Superconducting quench switches / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Superconducting quench switches

Move Thread

LAN_403

Marko

Wed Nov 09 2011, 02:58AM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Hello eryon

There is one very fascinating idea that came to my head recently. I was reading up about some commonly used superconducting materials and found out one interesting, yet mostly considered unimportant fact - a lot of these materials such as YBCO and similar high-temperature superconductor ceramics are actually very good insulators once out of superconducting state. Even strontium titanate ceramic (used for RF doorknob capacitors) apparently exhibits this at <0.35K!

the basic idea is having a superconductor carrying a (likely extremely heavy) current in it's normal, cooled state. To switch it off, a fast rising pulse of current is given to the superconductor by means of capacitive discharge, causing it to reach critical magnetic flux density and having the superconducting state collapse rapidly, which is sustained by positive feedback of the current heating it up. As the material transits to it's insulating state, it would halt the current flow rapidly and (hopefully) prevent it's destruction from ohmic heating.

Of course, a switch like this would only have a fast and reliable turn-off, and somewhat more troublesome turn-on as it would need to be cooled back below critical temperature.

The idea seemed so interesting that I thought someone might have thought of it before, but I could barely find information on conductivity of few superconductors at room temperature.
If something like that could succeed, it would be a great means of switching the superconducting energy storage coils to various loads or just for producing extremely high voltage pulses (producing lightning, powering pulsed linear accelerators and free electron lasers).

Obviously many problems remain unsolved, such as the question whether the useful material actually even exists, and the reduction of stress on the superconductor in rapid quenching (here we can't have the aid of metal plating to carry over the current during quenching). This would likely be exaggerated by usual brittleness of superconducting ceramics.

One would need to find a solution for total energy absorbed during quenching transition (turn off loss!) and find ways of decreasing it to minimum value.

One would probably also want to control this switch from low side (while the switch may be passing current from a heavy superconducting storage which may be impossible to change rapidly enough). To solve this two superconductors may be used in a form of transmission line and pass the current in parallel, while the capacitive discharge current (orders of magnitude bigger than the load current) is discharged into the bottom end of the line.

And in some applications, quenching might even be attained using external magnetic field, or a radiation pulse.

So I was wondering if anyone knowledgeable of superconductors could offer advice whether the idea holds any water at all, as well as properties of some superconductors. (If anyone owns a YBCO pellet, perhaps nuking it with some HV to check it's conductivity at room temperature.

In the end someone ambitious might even try creating a (sort of) a tesla coil driven by this method, if affordable enough superconducting material can be found and cooled cheaply by liquid nitrogen.

Marko

Adam Munich

Wed Nov 09 2011, 04:50AM

Registered Member #2893 Joined: Tue Jun 01 2010, 09:25PM
Location: Cali-forn. i. a.
Posts: 2242

Pretty sure this is how they short and interrupt the coils in MRI machines.

radiotech

Wed Nov 09 2011, 08:57AM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

One issue with a superconductor coil carrying a large current is when a quench
occurs there is a substantial magnetic field collapsing into a conductor which now
has a high resistance and might be damaged. Methods of detecting the quench
and applying a dissipation resistance to limit the voltage are needed.

Many of these issues were solved around 50 years ago with helium superconductors
and came back to consideration when marine applications seemed practical for weight
reduction of electrical generators.

Try and find J R Bumby, Superconducting Rotating Electric Machines 1983 University
of Durham

Marko

Thu Nov 10 2011, 09:35PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Hi guys

yeah, it's true they use a sort of a switch like this in MRI magnets, but only to handle over the supply current and not to interrupt heavy currents.

One would probably need to find a material that works well enough as moderately compressed powder in a flexible tube to prevent possible cracking that might occur with solid ceramic (such a contraption might be prone to arcing though).

Anyone experienced with superconductors around here?

Marko

Conundrum

Mon Nov 14 2011, 09:45PM

Registered Member #96 Joined: Thu Feb 09 2006, 05:37PM
Location: CI, Earth
Posts: 4061

Soon will be, am attempting to make some of the 212K material (Sn5In based) at the moment.

I would start with YBCO but its a pain to get yttrium oxide whereas the 212K material just needs tin, calcium, barium, copper and indium in the right proportions.
Googling "molar quantities" helped a lot, as for this to work you need to get everything right at the same time and a 0.2% change in quantities or iron contamination is enough to drop the useful fraction to zero.

Also has the advantage that it is possible to use magnetic susceptibility tests (aka hysteresis) to tell if any of the material is changing resistance at a given temperature which allows you to fine tune things/regrind/etc without starting from scratch
each time.
The best method I have learned of is to wind a coil and energise the material with an RF field in a quiet area of the spectrum and tune up the circuit for resonance. At or near Tc there will be a sudden change in the resonant point which will generally be repeatable as the temperature is cycled.

For fine tuning etc, my 0.01g accuracy "Ebay Special" scales work well BUT to increase the sensitivity keeping the scales at the same temperature also helps a lot as they drift over five minutes or so by 0.1g in either direction.
Apparently expensive analytical scales have this built in and typically take half an hour or more to settle to full accuracy.

Maybe connect a unity buffer and A-D to the output and monitor on the PC?

An idea I did have is to locate areas of interest in a gradient sample made by applying differential induction heating during annealing in flowing O2 using the RF technique.
Even magnetic viewing film will often show up an interesting pattern you would otherwise miss.

EDIT:- also see

Seems that one crucial step is pre-making the CaCuO3 which requires extended heating over a long period of time.
Without this step many superconductors using calcium won't work well if at all.
Also may be needed for BaCuO3 as well, which is why most researchers buy this ready made and tested.

-A

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob