Beginning Coilgun design
CliffB, Thu Mar 29 2007, 04:16AMI'm trying to build a coilgun and have been working on the basic parameters for the design. I was looking at using large capacitance and low voltage then headed over to barry's RLC simulator and saw that high capacitances give really long discharge times.
I was trying to decide between two setups for the capacitor, either using some combination of 1500uf, 350v caps or a combination of 3,399uf 250v caps. Both give near the same current peak through the inductor, but the 350(which I ran the simulation for at 280V, you'll see why) discharged fairly significantly faster and had the higher peak. I was wondering what reasonable values for l and r would be, I need to dig out a physics book and do some work on that, but at the moment I am more concerned with the charging circuit. I'm also interested in finding a way to approximate how long a projectile will be in the coil(still have a bit of reading to do).
What I was thinking about doing for charging the capacitors was getting a variac:
I like that one cause it appears to be reasonably priced and goes up to potential differences of 280v, wheras most, on ebay at least, seem to only go to 140ish.
I would then feed this through a rectifier, I was looking at:
I understand the general idea of the rectifier, but not the specific implementation, so I don't know if that would work.
I suppose I also need a switch for this circuit, is there any specific requirements for this? I know the other circuit needs a beefy switch, but for charging I think I'd be okay.
I'd like to use the variac for charging cause I want to be able to use it for other projects. I am new to this so if I am completely off I would be very happy to know so. Just did a calculation for final velocity by doing energy=1/2cv^2 then assuming, say 1% efficiency, and setting that equal to 1/2mu^2 and got a result of 1.084 for u. So: .01(.5*.0015*280^2)=.5(1g)u^2. u=1.08 I think this would be meters per second, which seems to translate to 2.43. This seems a bit low, anyone know what I'm doing wrong?
So I've done some more research and using the same basic design as is shown at:
I'd be building something with the parts list as follows(labeled relative to the diagram):
B1 - 9v transistor battery
C1 - 1500 uF 300wvdc, electrolytic capacitor
D1 - 400v bridge rectifier (linked in above post)
L1 - launching coil
P1 - iron projectile
R1 - 500 ohm 1/2 watt, charging resistor
R2 - 11K ohm 1/4 watt, safety bleeder resistor
R3 - 4.7K ohm 1/4 watt, trigger resistor
S1 - 600v 20A SCR, "silicon power cube"
SW1 - momentary action pushbutton switch
T1 - transformer, 110v primary, 0-280v secondary
I've not changed the charging resistor cause It was about the value for resistance i figured I needed, calculated based on a page somewhere on his site.
For a coil I'm looking at 10mm interior diameter, and between 75 and 115 turns of 12 AWG magnet wire. With the lower value i have a resistance of .024ohms inductance of .034mh , with a .75ms discharge pulse(from the RLC circuit modeler) peaking at 1622 amps.
This pulse seemed short, so I'm thinking about going up to 115 turns as well, but tapping it out at several places along the coil to allow for tuning. It seems that the magnetic field strength at where our particle is will not increase that much from the extra turns, and the increase in resistance and inductance may even cancel that out.
The main problems I have right now are that I don't think the SCR he uses would be able to handle the current that will be being delivered to the coil.
I am also not sure about the charger resistance value, or if the rectifier I linked is an appropriate part for what I intended to use it for.
Any help would be greatly appreciated.
[Edit: Fixed double post]
Re: Beginning Coilgun design
..., Fri Mar 30 2007, 03:44AM
Usually you want to use a projectile as long as the coil. A reasonable coil length is .5 to 1" (shorter coil = smaller projectile = faster moving, but higher change of suckback = lower eficiency)
I am a little sceptical about that variac, as its entry in the catalog only shows it as a 280v module (240v in 280v max out), not a 140v in 280v out one. Might check with the seller...
As to the caps, generally with a coilgun more voltage is better, as if the pulse is too long you get 'suckback' which kills efficiency.
As to the diode, you best best is the good 'ole 1n4007 diode. If you are in the US I could ship you en envelope of 20 of them for free.
I would use a normal 140v variac, and then a full wave voltage doubler which will give you a max voltage of about 400v.
If you use a variac, you do not need a charging resistor, just turn the variac on at 0v, and then turn it up (with that size of a cap bank you can't turn it fast enough to hurt anything).
As to the switch... You can use a big fat realy, although you will loose a small amount of efficiency. A SCR good for 2ka peak isn't all that hard to come by either, and you can parallel several SCRs (generally it is safe to assume the max current is the peak current rating of the scr time the number of SCR's, as the peak current is measured for a 20ms pulse, and you have less then a 1ms pulse.
..., Fri Mar 30 2007, 03:44AM
Usually you want to use a projectile as long as the coil. A reasonable coil length is .5 to 1" (shorter coil = smaller projectile = faster moving, but higher change of suckback = lower eficiency)
I am a little sceptical about that variac, as its entry in the catalog only shows it as a 280v module (240v in 280v max out), not a 140v in 280v out one. Might check with the seller...
As to the caps, generally with a coilgun more voltage is better, as if the pulse is too long you get 'suckback' which kills efficiency.
As to the diode, you best best is the good 'ole 1n4007 diode. If you are in the US I could ship you en envelope of 20 of them for free.
I would use a normal 140v variac, and then a full wave voltage doubler
which will give you a max voltage of about 400v. If you use a variac, you do not need a charging resistor, just turn the variac on at 0v, and then turn it up (with that size of a cap bank you can't turn it fast enough to hurt anything).
As to the switch... You can use a big fat realy, although you will loose a small amount of efficiency. A SCR good for 2ka peak isn't all that hard to come by either, and you can parallel several SCRs (generally it is safe to assume the max current is the peak current rating of the scr time the number of SCR's, as the peak current is measured for a 20ms pulse, and you have less then a 1ms pulse.
Re: Beginning Coilgun design
CliffB, Fri Mar 30 2007, 07:32AM
I've purchased said variac, unfortunately, I suppose did a bit more research into it and it appears it does take 120v input just above 150v out drops the current down. The resistor you mention, could be used to make a bridge rectifier for this though? I sent you a pm reguarding the SCRs.
Thanks for the response.
CliffB, Fri Mar 30 2007, 07:32AM
I've purchased said variac, unfortunately, I suppose did a bit more research into it and it appears it does take 120v input just above 150v out drops the current down. The resistor you mention, could be used to make a bridge rectifier for this though? I sent you a pm reguarding the SCRs.
Thanks for the response.
Print this page