Why Metal Spinning?

I have a lot of trouble starting the blog for this project, because I have a million thoughts about it and I haven't known whetner the try and express them all in a shorter form and follow up with further detail, or some other approach. What I am actually going to do, is address one small thing, and see where it takes us.

If I were looking at the Facebook and Twitter feeds, one of the biggest questions I would have is, "why is he focusing on metal spinning?" It turns out I have reasons!

It comes down to the mechanics of how microwaves are reflected in a resonance chamber. The photon hits the walls and induces an electric current in the walls and this electric current in turn emits a new photon that follows a course as if it reflected off the wall. Which it did; that's how reflection works. Anyway, the point is that without the electric current, it doesn't work. So, the material needs to be highly conductive, and the current has to be complete, or nothing happens. And there's another piece of the puzzle: skin depth. Skin depth is how far into the copper the photon penetrates before it creates the current. A high frequency photon will go farther in, a low frequency photon will stop closer to the surface. So if the photon doesn't penetrate all the to some copper, it doesn't work. Say, if there's copper oxide or some schmutz on the wall. And the skin depth is micrometers thick; the copper can't have anything on it. And it should be very highly polished, forcing the current to flow around surface blemishes, like the tiny scratches the keep metal surfaces from having a "mirror finish", increases the electrical resistance and cuts down the reflectivity. This, in turn, reduces the "Q factor", which I will explain later and is really, really important for the EM drive.

Which brings us to metal spinning. My first attempt at building a cavity, in addition to having nearly every dimension wrong, was made by cutting the shape out of a single sheet of copper, wrapping it around a mandrel, and welding the ends together so that it was in the shape of a frustum.


Which went OK except that it wasn't very smooth, and I didn't make my own mandrel so it wasn't shaped smoothly, and a million other reasons I am unhappy with it. But I had my biggest problem with the seam. It really bothered me to have a seam, because:

  1. I expect it would interfere with the electrical connectivity, since the solder probably conducts less well than the copper,
  2. I expect the huge step between the two ends as they had to overlap to get the weld to stick to be a problem with the skin depth,
  3. It just didn't seem right.

I am also looking at other designs where other groups have created their walls and then bolted or affixed the base plates to flanges on the walls. A mechanical connection like that does not work as well as a solder connection, just because it often better to have a liquid that can flow into the little cracks between two objects. Now, I am not going to knock Anyone else; many of them know what they are doing better than I do; I am just saying that I cannot do things their way and feel comfortable that it is the best I can do.

Then, I heard about metal spinning. It is a technique that is not often done today, and actually uses a lot of tools in very unusually way and, in a space like the Artisan's Asylum, is very cross-disciplinary and complicated. We do not have any experience with it amongst our instructors, and just having some one trying it at the Asylum has generated a lot of interest. But, what it means, is: I can create a frustum whose walls and one of the base plates are all one single sheet of metal and there virtually no seams anywhere. This is very important to me; I think this will increase the Q of the chamber, and it may make fabricating multiple copies easier, if it comes to that.

Som, months of yak shaving later and I have started to produce some spun parts.
Metal Spinning Night 1 Metal Spinning Night 2
Nothing that looks completed, but I am almost there.