5 ms·

Yes, it's possible but it requires a thick jacket. the formula for a thin-shell cylindrical pressure vessel is \sigma = rp/t, where sigma is tensile strength of

Yes, it's possible but it requires a thick jacket. the formula for a thin-shell cylindrical pressure vessel is \sigma = rp/t, where sigma is tensile strength of the material, p is pressure, r is radius, and t is wall thickness. For thick-shell jackets, the formula is slightly more complicated [1] but you get the idea. For Kevlar, max \sigma is 3.6GPa. So it's possible to maintain an internal pressure of 200GPa (2MBar), but the wall thickness must be much larger than the inside radius. [1] https://tge.readthedocs.io/en/latest/content/mechanical/thick_wall_pressure_vessel.html https://tge.readthedocs.io/en/latest/content/mechanical/thic...

semi-extrinsic 7y ago
We quoted the same formula, but AFAIK the radius is the outside radius, right? So you cannot go above t/r = 1.0?
tgsovlerkhgsel 7y ago
Wouldn't that mean that we can get a 1 mm radius almost-room-temperature superconductor with a wall only 55 mm thick? That seems like it would almost be practical.
vlehto 7y ago
If you use the formula for thick walled cylindrical pressure vessel, you hit practical limit with outside diameter being twice the inside diameter. When this happens and you also take into account fatigue, usually about half of the yield strength of the material is usable. Your best bet is using thin shells of maranging steel wrapped around each other with thin layer of pressurized non-compressible fluid in between. The pressure control of the fluid is then absolutely critical. Alternate with 0,5mm thick fluid and 1mm thick steel sections. Rinse and repeat. The total thickness would not be all that crazy, but the pressure control would be incredibly expensive. EDIT: Did some calculations. Using roughly that system I described earlier you would only hit 10 GPa with roughly 50 meters thick piping system, that would have 75 alternating layers of pressurized oil. And each layer would have different pressure from all the other 74 layers of pressurized oil. So not possible. If you try using single thick wall in your cylinder, increasing the wall thickness means that your max stress gets closer to the internal pressure reading. So you would need material that can deal with 200 GPa of stress.

semi-extrinsic 7y ago

We quoted the same formula, but AFAIK the radius is the outside radius, right? So you cannot go above t/r = 1.0?

tgsovlerkhgsel 7y ago

Wouldn't that mean that we can get a 1 mm radius almost-room-temperature superconductor with a wall only 55 mm thick? That seems like it would almost be practical.

vlehto 7y ago

If you use the formula for thick walled cylindrical pressure vessel, you hit practical limit with outside diameter being twice the inside diameter. When this happens and you also take into account fatigue, usually about half of the yield strength of the material is usable. Your best bet is using thin shells of maranging steel wrapped around each other with thin layer of pressurized non-compressible fluid in between. The pressure control of the fluid is then absolutely critical. Alternate with 0,5mm thick fluid and 1mm thick steel sections. Rinse and repeat. The total thickness would not be all that crazy, but the pressure control would be incredibly expensive. EDIT: Did some calculations. Using roughly that system I described earlier you would only hit 10 GPa with roughly 50 meters thick piping system, that would have 75 alternating layers of pressurized oil. And each layer would have different pressure from all the other 74 layers of pressurized oil. So not possible. If you try using single thick wall in your cylinder, increasing the wall thickness means that your max stress gets closer to the internal pressure reading. So you would need material that can deal with 200 GPa of stress.