Sunday, March 27, 2011

BTW: New hot look

The machine is much more modular now. It has to be given how often it burns itself out for various reasons. Notice that the firing circuits (the boards nearest the camera with yellow and black wires coming out are now set into packs of 4 circuits (instead of a single board with all of them). Similarly, the firing computer is more detachable and can be swapped out in a few minutes if it ever burns. A 3d-printed bracket holds it in place. Finally, notice the new shielded sensors from all the gates. This helps get the signals back in the right order even when the EM noise from firing is intense.

More penetration testing

The timing is getting really close an questionable. In the meantime, this is the results of a 62m/s firing. It's running on 7 magnets.

It's going up against maybe 0.8mm of steel. I think it'd get penetration with a more pointed munition. I'll make more ammo later this week.

Monday, March 21, 2011

What the hell is going on here anyways?

What the hell exactly is going on here? I’ve talked about the machine I have for some time now but not fully explained what it is exactly that’s going on and why it’s such a bitch.

For starters, let’s recap how they work: For a given accelerator magnet, there is a wire wound many times around the barrel. When an electric current is passed through that wire, it is literally electrons circling the barrel. By the right hand rule, when an electron is moving, it creates a magnetic field. Because the wire is wound around the barrel, it creates that strong magnetic field inside the barrel itself.
If there happens to be a piece of iron in the barrel, the atoms within the iron will align themselves with the magnetic field and strengthen it. Now the wire and the iron are both acting as magnets and they will attract each other.

In an accelerator, the machine will turn on power to a given coil when the projectile is near that coil. It will pull the projectile into its center. Once the projectile reaches that center, it will turn off and allow the projectile to continue onwards.

Many of these magnets can be lined up where each one imparts a certain amount of energy on the projectile. Since there is a set force between the coil and the projectile and the projectile travels a set distance to reach the center of the coil, the work done (W = F*D) is a constant. Thus, every subsequent accelerating magnet imparts the same amount of energy on the projectile as did the first regardless of how fast the projectile is moving.

Minor explosion causes significant rework

You may have noticed that I've not posted for a while. Aside from an actual job that pays, I had a significant issue with the machine. I'd post pictures of what I'm about to describe but I'm not at my house and don't feel like waiting. Instead, here's picks of the chick working on... most just looking crazy while she cuts the mountings for her 3d light cube.

The explosion on the machine was more of a loud pop and some sparks. Never a good sign. In the end, it looks like one of the coils was touching the metal casing. I burned through it's enamel and took the voltage on the entire case up to 400V. When a wire from the optical sensors also touched the case, it fried about every piece of circuity in the device. Ever firing circuit, the computer, all the optical sensors. The only thing that survived was the charger (which is invincible as far as I can tell).

After a total catastrophe like that, one might as well rework the entire thing. I drew up new printed firing circuits. I made a new mount for the computer, I changed all but the very first accelerator magnet to be computer controlled, and finally I'm using shielded wire for all the sensors. Never again will I worry about EM noise, I'm shielding fucking everything.

As of yesterday, I finally have 4 accelerator gates working again. The machine was back up to 51m/s. This is about on-par with where we used to be. And of course now that I've enabled more control over certain parts and added EM protection, I think it'll be far more sustainable to expand.