happyjack27 wrote:i was going to say that part of the "spiraling" i mentioned is due to the electrons repelling each other - but my electron guns in my sims were firing at a very high rate, and this effect would not be nearly as pronounced at a low rate - and esp. if they travel practically single-file through the cusps.
but really once they leave a "magnetic field line" that goes onto the surface of the wb, they don't go onto its surface. and there is exactly one "magnetic field line" that goes onto the surface. so they're always going to have a small energy gap they need to jump to get in. whether that's done through scattering or gyration or both.
i suppose they can enter the wb with a little excess inertia, it should get scattered out fairly quickly.
still, the way i see it most of those injected electrons are going to spend some time "recirculating" before they make it into the wb.
Sorry, what you are saying here makes no sense to me. As I asked before, please provide a link to any of your sim that you think shows a wiffleball effect.
My reason for asking is that it may be that you have a complete misunderstanding of the concept. Or I do.
I am hoping that an example will be worth 10000 words.
KitemanSA wrote:Finally found time to review these. Still don't see a wiffleball formed.
Just a silly question here, Could you do a quick doodle in mspaint or something to show us what you think should have appeared in the youtube video?
Just stick it in http://imgur.com or something.
As for the linked graphic, yes, except I have always pictured the wiffleball bigger. The graphic seems to show what will happen at β<<1. As β -> 1, the region should expand quite a bit, no?
Oh, and the humps will get smaller and pointier relative to the rest of the surface area as the cusps get pressed toward closed. IIUC
The third link (cuboctohedron egun view) shows the wiffleball starting around the 2 minute mark of the video, when current is changed from a positive to a negative value.
Using the method of images the wb can be made any size less than the magrid. Once u can get peak fusion velocity, fusion is proportional to volume times density squared of fusion region. That determines ideal wb size. Except for grid losses, in which case i would think smaller is better.
Impedance of electron injection is a reasonable concern. After all, cusp confinement is a two way street. Bouncing away from the cusp,or mirroring follows rules based on B field strength and approach angle. The difference is that from the outside the mirroring is different because the B field lines are at different shape. The 'funnel' is longer, with less acute angle changes. This may effect the mirroring angle (?). As a worse case I could see the resistance to cusp passage (external containment if you will) may be similar to what is described as cusp confinement, which is ~ 60 passes before escape (from both sides of the cusp). This means you need to inject ~ 60 electrons to get one through the cusp and into the interior. This means that you need ~ 60 times the current to maintain a given internal density. This requires much stronger e-guns,but not necessarily more energy. At least theoretically the electrons that reverse before penetrating the cusp give up their KE to the potential well, which for external electrons is directed towards the magrid. The electron injection may need to be more robust, but the energy balance may not be changed, so the Q considerations may not change. The rejected electrons ground with only low KE's
Because of the different morphology, e-gun distance , degree of collimation, etc. there may be significant modification to the picture.
As mentioned above, confinement may be similar from both sides under cusp confinement, but there are two major modifications.
If a Wiffleball forms internal containment may improve by several orders of magnitude. So, once a Wiffleball forms only one electron needs be injected for every ~ 100 electrons internally contained. So, a very robust electron gun may be needed at the start, but once the Wiffleball starts to inflate, the electron flux requirements diminish considerably.
Also, if a gas puffer, most of the may come from ionization of the neutral gas. Thus the E-gun injection flux requirements is lessened. I'm uncertain how this would effect the energy picture. Something like microwave heating might mitigate the number of high energy electrons that need to be injected.
The report that a more robust electron gun is needed may reflect the increased difficulty of injection with ~ 0.8 Tesla fields, and be consistent with what Happyjack27 has reported. It is important to note though, that this means more electron current. But, the total injection energy needed may change little due to energy recovery from rejected electrons. The major consideration may be the effect on vacuum pumping requirements. If the e- gun rejected electrons can be decelerated (give up their KE to the magrid) enough and quickly grounded without sputtering,the vacuum considerations may be minimal.
PS: The rejected electrons may also undergo recirculation- not just decelerated and grounded, but turned and re accelerated towards the magrid. The electron may have multiple opportunities to enter the magrid. I suspect. there are all sorts of permutations that might be relevant. B field pulsations, e-gun pulsations, POPS techniques, etc. may not only effect processes inside the magrid, but also outside of it.
I had been guessing it was a thermalization problem, but with "several anomalies" I should just throw my hands up and say "not enough information." They really are in uncharted territory here.
n*kBolt*Te = B**2/(2*mu0) and B^.25 loss scaling? Or not so much? Hopefully we'll know soon...
