New to Coilguns - Coilgun Buildlog (Charger apparently fixed) / Electromagnetic Projectile Accelerators / Forums

Forums

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

« Previous topic | Next topic »

New to Coilguns - Coilgun Buildlog (Charger apparently fixed)

first previous 16 17 18 19 next

Move Thread

LAN_403

Artikbot

Mon Jun 20 2011, 09:17AM

Registered Member #3247 Joined: Mon Sept 27 2010, 09:42AM
Location: Spain
Posts: 137

Yep, saw it lookin at the datasheet.

Guess I'll have to route some cables to place it somewhere above the 555, or dunno. Maybe it fits under the board.

Going back to the 555 not charging the gate properly... In what sense?

Seemingly I output 160mA to the gate, at 11.1V. That is a lot of power to the gate, unless the square waveform isn't really square and it doesn't rise on time.

Mind explaning that a bit more please?

Thanks!

Adam Munich

Mon Jun 20 2011, 03:51PM

Registered Member #2893 Joined: Tue Jun 01 2010, 09:25PM
Location: Cali-forn. i. a.
Posts: 2242

Thing is you want to charge that gate capacitor as fast as possible, and at 30, 40, 50kHz you need to supply big currents to charge it up fast enough.

I'd get some sort of dual low side gate drive IC to run it. Here is a 9 amp one that you could sample,

Get the SOIC version because the MLP is unsolderable. Also, get more then one because it's easy to kill gate drivers if you mess up.

Artikbot

Tue Aug 09 2011, 09:05AM

Registered Member #3247 Joined: Mon Sept 27 2010, 09:42AM
Location: Spain
Posts: 137

Hey Grenadier!

Sorry for the delay in the response, I ran into the exam period and set aside the project again.

I see, so a driver is the way to go. I'm not running at insane frequencies, but well, I guess 160mA is a crap of current if it needs to charge the gate capacitor in 1/14000 of a second. Even if it has a couple of pF, that is too little of a current anyways.

I'll do a daughterboard and place it on top of the ICs I have, so I can add the driver to the circuitry without increasing the board space.

Will see if my local shop can supply me those, otherwise I'll need to pay 10 times the cost in terms of shipping... But I'm used to that

Will report back soon, promise!

Thanks!!

Forty

Tue Aug 09 2011, 05:55PM

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

if you can't get the driver ic, you can probably get by with a couple transistors such as in this circuit:

since that is basically what the driver does anyways, and you're not at any ridiculous frequencies.

I wouldn't worry about that spike that shows up on your meter as long as your charger shuts off when it is supposed to.

is your optic triggering system still working fine? i've got a circuit for dual triggers if you need it.

Artikbot

Fri Sept 02 2011, 05:59PM

Registered Member #3247 Joined: Mon Sept 27 2010, 09:42AM
Location: Spain
Posts: 137

Wow, it's been long since I last posted, once again.

Crazy summer, I got into model car racing and the coilgun was stored in the closet.

I have new plans and new stuff too.

So, the two transistors (NPN-PNP pair) are able to drive the MOSFET at 7kHz (IIRC this is the circuit's oscillation speed)? Off to try it, and pray for the mosfet to not be fried.

@Forty: The trigger works like a charm, even tested it with the SCRs and it works great. Tre triac is bugging me a bit more, but nothing serious. Thanks anyways :D

My new plans are:

1) Fix the charger up.
2) Finish the coilgun body.
3) Test and tune it.

Once all this is done, this will be the new roadmap:

1) Build a Mazilli ZVS driver
2) Get a new capbank, around the kilojoule.
3) Build a 3-stage (probably 5 stages, depends on the budget) cannon, for 12mm caliber projectiles. This one will use strawed projectiles to reduce eddy losses and stabilize the round. I might use silicon insulated wire for this purpose, as enameled wire is deadly expensive in high diameters. Or proabaly a double-wound coil. Yet to be decided which material will I use.
Plus I need a winder to wind a 12AWG coil, by hand it's very, very hard to do it right, tight and nice.

And as an ultimate stage (now it's just a couple drawings on a blank sheet), mount the 4mm cannon (the current one) into an arm-wielded module. That is, have the coilgun hanging from the arm, with the trigger in the hand. But as I said, this is nothing but a couple concept drawings.

Forty

Tue Sept 06 2011, 03:29PM

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

a double wound or thinner gauge coil might actually be better than the 12awg since the turns could be placed closer to the barrel. if you plan to divided the 1kj over multiple stages, then smaller wire like 16awg should work just fine (and it's not hard to wind by hand and can be found in some older mot primaries.)

Artikbot

Tue Sept 20 2011, 08:07PM

Registered Member #3247 Joined: Mon Sept 27 2010, 09:42AM
Location: Spain
Posts: 137

It makes a lot of sense, now that you say it. I'll probably wind each other layer with alternative wires, that is, first layer wire 1, second layer wire 2, third layer wire 1, fourth layer wire 2, and so on.

Have found a rather cheap supply of enamel coated wire, prices starting at 20â‚¬ per 500g spool, approximately. 12AWG wire.

But for now, I'll just fix the charger, tune it for lower voltages (maybe tuned down at 200V just to do rutinary firing tests that do not requiere the coilgun to fire at full power) and build a 100W or so Mazzilli ZVS. Still in process of findinga decent transformer core, as I do not want to damage my beloved flyback pulled out of an old Trinitron CRT, which is very well built and has a very decent connector system installed.

Been thinking about possible causes of failure on the bare boost converter (no drivers used), and my sturdiest hypothesis is that once voltage built up, the charger maxed up its power and heat started to build in a rather heavy way, causing the MOSFET failure. I have yet to build the FET driver (ah the university, takes sooo much time).

By the way, how is current regulation undergone in a ZVS driver? Can't happen to find documents about this. Besides inductance on the transformer core, I cannot happen to deduce anything else.

Forty

Tue Sept 20 2011, 08:28PM

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

I think the current in the zvs is limited only by the inductance and the melting point of the components.

Artikbot

Tue Sept 20 2011, 08:47PM

Registered Member #3247 Joined: Mon Sept 27 2010, 09:42AM
Location: Spain
Posts: 137

Hahaha then I'd rather find out which inductance I need, melting point is something I'd rather keep ignoring

Forty

Wed Sept 21 2011, 01:17AM

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

well any power source you use to supply it is going to have some inherent current limit. so as long as your mosfets can switch the current, and everything else can dissipate enough heat (inductor, primary, resistors, wires) then the zvs will draw all the current it can.

first previous 16 17 18 19 next

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