If we assume that the grenade is a cylinder with length L and radius R, then we have a total current of I=LRj Amperes if j is the critical current density. Using the cylindrical coil formula E=(1/2)RN^2I^2f where R is the radius, I the current, N the number of turns and f is a form factor I will just ignore. N also goes away, since if we split the coil into many turns the area decreases correspondingly. So the total energy will be E=(1/2)R^3L^2j^2. Let's assume L=10 cm, R=5 cm and j a critical current density j of a million A/cm^2 (this can be achieved in a helium cooled superconductor). That gives a total energy of 6.25*10^13 J... whoa. That is 15 kilotons! What went wrong? Sure, I might have miscalculated, but I think the real limit here is keeping the grenade unexploded. That magnetic field is going to push the material *a lot*. I doubt there is anything that can hold together. Also, superconductivity quenches above a critical field of around 15 Teslas; I think this can be handled by having counter-rotating currents so the fields balance. Hmm, the total force on the outer part of the coilgrenade will be something like the F=2 pi I B N/m. The B field is ~15 T, so I get F=2*pi*10^6*0.1*0.05*15=471,000 N/m. Ok, that is not too bad: about half what a big locomotive can pull. I *think* superstrong materials can keep that together. Hmm^2... this nice page http://openlearn.open.ac.uk/mod/oucontent/view.php?id=398540&section=2.4 suggests that the critical magnetic field strength is *way* lower. Now I get currents on the order of 2,400 Ampere instead. I have a distinct feeling that I have confused units somewhere. I will retreat to my lair for some further reading and error checking... But I think the basic concept of a coilgrenade is sound. One can likely make a scaling argument that since it needs to be held together by molecular forces, it will at most be as energetic as a device releasing the energy in its strained molecular bonds - it will not be more powerful than a fuel air explosive (which cheats anyway, because it borrows a lot of chemical energy from the air). But it likely can be made to detonate in *very* cool ways, both releasing EMP and accelerating pieces in certain directions - it could in principle explode anisotropically like a mixture of a shaped charge and a railgun. It is worth thinking about.

Ah, here are some real data:

Quote:

The persistent currents in a closed superconducting loop will flow for months, preserving the magnetic field. As we calculated in the lecture, the energy density of magnetic field stored in the wires is B^2/(8 π) = 4 x 10^7 J/m3, assuming B = 10 T. Although this number is still much smaller than the energy density in gasoline (3.5 x 10^10 J/m3), it could be a possible solution to store the excrescent electrical energy. http://large.stanford.edu/courses/2010/ph240/li1/

and http://www.ewh.ieee.org/tc/csc/europe/newsforum/pdf/CR5_Final3_012008.pdf

Quote:

Although the attainable magnetic flux density limits the energy per unit volume given by Equation (1) ( B^2/2μ), the real limit of the energy stored in a SMES is mechanical. The virial theorem [4] gives a relation between the minimum mass of the mechanical structure, Mmin, and the stored energy, Wmag. For a solenoid this relation is: Wmag/Mmin = σ/d where σ is the working stress and d the structural material density. The relation defines the minimum mass of the mechanical structure in pure tension to support the radial electromagnetic forces. ... Assuming a reasonable working stress of 100 MPa, the virial theorem gives for a magnet with steel structure the value of stored energy per unit mass (mass specific energy) of 12.5 kJ/kg (3.5 Wh/kg). The CMS (Compact Muon Solenoid) [7] magnet of the LHC collider almost reaches this value for its cold mass (2.6 GJ/225 tons or 11 kJ/kg). The working stress of 100 MPa may be increased somewhat, but the mass specific energy will still be limited to the order of 10 kJ/kg. Some high-strength composite materials offer interesting perspectives for the future, because their stress density ratio is very high. High-strength aluminium alloys are also excellent candidates: they have approximately 1/3 of the steel density.

Now, diamond likely has a tensile strength above 60 GPa. So if we assume nanotech allows roughly this strength and aluminium-like lightness, the energy density can be boosted by 1800 times. This means 22.5 MJ/kg. This is a lot more, but still comparable to coal (24 MJ/kg). Still, far better than TNT (4.6 MJ/kg). Not a doomsday weapon, but a plausible grenade. And I suspect releasing it all on a micro- or nanosecond timescale allows some pretty useful physics that cannot be done with mere chemistry.

NewtonPulsifer wrote:

All this being said, it is likely that everyday use superconductor batteries are designed to fail in as benign a manner as possible. Things like shredding into pieces that have high air drag (like confetti) so an explosion is converted to as much concussion as possible in an atmosphere, as well as destroying itself over the longest period of time to spread the energy out as slowly as possible.

Good point. Makes for a neat piece of description: "There is a dull thud and the air fills with confetti as the robot's core battery fails."

kigmatzomat wrote:

I recommend the failure follow the popcorn model. The energy is absorbed by a material that rapidly phase changes to 99% liquid with 1% going to gas, creating a foamy mass. The outer shell ruptures (absorbing energy) and the rapid volume increase enables a resolidification. That gives you a nice "ka-PAF!" noise when a battery blows.

Your design works much better for a failure in a vacuum , but might be less space efficient to implement (as in there might be less room for battery due to the fail safes taking up more space). There might be vacuum and non-vacuum rated batteries (for safety classifications). Ablation might be useful for safety too.