Grid evolutions

For Farnsworth type designs.

Grid evolutions

Postby Doug Coulter » Wed Apr 09, 2014 2:08 pm

Here are some grids as we've (mostly) used/tested, and with a new one I want to try.
Grids.JPG
Some recent grids. Click pic for a bigger version.


The one in the upper right is the oldest, and until recently, our most reliable record setter. The carbon was mirror smooth when I first put it in - you can see some roughening due to sputtering, and the fact that the rods are a bit "twisted" around the axis, which wasn't deliberate - it was a piece of junk under the circle table on my mill, which I used to drill that one. It has an OD of 1", and uses 40 mil diameter rods. The twist not only seemed to not hurt anything, if anything, it helped. When we went to untwisted grids - we get a very tight beam of "something" out of the end, pointed right at my glass viewport, hot enough to make a stainless steel screen red hot, and the first piece of glass (there's one protecting the thousand dollar door glass) fluoresce in both white and blue (different particles? Who knows? Cracks it circumferentially too.).

For various reasons, I wanted to play with scaling. The grid currently in the tank has about the same dimensions as the grid in the upper right here, but thinner carbon, and a female threaded base (the new system for carbon ends).
The grid on the upper left is Ti ends. Ti just didn't work out that well - not thermally conductive or radiative enough, and long before it hit max power I feared the end ring would melt and slump, so I took it back out and made about the same thing, a little shorter - which was an obvious move if you've seen the kernal of the poisser concentrate in the middle anyway - and it's our current record setter in both Q and in sheer output (which are different modes of fusor operation), showing us a 2800x Q improvement, and 10m neuts/second at max power - the new one will take the full output power of the Spellman SL2KW, no problem, at least for a minute or two, which is more than it takes to hit 4k cpm on our silver in the neutron oven. It throws that nasty beam as well. Lately, except for just wanting to see huge neutron counts for fun, we're not running in that mode anyway, but in an oscillating mode - we seem to see a lot more fusion and Q during onsets of these oscillations, so that's the general direction our work is going. Some of what we are seeing could be instrumental artifacts due to the slow samplng rate of our power supply monitor, but we are sure that not only do our counters never miss a pulse (faster counters than the sensors can go) but also....the silver never lies about the integrated neutron output, nor does the "low passed" power supply sensing lie about average power - it could be that the peak power isn't getting measured right, but the average is certainly correct.

All of them seem to give a very tight focus at the tank walls but not so much at the center - about human hair sized, but you can see the width "at all" with the naked eye. I am not sure how the rod spacing and space charge affect this, so for a first try at seeing something different, I kept with 8 rods (20 mil) which amounts to a bit closer spacing, but with a shorter distance to the center...Hopefully, careful observation will show if this is the right direction to be moving in, perhaps 6 rods (same geometric transparency) would have been better, but with this design - not much material is wasted, so we can try that too if it seems indicated by results.

Hopefully I'll get to try this new one soon. I have a visit from the VA Tech physics dept in 2 days, so I might just "not fix what ain't broken" till they see the current record setting lashup. The only truly nice thing I can say about that recent 15 min of fame is that now we are getting quite a lot of attention from sceptical but credentialed scientists now. When they see with their own eyes, it's good for our credibility, to say the least.


EDIT: FWIW, I've used both Th (or other electron emitting doping) and plain tungsten at various times. Plain seems best and we see a huge excess of negative charge on every faraday probe we put in there anyway, no need to increase that, or as far as I can tell it just wastes more power, but the effect isn't huge either way, even if you heat things to the point it would make a difference.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: Grid evolutions

Postby Jonathan Schattke » Tue Oct 14, 2014 7:16 pm

I'm looking at Rider 1995 "A general critique of inertial-electrostatic confinement fusion systems" and especially the section on grid losses.
E. Electron grid losses
Now consider the power losses that are caused by con-
fining the particles with an electrostatic grid instead of a
magnetic cusp system. The ion losses on the grid can be
minimized by making the grid bias large and positive, so that
ions hitting the grid will possess essentially zero energy;one
must then calculate only the electron grid losses. Assuming
that the grid has radius Y.&d and transparency 77, to electrons
passing through it, and choosing the electron energy, velocity,
and density to be evaluated at the grid, one then obtains

P_e_lossgrid = ( 1 - η_e) * 4*pi * r^2_grid * η_e_grid * ν_e_grid * E_e_grid

Noting that η_e_grid = sqrt(2E_e_grid/m_e), and expressing the power
in watts, energy in eV, and everything else in cgs units, one
finds that

P_e_lossgrid = 1.19 * 10^-10 * (1 - η_e) * r^2_grid * η_e_grid * E^(3/2)_e_grid W


Now, this is based on a spherical grid; your cylindrical grid will trade these for cusp losses along the axis.
But the paper goes on to place grid losses as ~3000 power produced.
He states that a biased grid leads to Ion losses instead, and pegs those at ~40x power
User avatar
Jonathan Schattke
 
Posts: 10
Joined: Wed Oct 08, 2014 12:31 am
Location: Rolla, MO


Return to Fusors, Farnsworth type

Who is online

Users browsing this forum: No registered users and 1 guest

cron