Transformer Primary / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Transformer Primary

1 2 next

Move Thread

LAN_403

currentkills91

Tue Mar 13 2012, 09:20AM

Registered Member #3831 Joined: Thu Apr 14 2011, 02:54PM
Location:
Posts: 265

Hey everyone, I'm rebuilding my mini capacitor charger, and I have a question about the transformer.

I need help figuring out which way to wind the 2 primaries, I have one done but I'm not sure which direction the other one should go.

Here's a picture, sorry about the quality

Pins 2 and 3 are the center tap, which would mean current flows in through the top of the wind and down to pin 4. So can someone tell me if I can wind the other primary the same way, or will I need to switch the wires around so the current flow starts at the bottom of the coil? If that made sense?...

Forty

Tue Mar 13 2012, 04:56PM

Registered Member #3888 Joined: Sun May 15 2011, 09:50PM
Location: Erie, PA
Posts: 649

This is for a zvs i take it? Picture it as one continuous winding with a tap in the middle. Since the windings have the same orientation but different current directions, that must mean they create opposite magnetic fields.

So we need to get your windings to make opposite magnetic fields through the secondary, but from different legs. Looking at your picture, your primary (+at pin 3 and gnd at pin 4) will make a magnetic field oriented downward through the secondary. This means the other primary will have to make a field oriented upward through the secondary.
I'm not sure if it matters which way you go about making the field in that direction (+ could be applied at either end, with the coil orientation being what controls the B direction)

I suppose you could make it consistent with the other primary, and have + applied at the top. This would make the second primary look like a mirror image of the first (wire from pin 2 up to the top of the transformer, wrapped around the outside of the core going into the page at the left, and out at the right, clockwise if viewed from above)

currentkills91

Tue Mar 13 2012, 08:28PM

Registered Member #3831 Joined: Thu Apr 14 2011, 02:54PM
Location:
Posts: 265

Yeah sorry, it's for a ZVS style charger. I went ahead and tried winding it anyway last night. I wound so the current goes clockwise if viewed from the top...sadly my mosfets self destructed as soon as I attached power to the charger, still don't know why. O.o

Forty

Tue Mar 13 2012, 09:56PM

Registered Member #3888 Joined: Sun May 15 2011, 09:50PM
Location: Erie, PA
Posts: 649

do you have a tv flyback or something that you could test the circuit on first, before you test the E core transformer?

If the circuit's fine then my only idea as to why the transformer didn't work is that the direction of the current induced in the "off" primary by the "on" primary is opposite what it would be on a non E core.

Let me try to (and hopefully correctly) explain:
With the primaries on the same side with a centertap (as usually illustrated) each coil produces a magnetic field when powered but in opposing directions. When coil A is on, it induces a current in coil B that is in the same direction that the current would be if coil B were on (since the induced field will oppose the one that created it)

With the E core, I imagine that the flux from each primary has two paths to follow (one through the secondary, and through the other primary.)

Calling the left coil A: when coil A is on, part of it's magnetic field passes through coil B (as it should) but it induces a current in coil B that is opposite the direction that the current would be if coil B were on.

Maybe that's the difference that caused the problem. To fix it, and provide better coupling to the secondary anyway, I would just wrap the center tapped primary around the secondary. It looks like you've got the winding room.

currentkills91

Wed Mar 14 2012, 03:40AM

Registered Member #3831 Joined: Thu Apr 14 2011, 02:54PM
Location:
Posts: 265

I understand how the ZVS driver works and that fancyness, I was just unsure as how I should wind the primary with it being an E core...I can test the circuit again when I get more, sadly I used the last two mosfets i had for this thing x.x

m4ge123

Wed Mar 14 2012, 03:56AM

Registered Member #4118 Joined: Mon Oct 03 2011, 04:50PM
Location: MD
Posts: 140

Wind it around the center part, not the outside ones.

currentkills91

Wed Mar 14 2012, 09:40PM

Registered Member #3831 Joined: Thu Apr 14 2011, 02:54PM
Location:
Posts: 265

I wont be able to fit the required number of turns I need if I do, I need 1.5 volts/turn so I need all 8 turns.

klugesmith

Thu Mar 15 2012, 01:34AM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

currentkills91 wrote ...
...So can someone tell me if I can wind the other primary the same way, or will I need to switch the wires around so the current flow starts at the bottom of the coil? If that made sense?...

Couple of things (no pun intended) to consider here. I'll try to add to what Forty has explained.

If the two half-primary coils are connected in series, to make a center-tapped winding, it doesn't directly matter whether their currents flow from top to bottom. What counts is whether their currents both circulate the same way (clockwise/counterclockwise, as viewed from the top).

But the three-legged core upsets the simple transformer model. In symmetrical operation, the outside legs each have only half as many volts per turn as the center leg. Does that defeat the purpose for which you want to use the winding window area twice?

[edit] Wait, it gets worse.

In a center-tapped primary sharing a leg with the secondary, or using the other leg of a UU core, the two halves aid each other. Their mutual coupling is almost +1. Put a volt on one half, the current will ramp up to induce almost 1 volt on the other half (as well as some volts on the secondary). Inductance of the entire primary is about 4x the inductance of either half. If one half were reversed, the combined inductance would be near zero.

In your outside-leg case, the flux generated by one half-primary will divide between the center leg and the far leg. K to the center winding will be significantly less than 1. K to the other half-primary winding will be much less than 1, and negative. (As Forty explained, the two halves must oppose each other, to reinforce each other in center leg. That may upset your driver circuit. )
Inductance of the entire primary is less than 2x the inductance of either half. If one half were reversed, the combined inductance would be more than 2x the individual inductance. But then a driver with Vdd on center tap would almost immediately saturate the core. Look into fuses that can protect MOSFETS.

teravolt

Thu Mar 15 2012, 03:34AM

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

on other option for winding would be to wrap the center with the primary and put your secondaries to the left and right on the out sides. if you aren't getting enough voltage at the output use a smaller secondarey wire gauge. typicaly everything is wraped around the center

klugesmith

Thu Mar 15 2012, 03:47AM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

Not a bump, a different subtopic. Practical construction.

CK apparently wants to add a 16-turn center-tapped primary, to existing core and secondary winding. Says the window area is too small for a conventional winding around the center leg.

Solution: use wire that fits better than the stuff in the OP picture. That can be done with no loss of current capacity. How 'bout winding with a bundle of several magnet-wire strands connected in parallel? See recent 4hv thread about Litz wire. Don't twist the bundle any more than you need to keep the strands close while winding.

Better yet, design a bundle which will fill less than 1/8 of the window area, and contains an even number of strands. Keep track of the two half-bundles -- you'll make a "bifilar" winding that can be connected as a center-tapped primary (tie the start of one thread to the finish of the other). Symmetry and coupling will be hard to beat. Don't depend on magnet wire varnish for insulation between primary and secondary.

Please PM if you want some suitable magnet wire to play with, and don't have a small solenoid, fan motor, stepper motor, or MWO turntable motor to unwind. What are the window dimensions?

1 2 next

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