maximizing mazilli charging power / Electromagnetic Projectile Accelerators / Forums

Forums

4hv.org :: Forums :: Electromagnetic Projectile Accelerators

« Previous topic | Next topic »

maximizing mazilli charging power

Move Thread

LAN_403

Forty

Sun Jun 05 2011, 08:35PM

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

I'd like to get some ideas stirring that lead to the optimal design of uzzor's zvs capacitor charger that allows for the fastest and most efficient charge time of a capacitor bank for a given input.
a few questions about the transformer:

-core size: i know larger flyback cores are capable of more power, but for an input power less than a few hundred watts (12v batteries), is bigger still better? it seems to me that a more compact core would have less losses and greater energy transfer because the flux path would be shorter, and the winding window more full.

-air gap: should there be an air gap for a capacitor charging flyback core? if yes, would a larger or smaller gap be better?

-coil placement: would it be best to have the primary and secondary be separate and opposite of each other on the core or one within the other on one side? for the latter case, which should be on the inside, primary or secondary?

-wire size and type: the primary should obviously be a large diameter wire, but what would be best: solid core, stranded, or magnet wire litz? for the secondary, should the largest diameter of wire be used that still allows the coil to fit within the window, or would the greater distance from the core on the outer layers decrease performance? we usually try to get away with using the least wraps of wire possible for each side by using primaries with 5-10 + 5-10 turns, but would a greater number of turns per side (maintaining the desired ratio) increase energy transfer?

-should i design the transformer to output exactly the voltage i wish to charge to, or should i have it output a larger voltage and then have a regulating circuit to cease charging at the desired voltage?

and about the mosfets used:

-i originally thought that using the beefiest fets i could find would be best ( currently working with FCA47N60F 600v, 47A, 62 mOhm rds on from fairchild) but then i read more into the operation of the circuit and found that the voltage on the fets rises until they reach their breakdown voltage and then avalanche through the coils. so would i be better off with a lower voltage mosfet?

Ash Small

Mon Jun 06 2011, 12:12AM

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

Forty wrote ...

.-air gap: should there be an air gap for a capacitor charging flyback core? if yes, would a larger or smaller gap be better?

-coil placement: would it be best to have the primary and secondary be separate and opposite of each other on the core or one within the other on one side? for the latter case, which should be on the inside, primary or secondary?

I can't answer your other questions, but I read a paper recently that stated that the primary should always be wound under the secondary if possible. Also that if using a gapped core, one gap is better than two.

It also stated on which leg to place the windings (gapped side or ungapped side) but I can't remember offhand which side it said. Something to do with flux leakage, if I remember correctly.

I'll try and find the paper.

Daedronus

Mon Jun 06 2011, 09:40AM

Registered Member #2329 Joined: Tue Sept 01 2009, 08:25AM
Location:
Posts: 370

mazilli driver is a push pull driver, energy is transferred in each half cycle, so there should be no air gap. (it is used in flyback mode, where it's storing energy in the first half of the cycle)

The wire should be litz wire, where the individual strands have a diameter in the order of the skin depth at the particular working frequency.

Pri and Sec should definitely be on top of each other, for maximum coupling.

The mosfets should have the lowest Rds on you can find, while still being rater above the working voltage, with some safety margin
But, don't use 600v mosfets for 12V :)
they will be crap at 12v.
Look for 40, 60 or 100V mosfets.
Like IRFP2907PBF

Last thing, mosfets should definitely not avalanche in push pull operation, since there is no back emf.

Forty

Sun Jun 19 2011, 09:08PM

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

alright, i made some litz cable to use for my primaries, 30 awg, about 70 strands.
I've got some more suitable (hopefully) mosfets coming from fairchild in a day or two.
i could do litz wire with a few strands of 30awg for the secondary too, would that be worth it?
also, another recent thought: should the bridge rectifier on the secondary be made of individual high speed diodes like uf1007 or could a bridge rectifier chip (have several BU2510) be used. I just wonder because these chips are most likely designed for 60hz operation and the frequency on the secondary will be the same as that on the primary: several khz.

any input on my original core size question would be helpful too

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