SOLENOID-DRIVEN ROD OSCILLATOR / Electromagnetic Projectile Accelerators / Forums

Forums

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

« Previous topic | Next topic »

SOLENOID-DRIVEN ROD OSCILLATOR

Move Thread

LAN_403

Signification

Tue Dec 01 2015, 04:16AM

Registered Member #54278 Joined: Sat Jan 17 2015, 04:42AM
Location: Amite, La.
Posts: 367

I have a rather unusual requirement that is similar to a coilgun in operation. I need to rotate a metal disc (3/8" x 6" dia) by displacing a 1/4" metal rod through a coil . The rod displacement is about 5.8" before it switches off and the wheel's momentum brings it back up. The metal rod's end is connected near the disc rim by a bar (similar to a piston drive). The coil firing will energize and rotate the disc from a few degrees after TDC (top dead center--in car engine speak) to --almost BDC. then the momentum of the disc will swing things back to the starting position and repeat accelerating just a bit to two, maybe four revolutions per second, here I will keep the RPM's constant. Only short runs (three or four minutes) will be necessary, but repeatability is most important.

Since the rod will try to center in the coil, I plan on first trying with about an 18" metal 1/4" diameter rod in a guide co-linear with the disc center, with perhaps a LIPO/IGBT setup optically triggered and manually adjustable. Maybe a heavy gauge wire coil of about 2" length. The rod will never exit the coil, but only oscillate approx 6" back and forth about it's center.

******************
Any suggestions (esp) on the rod and coil will be greatly appreciated.
******************

DerAlbi

Tue Dec 01 2015, 05:39PM

Registered Member #2906 Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 727

Relatively slow movement. No pulse load. uuhm.
I would suggest you use a permanent magnet inside the coil. With that you can even determine if you push or pull. This might help with the rotation.
The slow movement implies you can tolerate slow reaction times so you can have high inductance. After a short thought this brings me to the conclusion that you might want to have a coil similar to a mechanical relay. Thin wire, high resistance, medium or low voltage, not much heat, but steady pull force.
Coil length is somewhat in the order of the radius of your disc. Not so many layers because of the probable heat. Or maybe many layers for higher thermal capacity? Basically you are restricted there due to the maximum flux the magnet can handle. You dont need much i guess. I mean.. you do not specify any pull force

LiPo and IGBT are (most of the time) an oxymoron. You wont have a voltage high enough to make IGBTs the right choise. (I assume voltage below 150V..)
Think about it: you have minutes worth of operation. Its not practical to have a 1kW+ coil due to the heat. So you have either high voltage and low current (=Mosfet) or low voltage and high current (=Mosfet). I suggest lower voltage, just because winding thicker wire is less annoying. A given coil geometry for a given pull force will have the same losses no matter what.

Signification

Tue Dec 01 2015, 08:30PM

Registered Member #54278 Joined: Sat Jan 17 2015, 04:42AM
Location: Amite, La.
Posts: 367

DerAlbi wrote ...

Relatively slow movement. No pulse load. uuhm.
I would suggest you use a permanent magnet inside the coil. With that you can even determine if you push or pull. This might help with the rotation.
ll force

I simply want to get an Al disc up to a certain speed, then hold at that rate--NOT fast. I would need switch control of the 'push' of the rod that supplies force to the disc (via a separate rod like a piston crank rod), so I don't see a permanent magnet working....how would I switch it?

BTW: this idea of driving the rod comes from one of the most impressive youtube coilguns I have seen fired used a 37V LIPO batt that could burst >500A and an IGBT. It had four stages, but only one was used--IIRC, the video title was by "IVCOILGUN" I will find it and post it here.

I don't even really have to use a coilgun drive...would like too--on e that just pushes a long rod about 6" which is approx when the rod is centered in the coil....you suggest a coil length about the radius of the disc?

EDIT: Here it is...

hen918

Tue Dec 01 2015, 10:32PM

Registered Member #11591 Joined: Wed Mar 20 2013, 08:20PM
Location: UK
Posts: 556

Signification wrote ...

maybe he would have got the projectile through *both* sides of the box if he used a nice 100V MOSFET with a snubber!

DerAlbi

Wed Dec 02 2015, 05:25AM

Registered Member #2906 Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 727

One coil can either push or pull a permanent magnet depending on polarity on the coil. As far as i understand you want to translate a linear movement into a rotary action. That will do the job just fine. You want to go for low current, many turns to keep it simple. Less turns and higher current is only needed if things go fast. If the suggested coil geometry is not correct please forgive me.. maybe i wasnt able to put the exact same thing together in my head that you think you describe precicely. Its a shame people didnt invent pictures or technical drawings yet

ooh wait.... no. hmmh.

Well.. this video.. hmmh. These guys. What shall i say.
They are a typical representation of the coilgun-community. They dont know shit about what they do but think they should teach the stuff the think they know.Thereby adding bullshit to the internet as if there wasnt enough allready. Their model is so over simplified that they come up with hillaripous numbers like 42V/100mOhm = 420A. What about the Battery ESR, what about the IGBT drop at this current? Yes, The IGBT drop at this current is significant. thats why you use Mosfets at low Votlage wherever you can. Then they put a 2.2Ohm resitor for faster current decay in series to the flyback diode. Great idea! Because..... 420A*2.2 is ~900V and that voltage is seen by the IGBT which is a 650V-Type. Great. Perfect enigneering. They really know what they are doing and what they calculate and hhooooooly fuck. Idiots. IDIOTS. I D I O T S. !!!!
And whats even more fucked up is that such guys get a kickstarter project??? WTF. I asked kickstarter to suport me but i got rejected due to the subject of the project. So coilguns are allowed.. or only the pathetic ones or only dudes that will never get anything significant done or what?
Seems legit -.- Hillarious if people support them. Its just burned money.

Anyway... dont call this video impressive. Its pathetic. It really is. And reading the comments -.- ooohh faith in humanity.. it wouldnt even work try to educate them. They obviously hunt for big numbers.... Maybe they should just multiply everything by piÂ². i mean.. why not.

Dont get me wrong here.. coilguns are a nice beginners project and if it inspires people to learn stuff its really great. But with such sources at hand people get poisoned with wrong thoughts and thought-processes by people who just rest on top of mount stupid. And the beginners simply cant decide if its crap or not - its not even their fault. I wish so much that unscientific bullshit like these videos could be cleaned up. Like forums with plain wrong comments ranked high in google. People should be more responsible for that they are posting. Videos like this are a crime against humanity. I have no other words for that.
Btw: has anyone any idea why coilgun-videos are described by the stored electrical energy? Its not like you couldnt add more Caps by simply spending more money.

Sry. Back to topic.

Signification

Thu Dec 03 2015, 02:13PM

Registered Member #54278 Joined: Sat Jan 17 2015, 04:42AM
Location: Amite, La.
Posts: 367

OK, I see the -bad- 900V which could hit the IGBT. Maybe the MOSFET will do it. Can anyone tell me basically how the 'snubber' caps help.
Since I am simulating automotive internal combustion stroke action, and assume I don't need any pullback--except to re-center the rod for the next 'push', also since the disc is well balanced with ball bearings, I don't think it will take much time-and-effort to get it to a decent -demo- speed. Even though all 'drive' is planned out for the coil and long rod, I see that I may not even need this method--but I do think it will be employed--too much planning and $ at this point for this big of a change now.

I noticed they have more videos--but I listen, and can easily assume that all this posted info is incorrect (along with the response msgs). it just --sounds-- good and is very easy to accept and believe--I can do the 180, and correctly in time. I would like to get away from the LIPO batt + charger--have you all seen the prices here? SO...any suggestions...greatly appreciated.
ps
I ended up with a box of eight (8) block Darlingtons (100A @ 600V) that were advertised as IGBT's--"as is" *then* stated.

DerAlbi

Thu Dec 03 2015, 03:17PM

Registered Member #2906 Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 727

Ok ok. Then no permanent magnet if you dont need the both directions

I think in your applications LiPos are fine. You wont be able to do maintain the power otherwise except a big block of SMPS-Magic. Using capciors wont store enough energy (since you need it for a long time)
So... How to chose a coil?
1) Geometry
1.1) Length: the coil must be as long as the path you want the pullforce to act. (maybe a bit longer.. you dont hunt for max. efficiency here i guess)
1.2) Inner radius: the larger the coil the more force there can be before the iron saturates. Larger inner diameter makes the projectile havier. But you have no coilgun, so just use a really fat rod. If its 2cm in diameter who cares. (if there is any unknown boundary condition then make it smaller but as thick as you can)
1.3) Outer Radius: try so stay reasonable. You dont need high efficiency... so i would suggest not so many layers. maybe half the inner radius worth of copper thickness? maybe less, maybe more. (Meaning outer radius = 2x inner radius, contrary to what one says in coilguns where Outer radius = 3x inner radius.)

When you determined the geometry (that you like) then you
2) choose the wire.
2.1) a given coil geometry will produce a given heating for a given force. thats basically constant. (for a given conductor material) Accept that first.
2.2) Choose the inductance: inductance is what slows the current rise down. The faster the current must rise, the lower the inductance. Since you are reletively "ultra low speed" you will aim for a really high inductance.
2.3) Consequences: the high inductance is only achieved by many turns, which is achieved by many turns within your geometry meaning verry thin wire (maybe 0,3mm?), thereby increasing the resistance and lowering the maximum achievable current.
2.4) Dont care! If you had have choosen a thicker wire you had less turns (in a given coil volume), less inductance, less resistance... Yes. But: Force = (much more)Amps * (much less)Turns then. See the (much xxx) cancels out.
2.5) Achievements? You minimized your current by accepting the high inductance. Its now quite the same like a relay-coil for example.. it still works in 10ms, but it does not consume much current.

3) Choose the voltage for your coil: Due to the low speed, the time constant (L/R) wont matter much (and is constant for your chosen coil geometry no matter which wire you have chosen). Your maximum current will be simply Voltage/Resistance. This Current must be not much higher than whats nessecary to saturate your iron.

This setup now puts the mimum strain on your switching devices and power source. I hope thats in your interest.
Because: you can use the chapest crap as switching device that you can find. I figure you wont need much more than 5A at 20V. Maybe its even more relaxed. But that would be allready 100W worth of heat. It depends on what power you actually need.
Higher power would be achieved by bigger iron and a bigger coil with bigger wire. (the coil geometry and core just scales up) This enables you higher current, therefore higher Amps*Turns.

It might still be possible to use a permanent magnet. IF (i dont know if you do) you achieve the same pull force with less current then you have gained quite a lot. The coil will get HOT in time. Lowering the current to 70% will half the heat output as you might know allready.

PS: dont care about numbers i gave here, they are based on nothing.

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