CoilGun 'Coil' question / Electromagnetic Projectile Accelerators / Forums

Forums

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

« Previous topic | Next topic »

CoilGun 'Coil' question

1 2 next

Move Thread

LAN_403

mmt

Mon Mar 01 2010, 07:51PM

Registered Member #2333 Joined: Thu Sept 03 2009, 03:04PM
Location: Slovakia
Posts: 80

Hi,
ok i'm new, in coilguns, and i wanna make one soon, I'have been searching for this information on internt, and also here, on forum.. I dunno how many turns of wire should i wind for my coil for coilgun. Is there any up an down limit for coil turns, and also how many layers, thickness, and insulation of wire? Or is there any formula for these things?

my coilgun will have: 20 000uF/400V cap
charge votlage: 325VDC
switching by massive Thyristor: 1600V/160A

TechNerd

Mon Mar 01 2010, 08:06PM

Registered Member #2289 Joined: Thu Aug 13 2009, 02:49PM
Location: Jylland, Denmark
Posts: 19

Hi

have you checked this site http://www.coilgun.info/about/home.htm
I found it very use full

Barry

Tue Mar 02 2010, 01:08AM

Registered Member #90 Joined: Thu Feb 09 2006, 02:44PM
Location: Seattle, Washington
Posts: 301

You want the 'on' time to be just long enough to pull the projectile to the center of the coil, but no longer or else it will experience the Dreaded Suck-back Effect. If your coilgun is gated by an SCR (as most of them are) then the timing is a direct result of the LC time constant.

Since the LC time constant is so important, you want to design the coil's inductance L to yield the desired firing time for your capacitance C.

Try this LC timing calculator to help choose the inductance. My experience has been that stage 1 timing usually ends up in the range of 2 - 8 milliseconds, but it depends on many factors.

There are lots of different coil designs (turns, wire size, length, layers, etc) that end up with the same inductance. But it's easy to narrow the choices.

Coilguns work best with low winding resistance, so choose the thickest wire you can manage.

Coilguns are most efficient when the coil length is about the same as the projectile length, or maybe 10-20% longer than the projectile, so as soon as you choose what you're going to shoot then you know the coil length.

The coil's inside diameter must be wide enough to accept the projectile size and the firing tube's thickness.

Excess insulation on the wire is basically wasted volume, so we usually use "magnet wire" in which the insulation is a coat of varnish. Besides, it's cheap. But virtually any kind of wire will work.

Now you've got most of what you need to finish designing the coil. Try this inductor simulator and play with the sliders until you get the inductance you want.

Finally, build your coil with a few extra turns. It's easier to remove turns from the coil later on, than to add them. Note: Inductance is turns-squared, and 10% extra turns results in 20% more inductance, so don't get too carried away with adding extra. Just be sure to have some.

Have fun, Barry
"The majority of quotes on facebook are misattributed to famous people." ~ Abraham Lincoln

mmt

Tue Mar 02 2010, 03:13PM

Registered Member #2333 Joined: Thu Sept 03 2009, 03:04PM
Location: Slovakia
Posts: 80

Thanks for your advises Barry! u solwed, lot of my qestions.
In here

I choose capacity 20 000uF. but im kinda confused witch definition of period should i choose, and also the time u said: 2-8mili sec, but thats a quite big induction differance between 2 and 8 mili sec. so this is my new problem, can u help me?
this inductor sim

look really great! im lookin forward to finally design my coil, when i will know that induction.
thx
-Martin

Barry

Tue Mar 02 2010, 09:42PM

Registered Member #90 Joined: Thu Feb 09 2006, 02:44PM
Location: Seattle, Washington
Posts: 301

Use the setting for "Time for Zero Crossing Pulse". In my experience it yields the closest answers to the right value.

The "time for one full cycle" is only there for completeness and other people not who are designing something other than coilguns. The "time for half power pulse" option was something I investigated on the assumption that the initial leading and trailing edges which represent low current were not going to be important. But it turned out that the whole waveform will definitely affect the projectile.

To estimate the necessary pulse time, we can make a rough guess by assuming constant acceleration just so that we can use basic physics equations. Here's a page about estimating timing. The end result: t = 2d/v

And then your next question will be "what velocity V can I assume?" Well, here's another web page that uses the stored potential energy, the projectile mass, and the assumed efficiency: Estimating Kinetic Energy.

Hope it helps!
Cheers, Barry
A dyslexic man walked into a bra.

mmt

Sat Mar 06 2010, 03:17PM

Registered Member #2333 Joined: Thu Sept 03 2009, 03:04PM
Location: Slovakia
Posts: 80

again, there is another question

for calculation v(speed) i need to know 'e' efficiency and e=KE/PE ,but for know KE i need 'v' again , witch is a circiut error :D , can u help me again?, or here is my specs, as i am soo stupid

,i cant caculate these things myself:
V=315V
C=0,019F
m=0,002kg

Thanks!
-Martin

Turkey9

Sat Mar 06 2010, 09:19PM

Registered Member #1451 Joined: Wed Apr 23 2008, 03:48AM
Location: Boulder, Co
Posts: 661

it's going to be much easier to measure speed, and I think it's the only way. There are many cheap ways to do it:

Shoot horizontally and measure distance before it hits the ground. Use the position function to find how long it was in flight (same as how long it takes to fall from your barrel height to ground). Then use the good 'ol distance/time to get your v.

Shoot at a board at a known distance and record the sound of it firing and hitting on audacity. Then use the spikes to find your time.

And of course there's always the chronograph!

Barry

Sat Mar 06 2010, 10:25PM

Registered Member #90 Joined: Thu Feb 09 2006, 02:44PM
Location: Seattle, Washington
Posts: 301

mmt wrote ...

Again, here is another question

for calculation v(speed) i need to know 'e' efficiency and e=KE/PE, but to know KE I need 'v' again , witch is a circuit error :D Can u help me again?, or here is my specs, as I am so stupid

,i cant caculate these things myself:
V=315V
C=0,019F
m=0,002kg

Thanks!
-Martin

It's hard to know where to start, since it seems like everything depends on everything else. So to break the loop, we assume something for efficiency, then design and build the dang thing.

So we generally assume efficiency in the range of 0.1% - 1% for a single-stage coilgun and work the numbers from there. Here is an example using your own numbers:

PE = 1/2 C V[sup]2[/sup] = (0.5) (0.019) (315)[sup]2[/sup] = 939 joules

KE = e PE
At the lowest efficiency KE = (0.1%) (939) = 0.939 joules
At the highest efficiency KE = (1%) (939) = 9.39 joules

Solving KE=1/2 m v[sup]2[/sup] for velocity gives us velocity v = sqrt( 2KE/m )
At the lowest efficiency v = sqrt( 2 (0.939)/(0.002) ) = 30.6 m/s
At the highest efficiency v = sqrt( 2 (9.39)/(0.002) ) = 96.9 m/s

From this velocity range you can estimate the required timing, as outlined in previous notes, which subsequently leads to an estimate of the required inductance. And this will give you enough information to design the coil.

However, imho it will take considerable tuning and adjustment to get a 2-gram projectile moving at 30+ m/s. I'm not sure a single-stage coilgun can put that much kinetic energy into such a small projectile; it doesn't have much mass and will most likely saturate magnetically before it reaches those speeds. You should probably consider a heavier projectile (larger cross section) if you want to obtain an efficiency of 0.1% or better. It would move at lower velocity but carry KE and therefore more "punch". Just an opinion. This is why we have to build it and see.

Cheers, Barry
If a beer taste-tester was addicted to his job, would you call him a workaholic or an alcoholic?

mmt

Sun Mar 14 2010, 11:47AM

Registered Member #2333 Joined: Thu Sept 03 2009, 03:04PM
Location: Slovakia
Posts: 80

thx, Barry.
I've made some coils and tested it, and seems i have found the right one,
now i am searching for some LV portable capacitor charger, photoflash charger is very weak and slow..

anyone has some scheme for quite powerfull coilgun capacitor charger?

Turkey9

Tue Mar 16 2010, 12:15AM

Registered Member #1451 Joined: Wed Apr 23 2008, 03:48AM
Location: Boulder, Co
Posts: 661

Rewind a flyback secondary to give off about 400 volts and run it off a ZVS driver. Works great and a lot of power. Only takes a few layers the length of the side of the core to get to 600v in my experience.

1 2 next

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