torus electromagnets

Discuss the technical details of an "open source" community-driven design of a polywell reactor.

Moderators: tonybarry, MSimon

KitemanSA
Posts: 6114
Joined: Sun Sep 28, 2008 3:05 pm
Location: OlyPen WA

Post by KitemanSA »

D Tibbets wrote: My speculations is about the curvature of the magnetic field facing the core as it is pushed out by the Wiffleball. An ornginally almost conical field will be pushed into a more gentle curve. If the magnetic field is less conical to start, the end shape may be less convex- ie- more quasi spherical. At the regions where the field dives into the cusps, there will be a curvature based on the deformed field lines. I wonder if a flatter layout of the windings facing the center would increase the steepness of the cusp throats, essentially shrinking the size of the Wiffleball holes and improving the wiffleball trapping factor.
I've been pondering "holey" pancake magnets myself. Should be possible to get an almost tangential field across the face of the pancake and have small cusps at the holes and the X-cusps (modified funny cusps). We'ld need squarish and trianglish holey pancakes, but should be fun to wind!

MSimon
Posts: 14332
Joined: Mon Jul 16, 2007 7:37 pm
Location: Rockford, Illinois
Contact:

Post by MSimon »

Famulus is in the process of doing LN2 cooling.

Let us get rational for a moment and think high end hot rods. Building a Polywell is no more complicated. The complications are just different.

Once there is enough open source knowledge "everyone" will be doing it.
Engineering is the art of making what you want from what you can get at a profit.

ladajo
Posts: 6204
Joined: Thu Sep 17, 2009 11:18 pm
Location: North East Coast

Post by ladajo »

Will Famulus run into difficulties finding a way to cool his YBCO coils in the machine and operate it at the same time: yes
Will he look for a way to solve it: yes
Will he succeed: That is why he is trying

I actually had a thought where he could use copper tube to insert his YBCO into in the form of a coil, then run LN2 through each tube. The rub is how he contructs the core. Wall mount would seem to be the way to go, but given his current finances and material on hand, not so practical.

Maybe Marc can chime in on how he plans to cool the YBCO, and have it in a Magrid construct?

GIThruster
Posts: 4686
Joined: Tue May 25, 2010 8:17 pm

Post by GIThruster »

I think these questions and objections folks have to YBCO use have all been answered in the past. SuperPower has miles of this stuff in operation for several years now. All the hardware has been designed, tested and put into practical use, and the SP people are sharing their expertise about how to use what they've designed.

Just doesn't look like a big problem to me. You have to have the money to get the experts in, but other than that, all the science and technology is running. No new stuff needed. You don't for example need to worry about how to cool the YBCO. Years ago, SP had perfected their cabling so that each tape was wound helically, then surrounded by an outer casing which then had the LN2 pumped through. I'd be interested to see what they do with the newer 12mm tape especially since one wants as compact a magnet as possible. I'd bet it has it's own casing and is NOT helically wound but rather has the LN2 flowing along both sides of each tape.

Only way to find out is contact the SP people, but wondering HOW to do these things when someone has already done them seems a bit reinventing the wheel.
"Courage is not just a virtue, but the form of every virtue at the testing point." C. S. Lewis

MSimon
Posts: 14332
Joined: Mon Jul 16, 2007 7:37 pm
Location: Rockford, Illinois
Contact:

Post by MSimon »

Early MRI magnets had a two jacket solution: LN2 for the outer jacket. LHe for cooling the coils. Now a days many are conduction cooled.

So that has been done. Once you have a two jacket design going to a three jacket solution (LHe, LN2, H2O) is not so tough.
Engineering is the art of making what you want from what you can get at a profit.

ladajo
Posts: 6204
Joined: Thu Sep 17, 2009 11:18 pm
Location: North East Coast

Post by ladajo »

The YBCO tape that Marc bought has no casing.

WizWom
Posts: 371
Joined: Fri May 07, 2010 1:00 pm
Location: St Joseph, MO
Contact:

Post by WizWom »

I thought YCBO lost superconductivity at LN2 temperatures around 0.5 to 1 Tesla.
Wandering Kernel of Happiness

MSimon
Posts: 14332
Joined: Mon Jul 16, 2007 7:37 pm
Location: Rockford, Illinois
Contact:

Post by MSimon »

WizWom wrote:I thought YCBO lost superconductivity at LN2 temperatures around 0.5 to 1 Tesla.
I think you are in the right ball park. Even so LN2 is a start on the right path. And don't forget: even if it is only capable of realizing .1 T in the bore it will be continuous. Which makes measurement cheaper.
Engineering is the art of making what you want from what you can get at a profit.

KitemanSA
Posts: 6114
Joined: Sun Sep 28, 2008 3:05 pm
Location: OlyPen WA

Post by KitemanSA »

Colonel_Korg wrote:
KitemanSA wrote: So make a square coil and place a round sheath around it.
Don't not work. The magnetic field will conform to the shape of the coils, so the container will have to be the same shape. WB5 had square cross-section coils while WB6 has circular cross section coils.
Before too far, the field becomes effectively round and will follow a round sheath. Indeed, most SC designs I've seen on this forum use a square SC coil surrounded by many layers of TPS (thermal protection system) which becomes round at the outer layer. By that time, the field conforms.

And in truth, if the "square field" keeps the electrons away from the "round sheath", who cares? As long as the cusp is smaller than the gap between the coils, the minor X-section of the coils is immaterial.

GIThruster
Posts: 4686
Joined: Tue May 25, 2010 8:17 pm

Post by GIThruster »

Has anyone ever tried shrining the cusps with Mu Metal?
"Courage is not just a virtue, but the form of every virtue at the testing point." C. S. Lewis

D Tibbets
Posts: 2775
Joined: Thu Jun 26, 2008 6:52 am

Post by D Tibbets »

GIThruster wrote:Has anyone ever tried shrining the cusps with Mu Metal?
I'm confused, Wikapedia says it has high magnetic permeability, but state it is used for shielding against magnetic fields. I guess I should bone up on my terminology. In any case, I don't think you want to shield against the magnetic fields produced from the copper or super conductor magnets. Otherwise, how would the field get outside the casings and shield the structures, etc.

Dan Tibbets
To error is human... and I'm very human.

GIThruster
Posts: 4686
Joined: Tue May 25, 2010 8:17 pm

Post by GIThruster »

Mu Metal is used for shielding becasue its super-high permeability gives it the ability to draw magnetic flux into itself and away from other places. If you make a box out of it, imposed flux will travel through the MM walls and not penetrate the box because the reluctance of the box is many thousand times less than that of the air or vacuum in the box. It is highly anisotropic so constructing a box is not as simple as one might imagine.

Careful placement in places like the cusps might close them up entirely but you'd want to model this in 3D Mag field modeling software like Amperes to know for sure.
Last edited by GIThruster on Sun Jul 11, 2010 5:06 am, edited 2 times in total.
"Courage is not just a virtue, but the form of every virtue at the testing point." C. S. Lewis

D Tibbets
Posts: 2775
Joined: Thu Jun 26, 2008 6:52 am

Post by D Tibbets »

Comments on several posts above. Liquid nitrogen, if contaminated with oxygen or air (with water vapor) would cause all sorts of problems. But, I think the normal process of liquidizing nitrogen produces a fairly pure product. After that simple purging (with gaseous nitrogen, and keeping a closed system closed, or preventing backflow in an open system would not be challenging.

I'm not sure the magnetic fields would eventually reach a spherical shape from originally square crossection magnets, but it probably would at least trend towards this. In any case, I don't think the magnetic field shape is important from the standpoint of insulating structures, so long as the strength is strong enough to prevent unreasonable cross field transport. The volume of this strength (especially where opposing fields push against each other and thus compress this volume (like in cusps)) needs to be great enough so that the gryroradii of the charged particle do not allow them to reach the surface of the magnets. Having rounded casings helps this at any given strength, stronger fields with non rounded surfaces could compensate, at least to a large extent. That is why a sharp cornered rectangle is probably to vulnerable, but a mostly rectangular shape* with moderatly rounded corners should work if my reasoning is sound, with possible advantages that I mentioned before.
I have not seen any breakdown of the contributions, but I suspect the spacing between the magnets contributed more to WB6's performance, than did the rounded shape. Admittedly, I think the shape has increased importance in 'weak' magnets like those in WB6. I don't know the magnetic strength in WB5, though it looked like it had flat faces and perhaps some curvature on the sides as the wires bulged outward some. I'm not sure what in this arrangement of WD5 (and WB4) magnetic fields led to Bussard, etel's YUREKA moment about spacing, and recirculation. Perhaps (pure speculation) the WB5 was the best containment machine that Bussard could conceive, and when it showed insufficient containment, his brain was forced to conceive of alternatives outside the box, or areas where his assumptions went wrong. Perhaps it was here that he realized (he says he did realize this at some point) that his handling the magnets as theoretically infinitely small lines without any intercepts was inappropriate for the real world. As he said in his Google talk, this seemingly simple error was not appreciated by his team or the reviewers for over 10 years.

* Another consideration is that sharp corners could lead to arcing at lower densities. Even with magnetic shielding, this might have limited efforts to reach high densities within the Wiffleball.

Dan Tibbets
To error is human... and I'm very human.

Post Reply