Electron thermalization time versus confinement time.

[TallDave]

I like alex's (Dan's?) conceptual picture of a negative potential at the middle of the cusp. The electrons on the inside half tend to get "plugged" while those on the outside eventually reach a point where they want to move back toward the Magrid, so they tend to oscillate. There's some recirculation but a general flow outward because we're pumping 10MW of electrons into the WB.

So you have potential sloping up from wall to Magrid, then down from the Magrid to the center, with dips in the cusps (or, from the electrons' perspective, sloping down from wall to Magrid, then up to the center, with "hills" in the cusps).

Maybe an actual picture would help.

[alexjrgreen]

???

I said

if the potential of the wall is more negative than the space potential near the throat of the cusp

You say

the magrid, which is more positive than the wall.

where is the contradiction?

The space potential near the throat of the cusp I was expecting to be slightly negative.

[TallDave]

It might be worth noting too that we know the wall probably isn't absorbing many electrons. Rick has said the hot spots were on the machine, not the wall. Someone correct me if this is way off, but that seems to argue the electron losses are dominated by losses to interconnects (which are poorly shielded) and Magrid (through cross-field diffusion).

[Art Carlson]

???

I said

if the potential of the wall is more negative than the space potential near the throat of the cusp

You say

the magrid, which is more positive than the wall.

where is the contradiction?

The space potential near the throat of the cusp I was expecting to be slightly negative.

I thought you meant "slightly negative with respect to the magrid". Please tell me that is what you did mean, because "slightly negative with respect to the wall" really makes no sense at all.

[alexjrgreen]

The space potential near the throat of the cusp I was expecting to be slightly negative.

I thought you meant "slightly negative with respect to the magrid". Please tell me that is what you did mean, because "slightly negative with respect to the wall" really makes no sense at all.

I did mean with respect to the wall.

Without a cusp jet, the potential at the centre of a magrid coil is close to ground because of the negative potential on the wiffleball.

Add a cusp jet with a net negative charge and that should put it at a lower potential than the wall.

Did someone suggest a picture?

[TallDave]

Since there are many more electrons in the wiffleball than in the cusp jet, and a net negative charge overall, the ion is attracted less strongly than before to a point between the cusp and the centre of the wiffleball, but much closer to the centre.

Agreed. Keep in mind too, the center is a lot easier to get to. If alphas are making a thousand transits at MeVs, fuel ions will have a hard time making it into the cusp even if they do see a competing potential there within their Debye sphere. Also, since electron density inside the WB is 1e4 times greater than out, the cusp potential should be relatively small.

[KitemanSA]

It might be worth noting too that we know the wall probably isn't absorbing many electrons. Rick has said the hot spots were on the machine, not the wall. Someone correct me if this is way off, but that seems to argue the electron losses are dominated by losses to interconnects (which are poorly shielded) and Magrid (through cross-field diffusion).

There is a distinction to be made between absrbing electrons and absorbing energy. The wall may be absorbing 10 times as many electrons as anywhere else, but since they've given their energy up to the potential field, they don't have much left when they reach the wall. The electrons hitting the MaGrid however give up their maximum energy. Fewer electrons, MUCH more energy = hotspot!

Oh, and the relative areas may also come into play.

[TallDave]

True. I should modify my statement to "electron energy losses." And still subject to the areas involved.

Come to think of it, some are claiming ions are slamming into the walls too, so I should probably further pare that to "energy losses."

[rcain]

...The electrons hitting the MaGrid however give up their maximum energy. Fewer electrons, MUCH more energy = hotspot!
....

.. at the risk of taking things off topic, i have been much more concerned of late about fusion products impacting the magrid, assuming its operating at full commercial whak, and not just an experiment in containment.

bytheby, has anyone done any math regarding trajectories, impacts and damage? mitigation?

[TallDave]

rcain,

Alphas will be corralled into cusps. According to Rick, they make 1000 transits on their way. So, the number smacking the Magrid is presumably small. I don't know if anyone calculated how many would do so at reactor rates and what the spallation rate might be, but it seems it ought to be low.

Neutrons will still be a first-wall problem in D-D/D-T. M Simon did some BOE work way back and found this may mean the sweet spot for such reactors is around 100MW, because past that point costs rise as R cubed while power can only rise at R squared due to the heat load (inverse square law). There were some thoughts on materials as well (boron plus water? I can't recall).

If you search "heat load" I think you can find that thread.

[rcain]

thanks TallDave. maybe i'll reinvigorate that particular thread.

however, 'corralling' anything anto a cusp, if ithey are travelling that fast, seems like a big quetsion to me. as to 1000 transits - i thought that was lifetime for injected ions, before upscattering/escape, not direct fusion events.

fast neutrons, yes... i'll take it up again on the other thread.

[TallDave]

I was surprised too.

Way back in the pre-TalkPolywell days, there was a discussion on the Polywell wiki in which we agreed alpha sputtering/spallating was a major issue, maybe a showstopper. But as it turns out, at reactor magnet strengths the alphas swirl around and leave through a cusp, according to Rick (I imagine he, Park, Wray, and etc. have some nice simulations going for their reactor designs).

It was quite a facepalm moment for myself. I had just assumed at MeVs even 10T fields wouldn't deflect them much.

I'm pretty sure Rick's comment on 1000 transits is here somewhere. Google his posts with "alpha" and it may turn up.

[TallDave]

OK, I was curious so I went and found it myself:

4. As for Mr. Tibbet’s questions relating to alpha ash, these devices are non-ignited (i.e. very little alpha heating) since the alpha particles leave very quickly through the cusps. If you want to determine if the alphas hit the coils, the relevant parameter is roughly the comparison of the alpha Larmor radius to the width of the confining magnetic field layer. I’ll leave that as an “exercise to the reader” as well.

Haven't found the # of transits mentioned yet though. I'm pretty sure it was said to be 1000.

[TallDave]

Ah, here we go:

viewtopic.php?p=18441&highlight=alpha#18441

The alphas make about 1000 passes before they exit through the cusps. They leave at essentially full energy.