Is There an Optimal Size for Magrid Casings?

[MSimon]

If superconductors are an absolute must, then you need to try more than one approach to answer the cooling requirements. Keeping the two water jackets full of liquid rather than steam across every magnet in the system is going to demand the use of two great big pumps and the flow rates are bound to cause trouble with vibration, not from the pumps per se, but from the turbulent flow inside the torii.

I'm thinking of pressure headers with every coil fed individually. With DP transducers, P transducers, inlet and outlet temps, and valves to see that the flows are even and sufficient.

A power plant is going to need a lot of control. Fortunately the control on the second or third one will be cheaper. A microprocessor assigned to the cooling circuit of each coil feeding a Master Controller should do nicely.

And yeah. Love that turbulence. It is great for heat transfer. But it brings problems.

[Billy Catringer]

So I can see, but as far as I can tell, and this may be simply be because of my faith in Doctor Bussard. He said the physics is done, so it's done. What I think is left boils down to two hairy engineering problems. First is the electromagnets, and how do we capture extremely irate He nucleii so as to derive current from their passing. Best I can tell, the latter problem as not been addressed at all. I would also think that the particle decelerated/generator problem could be studied without a working BFR. That could be done with a little LINAC or a gyratron flinging 4He at various decelerator designs.

Without the electromagnets though, nothing happens and it is the magnet problem that seems to be the most difficult one. I guess I am harping on you because you are the one working on the magnets and that involves the mechanical stuff I understand best, but I have a question about your superconductors which, I don't really understand all that well.

Presumably, you can put a current on the SC coils and they will maintain a current flowing through them so long as they are kept cold. Because the current keeps running, they maintain a strong magnetic field. So far, so good, but that is for undisturbed magnetic fields, right? When the electrons start following the field lines around in large numbers, especially after there are enough electrons trapped in the center of the well, a lot of stress is placed upon the magnetic field lines and I have to think, given that conservation laws are so ruthless, that this will result in some kind of effect on the superconducting coils. Either heat builds up in them, or they need more current to maintain the fields or both. Am I wrong?

[Billy Catringer]

I'm thinking of pressure headers with every coil fed individually. With DP transducers, P transducers, inlet and outlet temps, and valves to see that the flows are even and sufficient.

A power plant is going to need a lot of control. Fortunately the control on the second or third one will be cheaper. A microprocessor assigned to the cooling circuit of each coil feeding a Master Controller should do nicely.

And yeah. Love that turbulence. It is great for heat transfer. But it brings problems.

Do you want to sketch it, or do you want me to try? I'll need to ask about some dimensions. Have you done a flowsheet yet?

[MSimon]

I'm thinking of pressure headers with every coil fed individually. With DP transducers, P transducers, inlet and outlet temps, and valves to see that the flows are even and sufficient.

A power plant is going to need a lot of control. Fortunately the control on the second or third one will be cheaper. A microprocessor assigned to the cooling circuit of each coil feeding a Master Controller should do nicely.

And yeah. Love that turbulence. It is great for heat transfer. But it brings problems.

Do you want to sketch it, or do you want me to try? I'll need to ask about some dimensions. Have you done a flowsheet yet?

Flow sheet with respect to what? Project management? Control hierarchy?

Dimensions of what?

I'm already working on processor design, A/D and D/A converters. code etc.

[Billy Catringer]

Flow sheet with respect to what? Project management? Control hierarchy?

I'm already working on processor design, A/D and D/A converters. code etc.

Er, a flowsheet of the cooling system for the electromagnets. You've been working on the instrumentation, I assumed you'd have been working on flowsheets at the same time. If I'm wrong, please forgive my presumptiousness.

The heirarchy is already set, I think. I hereby nominate you to be the Boss Goat. Dimensions? Give me the ID and cross section of the conductive core and I'll start sizing tubing...er...stampings. The pieces will likely need to be made in a shipyard. Shipyards are the only outfits likely to have a big enough press brake.

EDIT: The way I see it, we can't do any harm by drawing up something based on what we know. We can buil.d it in Blender so that we can see what it might look like and start eliminating problems with fit up and what have you.

[MSimon]

The only flow sheet (and crude at that) was done for LN2. I came up with 100 gpm @ 100 psig max. For H2O I like 100 deg C delta T.

We can start out with a 10 Tesla coil 2 meters in dia. 16 million ampere turns. Figure a Jc of 10,000 A/sq cm That is 1,600 sq cm. That makes a radius of 29 cm just for the superconductors.

Any way - those are some rough numbers. One to ten Tesla coils.

Here is a magnetic field calculator.

http://hyperphysics.phy-astr.gsu.edu/hb ... urloo.html

[Billy Catringer]

Okay, good! Now I have something I can chew on. I'll be back w/ some pictures in a few days. The way I see it, we have a lot more to go on that Leslie Groves had. It's gonna be rough, but it ain't like I have anything better to do at the moment.

[MSimon]

BTW that is 100 deg C in the outer loop.

The inner loop should be more like a 10 or 20 degC delta T.

[Billy Catringer]

BTW that is 100 deg C in the outer loop.

The inner loop should be more like a 10 or 20 degC delta T.

By this I gather you intend to pre-heat the water. It makes good sense.

[MSimon]

The main thing is to keep the delta T between the inner water jacket and the LN2 jacket as low as possible.

I think a reflective coating between the jackets to lower the albedo is a good thing.

[Billy Catringer]

That's a matter of how you finish the stainless. I figure most of the inner tubing should be 316L or something similar. Aluminum coated Mylar does just about the best job at dealing with infrared. I suppose we could use it for lining in the insulation jackets. They used Mylar for a Echo 1, I think, so it would hold up to vacuum and extreme temperature changes well enough. I just don't know about how much outgassing we might get from it over time.

I went to bed about two hours ago then started tossing and turning over this gorramned thing so now I am up drinking coffee and smoking. What are you going to do about instruments inside these monsters? I don't think that there are many such designed to function inside of a 10 Tesla field. I figured out how to install thermowells, but will the thermocouples work?

I think you are going to have to go with by-guess-and-by-gosh on the coil temp after start up. For that matter, I don't know don't know that there is anything that you can wrap inside the coils that would give you an accurate temp as the coil temperature was being brought down.

Gonna hafta pump a vacuum on every cooling jacket prior to filling. High point bleeds would play merry hob with the magnetic fields and likely cause arcing to boot. Bloody thing is going to need several days for cool down prior to filling and start up. I expect that cool down will have to be done with gas phase He while cycling LN2 through it's own jacket.

I'm thinking that the outermost water jacket will be our main structural element that holds up against all the forces these things are gong to see. I'm thinking something like a 4cm wall thickness minimum, but probably thicker. I might even be paranoid and stick a joint of heavy wall mechanical tubing in at the bottom. The four upright magnets will get a lot of torque put on them where they are strapped to the standoffs.

See ya later, Chief.

[MSimon]

4 cm wall thickness? that seems excessive. BTW I was hoping Cu was good enough for the outer wall. It's conductivity would give some help with cooling. And later it might be good to make the first wall such that it had "fingers" extending into the H2O to improve cooling.

The main limitation in reactor size for pB11 is the thermal load on the first wall.

[Billy Catringer]

4 cm wall thickness? that seems excessive. BTW I was hoping Cu was good enough for the outer wall. It's conductivity would give some help with cooling. And later it might be good to make the first wall such that it had "fingers" extending into the H2O to improve cooling.

The main limitation in reactor size for pB11 is the thermal load on the first wall.

Can do copper on the outside. Is not problem as sheathing, but remember the superconducting coil alone weighs ~ 5,100 kg and the entire assembly is being pushed around by a 10T field. I haven't figured up all the weigths just yet, but you also have to allow for some major hoop stress in the water jackets. Nitrogen can run up some very high pressures if you have a failure in one of the outer jackets.

The only truly strong support that the upright magnets will have is at their bottoms where they are held up by the standoffs. Hoops are fairly strong when standing up, but tend to go egg shaped. The upper and lower magnets will be lying flat and they will need to be fairly stiff to avoid unwanted sag.

I have yet to figure out a way to prevent racking of the magnet assembly without very stiff outer tubes. There is not a good way to install diagonal supports. Also, the best way I know of to eliminate uwanted vibration is to use more mass.

Lemme do some more head scratching and fiddling around here. I'm guessing from experience on this stuff anyway. Everything I do will need to be checked by someone who can do a real stress analysis. I know what is about right, not what is exactly right. I have always been on the builder's side of this kind of thing. My goal is to give the smart folks on this forum something to look at, shoot at and otherwise criticize.

I was thinking of putting some vanes inside the outer water jacket so that the water would be forced to circulate around as it traveled from inlet to outlet. The inward fingers to improve conduction into the water is probably a better idea.

I suggest using Kel-F, or Teflon-FEP, or Neolon (all the same thing) for supporing the SC coil in its tube. This stuff remains an insulator at low temps and will creep a little under load. The creep is necessary to allow for contraction in the SC coil.

[KitemanSA]

Please remember that Dr. B had envisioned that the final coils would be either square or pentagonal in plan-form (not cross section, plan-form)with rounded corners. I think this will tend to reduce the overall stresses in the magnet because the coil and virtual coils forces will be better balanced. With the "straight" sides of the coil actually following a "great circle" along the sphere of the well rather than the edges of the cuboctohedron, the primary forces might resolve into simple hoop stresses in most of the magnet.

I think there may have been other reasons like better sphericity and smaller "funny cusps" that made him want these flat sided coils.

MSimon was kind enough to place some graphics of these concepts in photobucket, but I guess I just don't get it cuz when I tried to "img" them, they came out ity-bity unviewable postage stamps.

{Edit=replace vertices with edges. Folks I don't know why, I DO know better, but somehow I got "vertex" in my mind when I ment edge. DWI. Discussing while idiotic!}

[MSimon]

Please remember that Dr. B had envisioned that the final coils would be either square or pentagonal in plan-form (not cross section, plan-form)with rounded corners. I think this will tend to reduce the overall stresses in the magnet because the coil and virtual coils forces will be better balanced. With the "straight" sides of the coil actually following a "great circle" along the sphere of the well rather than the vertices of the cuboctohedron, the primary forces might resolve into simple hoop stresses in most of the magnet.

I think there may have been other reasons like better sphericity and smaller "funny cusps" that made him want these flat sided coils.

MSimon was kind enough to place some graphics of these concepts in photobucket, but I guess I just don't get it cuz when I tried to "img" them, they came out ity-bity unviewable postage stamps.

It probably has to do with how you saved them. You need more dots.