transformer question / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

transformer question

1 2 3 next

Move Thread

LAN_403

thrival

Mon Oct 08 2007, 12:33PM

Registered Member #1019 Joined: Sat Sept 22 2007, 02:39AM
Location:
Posts: 29

OK, just so anyone reading this understands, this is NOT about "free-energy," "over-unity"
or anything like that, but a SPECIFIC question about increasing VOLTAGE in a transformer
with corresponding current trade-offs.

I dissassembled a couple of MOTS and harvested the secondaries of about 2,160
turns each. My intent is to wrap a primary around one of the secondaries of about 10
turns. If I run 12V sine wave pulses into my primary I should pull about 2,592V from
positive to negative peak, or 2,592V clamped DC.

Now I'd like to run that output into the primary winding of 10 turns of another transformer,
just like the first. Since my voltage amplification factor is approx. 216X, my output from
the second transformer should be 216 X 2,592 = 559,872V. Again this is + to - peaks, AC
reading would be exactly half as much.

Theoretical question here: assuming the baked varnish winding coatings can handle the
stress, what would happen if I mounted both set of transformer coils on a common iron
powder core? (Note the core might actually be a dielectric with greater or lesser amounts of
iron powder added, to provide an inductance between that of air and typical iron cores.) Or
should I omit any iron in the core to prevent saturation.? An interesting phenomenon
I forsee if both transformers are magnetically coupled, is the field of transformer #1 initially
influencing transformer #2; however since transformer #2 will produce a much higher voltage
due to the method of winding, it's attendant magnetic field will influence transformer #1 to
an even greater extent. Of course without a core there shouldn't be any saturation, but the
coupling effect would still apply.

My further intent is to ground the secondary output of one lead from transformer #2 with one
of the primary leads from transformer #1, and running this as an ignition coil with square-
wave, flyback resonance.

The point is to make a DIY ignition coil of higher parameters than commercially available
and at zero cost.

Of course I cannot assume (even though I did for sake of the thought experiment) that the
windings which are multi-layered and in close proximity will handle the stress and not
arc-short internally. My point of writing this is to ask if anyone else has done this or a similar
experiment, so I might learn from their experience so I don't destroy a couple of perfectly
good windings; (hack-sawing into MOTS and removing them is a fair amount of work.)
thanks.

Capper

Mon Oct 08 2007, 01:29PM

Registered Member #914 Joined: Fri Jul 20 2007, 06:22PM
Location: South Bend, IN
Posts: 85

Theoretically, You would have to pump 46 A @ 12 V into it just to get 1 mA out (assuming zero losses), and your coils would get very hot just from resistive heating.

In practice, none of the wires could handle the high voltage (including the spark plug wires).

ragnar

Mon Oct 08 2007, 02:05PM

Registered Member #63 Joined: Thu Feb 09 2006, 06:18AM
Location:
Posts: 1425

Seriously, seriously, have a look at some of the voltage multiplier circuits out there.

If you just take a perfectly ordinary, MOT, don't modify it, you can build a driver that will run a square (or sine, if you want) through the primary at hopefully < 200Hz so the iron can handle it.

You could even use a self-oscillating push/pull like the "Royer ZVS" seen around in these forums. If you can find a MOT with a centre-tapped primary for 120V/240V, the hardest part of your job is done.

A well-design cockroft-walton multiplier on the output will take the voltage up to some reasonably high (but sane) values.

As for two coils on the same core with 500kV? You are dreaming.

Somebody here has made a neat high-frequency driver using a ferrite CC core with a MOT primary just threaded into it. Can anyone remember the link?

uzzors2k

Mon Oct 08 2007, 02:29PM

Registered Member #95 Joined: Thu Feb 09 2006, 04:57PM
Location: Norway
Posts: 1308

Matt Bingham wrote ...

Somebody here has made a neat high-frequency driver using a ferrite CC core with a MOT primary just threaded into it. Can anyone remember the link?

Absolutely.

Antonio

Mon Oct 08 2007, 04:45PM

Registered Member #834 Joined: Tue Jun 12 2007, 10:57PM
Location: Brazil
Posts: 644

If you put the four coils in the same core, you will have a normal transformer from the first winding to the last, with the two other windings in parallel acting as a pair of short-circuited windings, with a high current on them limited by resistive and inductive losses. With two cores, the second transformer will saturate and clamp its input voltage to a few tens of Volts. The input current will be huge. Anyway, the final result will not have a nice smell...

Shaun

Mon Oct 08 2007, 05:03PM

Registered Member #690 Joined: Tue May 08 2007, 03:47AM
Location: New Jersey, USA
Posts: 616

Listen, if really do want a few hundred kV out, there are several proven ways to do this.

1. Tesla coil. Not the easiest, but more than capable, and a thousand times proven.
2. Marx Generator. Easily capable of over a megavolt, the disadvantage being short, intermittent pulses.
3. CW multiplier. Capable of continuous output. (you will need an HV-AC supply to power it with, AC flyback would be excellent)

You should really forget about cascading transformers (especially solid-cored ones). Even under oil, the wiring will just not be able to handle more than a few tens on kilovolts.

Check out Reaching's "1MV Bipolar Marx" thread, and Christopher Miles Hooper's "Bipolar Marx" thread (both in Projects section), maybe you'll see something you like...

thrival

Mon Oct 08 2007, 09:27PM

Registered Member #1019 Joined: Sat Sept 22 2007, 02:39AM
Location:
Posts: 29

Thank you, all.

Antonio: I get it. So even if the windings could take it, I would
need to avoid magnetic coupling between transformer #1 & #2, for
them to behave like two separate devices.

Matt & Uzzors:

Thanks for the flyback link. That's the kind of plasma I'd like to make
at the plugs, I have a nice mini-MOT, removed the 3V winding and shunts,
had enough room to wrap about 10 or 12 wraps of #16 insulated wire between
the primary and secondary. But I'm afraid that at flyback resonance, will cook
the insulation. I ran some square waves from a frequency generator (12 - 35V,
very low current) through the primary, with a voltmeter on the secondary, and
found this core likes to resonate around 1,8 - 2 khz. But I think I need to run
it faster than that just to stay ahead of engine RPM's. Anyway I still need more
V than the MOT can provide, to jump the spark plug gap. Also the MOT
secondary only has one output unless I tape off the core where its grounded.
A CW multiplier requires AC. Too many stages for the frequency, me thinks.

Does anyone know how quickly a flyback starts to oscillate from ON/OFF? Is it
instant, or does it take time to begin oscillating? Reason being I need to time
the ON/OFF cycles from the module, through a secondary transistor that can
handle the ignition timing AND the AF/RF passing through.

Sam Barros @

has a nice flyback driver using but one
transistor.

ragnar

Tue Oct 09 2007, 03:05AM

Registered Member #63 Joined: Thu Feb 09 2006, 06:18AM
Location:
Posts: 1425

For ignition purposes, the startup time of most of these (transformer + voltage multiplier / flyback) should be plenty fast enough.

cjk2

Tue Oct 09 2007, 03:18AM

Registered Member #51 Joined: Thu Feb 09 2006, 04:17AM
Location:
Posts: 263

Just as a side note, the 2n3055 driver on Sam's website is a very poor design, you will find that a royer style driver gives Much better results from the same input.

Dr. Slack

Tue Oct 09 2007, 07:07AM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

Now I'd like to run that output into the primary winding of 10 turns of another transformer,
just like the first. Since my voltage amplification factor is approx. 216X, my output from
the second transformer should be 216 X 2,592 = 559,872V. Again this is + to - peaks, AC
reading would be exactly half as much.

No, No, nien, fie, NO, pshaw, Noooooooo, nyet, never, negatory, non, definitely not (got it yet?)

Voltage amplication factor is 216x, up till about 1 volt per turn, governed by the frequency and the core dimensions of a MOT. Above that the core saturates, and it stops "obeying" the niave transformer equation and starts doing something more complicated, including changing the operating conditions of the preceding stage. RL is more complicated than grabbing a simple picture, and then blindly applying equations outside their area of applicability. You can't run a 120v input transformer at 2kV input and get good results. If you could then people would have done it already, already. Same with ignition coils. If you want X volts, then use an appropriate design, rather than looking to cobble together otherwise good components in inappropriate ways.

1 2 3 next

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