioned is current. Are we talking about microamps or Amps (or something in between).

I hate using the word "impossible", so I'll state it like this: getting an +-40kv transformer is really really really difficult and expensive. I strongly doubt you will be able to make one given the level of experience you have with transformer calculations.

You havent mentioned your application at all, which doesn't make things easier. If you dont want to state it because it's secret or something like that, get a clear list of specs, clear meaning

-voltage
-current
-frequency (30hz what, AC or pulsed hf); what shape (sine vs square)

I still feel you won't be able to make the transformer you want; you are probably approaching whatever you application is wrongly. Try giving us at least a vague idea what you want to use it for (Xrays? HV experiments? arcing? say something)

Ben Solon

Sat Oct 13 2012, 10:23PM

Registered Member #3900 Joined: Thu May 19 2011, 08:28PM
Location:
Posts: 600

Bored Chemist wrote ...

Err well, no not really because 211 cm would barely fit in a small room.
Which value is right?

Oops. Decimal point error somewhere. It checks out to 211cm and ~6.5 feet!

Dr. Brownout

Sat Oct 13 2012, 11:58PM

Registered Member #2405 Joined: Fri Oct 02 2009, 12:59AM
Location: Wisconsin
Posts: 140

ben123324 wrote ...

Pinky's Brain wrote ...

Lets say 400 turns primary so you'd get ~40 kV out with 1 kV in and a 16000 turn secondary ... you'd need a core area of 1000/(60*400*0.3) = 0.14 m2 :)

and for those of you out there not so mathematically inclined: the core(if circular in cross section) must have a radius of ~8.31102 inches! 211cm in all important metric, but the imperial value inspires more fear don't you think

more: 16000 turns without stacking factor on that core is 21.222 kiliometers. try finding 13.1867 miles of continuous wire with thin enough gauge that it wont fill your core window. then wind it!

Thats going to take some mighty fine wire, not to mention wire that small will mean lots of turns per layer which also means huge voltage difference between the start of one layer and end of the layer on top. I did a design for a 70kv transformer for an xray machine. There were many issues with the manufacturing of that particular winding because the wire was so fine even though it was done a commercial winding machine. They were stick wound 6 at a time. I had to use larger than normal end margins so the windings could be stair stepped to increase the creepage distance. Your best bet is going to be mylar layer insulation with crepe outer wrap. This transformer was submersed in oil. The stair stepping of the winding corrected the problem of layer to layer flash over. It can be done but special attention needs to be given to the details. These windings were also vacuum impregnated with a nitrogen push. We had a few break downs before we added the nitrogen push to the mix.

Also we baked them before submersing in the vacuum varnish tank to make sure ALL the moisture was out of the coil that we could get out of it. But they did work and are still being made.

Andre

Sun Oct 14 2012, 08:39AM

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

so many replies :) I was just out for a day,
what I'm trying to accomplish is compact transformer that would have a nearly square wave on the output at 30hz at 40kv 0.3 mA, the core I'm using 2 of this u cores together 81-33-16

thank you guys I'm very interested in the calculations more than anything else :)
that way I can understand the physics of it

Ben Solon

Sun Oct 14 2012, 01:19PM

Registered Member #3900 Joined: Thu May 19 2011, 08:28PM
Location:
Posts: 600

Given 1000v input, 30hz drive, .000396m^2 Ae, and 400 primary turns, your flux density is 189.39 teslas. The max for ferrite is .3.

Given 1000v in, 30hz drive, .000396m^2 Ae, and .3 T max density, you must have 252,100.8 primary turns. To get 40kv from that 1000v, you must have a bit over a million secondary turns.

Now use the given formula to figure out how many turns you meet once you see the value of high frequency. Notice how everything is getting smaller these days? That's because ferrite is being used in power management, and the higher the frequency you use with a core, the smaller the core can be. You already have a "small" core, so in order to use it you must use a frequency high enough to avoid saturation. See how the flyback power supply's around here can generate tens of kilovolts while only using a few primary turns and ranging input voltages? They all operate using high frequency.

Pinky's Brain

Sun Oct 14 2012, 02:30PM

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

So core area is ~2e-5, so you would need 40000/(60*2e-5*0.3) ~= 100 million turns on the secondary to not have it saturate with light loads. It simply doesn't work ... you'll need a huge transformer to keep the turn count down. The only reasonable way to switch that much voltage at those low frequencies is with stacks of semiconductors.

Andre

Sun Oct 14 2012, 06:08PM

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

i was looking at this design notes

and on page 3 it says the pulse duration is 5us and that's at 200KV. so I was thinking that building a slower transformer would with lower voltage and lower current would be a lot easier.
I need to get back to the drawing board.

Pinky's Brain

Sun Oct 14 2012, 06:11PM

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

Your pulse duration is 3000 times larger and your voltage only 5 times lower.

As I said, the only reasonable way to create pulses that long at those voltages is with a stack of semiconductors.

Ben Solon

Mon Oct 15 2012, 01:48AM

Registered Member #3900 Joined: Thu May 19 2011, 08:28PM
Location:
Posts: 600

slower does NOT mean easier. and 200kv from a stacked coil is impossible for us hobbyists. it's far easier to ramp up the frequency.

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