Slashdot Mirror
Superconducting Cables To Carry Power In Detroit
bewert writes: "Check out [this Knight-Ridder wire story.]
This could change electricity distribution economics as
we know it. A project is under way to replace 9 major
copper power distribution cables with 3 smaller ones
made from a high-temperature superconducting material
called BSCCO (pronounced bisco). Pretty interesting technology,
and one that could have huge implications for reduction of
transmission power losses and the need for more generation." Not to mention that it means a 25-fold reduction in the weight of the cables used to carry electricity for a large chunk of Detroit.
Yikes, let's clean this engineering up.
1. One of the main reasons for using the HTS cables is space. Detroit Edison needs to increase the current carrying capability at the station, and using the HTS means they get more capacity with the same underground conduit, so they don't have to excavate to improve the circuit. big savings. Plus, the smaller diameter and weight make it easier to pull the cable through the conduit, and eliminates the need for splices in the cable due to maximum pull weights. Bad splices are a common cause of failure in underground cables. Of course, if you are ComEd in Chicago, you just ignore that until the city goes dark...
2. Cables have higher capacitance than overhead transmission lines, because the conductor is closer to the ground potential. It also has lower inductance for the same reason. There is no external electric field. The sheath is at ground potential, so the field is between the conductor and the sheath.
3. Long distance cables are typically DC because in high voltage AC cables the voltage increases at the sending end due to the high capacitance of the line. That's why underground AC networks have shunt reactors, to keep the voltage down.
Although inductance does reduce the maximum power transfer in a circuit, it's due to its affect on voltage.
4. Although the lossless characteristics of HTS are important, that doesn't by itself make the economics attractive. Avoiding construction costs and pushing more power through the same rights of way due to higher current density is the niche that HTS is currently filling.
Elvis, the power engineer-nerd.
The first bird that perches on these power lines and puts a talon through the insulation is going to get a nasty surprise.
I wonder what the cooling system looks like for these new lines? It seems quite challenging to cool all that cable and prevent any LN2 leakage. More importantly, if a leak happens, and the cable rises above its transition temperature while carrying a large current, there must be some kind of backup system to shunt off the current and prevent the heat generated by the sudden resistance from damaging the cable. Perhaps that's why the ceramic ribbon is wrapped in silver.
I don't do hardware -- would some actual power engineers care to comment on the cable design?
nature already delivers free power to each and every home in the USA. It's called sunlight.
But I guess if we learned to take advantage of that, then there would be no use for bloated inefficient electricity utilities.
These are my friends, See how they glisten. See this one shine, how he smiles in the light.
You didn't miss a thermo lecture, just a economics one. A given commody market value can vary over time. Storing it and paying rent on the storage and selling it later can be quite profitable. It can also lead to quite a loss if the price never goes up enough (or you can't wait that long).
That's what makes the stock market work. And the power market. And the futures market. And...
According to the article the existing copper cables are cooled with oil. I expect that means they are only replacing existing high mantinance (high capicity?) cables with these things.
I don't know enough about power distribution systems to know where these cables live, but I'm betting they are not the overhead phone pole kind. Maybe they are only found much closer to the genneration systems.
--
--
For instance, temperature is just one parameter to look at when you're looking at superconducting cables. Increasing the current density and/or the magnetic field will also tend push you out of the superconducting state. Tc is the temperature when current and field is zero, and the trouble with the high Tc superconductors is that you don't have a lot of clearence between the temperature of liquid N2 (77K) and Tc. When you try and load the cable with current, what you might call the "effective" critical temperature is going to be lower. The easy form of BSSCO is only something like 90K -- can't believe I don't remember more precisely than that, I used to work on this stuff -- anyway, I refuse to believe that they've managed to reliably come up with the 125K form of BSSCO, that's one holy grail that was looking pretty elusive, at least as of ten years ago.
A minute with google turns up what looks like a pretty good technical article about the processing of BSCCO/Ag tape: Is Low Cost BSCCO Tape Just Around the Corner?. (ObGripe: sure would be nice if the slashdot crew would do a teeny bit of background research on these stories, instead of just pointing us at junk news sources). Looks like I might be wrong about the 125K form of the stuff: they talk about working with both the 2223 and 2212 compositions (the numbers there are the main stoichiometries of the compound, e.g. Bi2 Ba2 Ca2 Cu3 Ox... as I remember it they don't usually specify the amount of Oxygen in the mix, because it's a bitch to measure it, and it tends to vary anyway). But then, they wouldn't be talking about both forms if they had the 125K form working really well.
Looks like they've got some decent numbers from direct measurements of current/area, which makes sense, or they wouldn't be announcing projects like this.
(By the way: one of the cool things about BSCCO -- I wonder when they made up this "Bisco" business, that's a new one on me -- but all the components are relatively non-toxic. At least they're not using something really evil like Thallium.)
Fissure erupts, turns citizen into pillar of salt! Story at 11.
(so what's ksu's analogue to ku's kulua?)
Things that can make logic gates:
Vacuum tubes
Transistors
Relays
Streams of water
Ropes and pulleys
Brain cells
If you can create an intelligent out of semicondutors, then you can do it with a sufficiently complex plumbing system.
Excellent! Now I can finally replace my gold metal speaker wires with liquid nitrogen cooled, silver encased, bisco ribbon cables! Then I'll finally be able to sleep at night. Well, as long as I have the cryo, can I replace my tubes with Josephson junction? Thermal noise begone!
It's not that the single electrons 'don't collide': coupled pairs of electrons are bosons, since spin 1/2 cross spin 1/2 yields either spin 1 or spin 0: both of which are bosonic.
You're probably familiar with the Pauli exclusion principle for electrons - this says that no two electrons (fermions) can occupy the same quantum state. This is because electrons, having spin 1/2, are fermions, and obey Fermi-Dirac statistics (interchanging fermions changes the sign of the wavefunction).
Bosons don't have a Pauli exclusion principle - interchanging bosons doesn't change the sign of the wavefunction at all, and so the number of particles which can occupy a state is unlimited.
Thus, imagine a pair of electrons in a material- nothing is stopping them from radiating their energy away and settling into the ground state- it doesn't matter that there are other pairs of electrons there, since a pair of electrons is a boson. They then settle into the ground state - the lowest energy state.
Now, you've got a curious situation. The electron pair is propagating through the material since there's a potential well - there are particles which they COULD scatter off of and lose energy - but they're in the ground state - they CAN'T lose energy, so they CAN'T scatter. It's not that the electrons don't collide 'as much': ideally, they don't collide at ALL - resistance zero.
As for the 'electron-pair' waves: the idea of waves and particles being separate things is a classical idea: Nature doesn't quite agree with that. Particles are waves, waves are particles, it's all the same bloody thing. Calling them 'electron-pair' waves is fine: calling them 'electron-pair' particles is also fine. de Broglie matter waves are also a semi-classical idea: halfway between QM and classical mechanics - it would be a good thing to abandon that idea, and just accept that fundamentally, all matter has a wavelike nature.
For a good start in QM, read D. Griffiths "Introduction to Quantum Mechanics": Griffiths's texts are usually quite amazingly good at allowing students to actually understand the physics of the situation. Getting rid of classical intuition like "physical location of a particle" and "momentum of a particle" and "particle" in general is a good thing to learn as soon as possible if you are interested in higher physics.
There are other processes in superconductors which do cause resistance-like effects, but fundamentally, the superconductor itself has zero resistance - you might have things like thermal resistance, or 'magnetic resistance', or impurity resistance, but the superconductor literally has zero resistance, almost (note almost) by definition.
No, actually, impedance is a 'resistance' to current flow due to a changing magnetic field, which induces an opposite EMF in the circuit. Physically, the best analogy would be if you imagine resistance to be traffic slowing you down in a car: impedance would be similar to something like bad gas in a car reducing the amount of power available, or going up a hill (going up a hill is a weak analogy, but it has a strong parallel in the whole EMF/potential well thing).
Argh, argh, argh. Somehow hit 'submit' before I was finished typing. How aggravating.
It's true - there is no resistance in a true superconductor, isolated from all else, with an isolated current inside of it. However, place that superconductor in a circuit, and many various issues will arise, again, especially in T2 superconductors which will admit magnetic fields through their volume.
Still, the best way to describe it is to say that the superconductor has zero resistance, and that effectively you have additional objects introducing mitigating effects.
The first superconductor wasn't lead - distinctly not! - it was mercury. 1911, Heike Kammerlingh-Onnes, using liquid helium to cool mercury below 4K.
Actually, no... its resistance does drop to zero, literally. However, the behavior becomes 'curious' when you actually try to drive current through it - especially since this is a T2 superconductor,which means that it forms magnetic whorls at any non-zero magnetic field. As you know, a current creates a magnetic field, and so you get a sort of 'balance' between the amount of current you can drive and the maximum field that the superconductor can support. If you try to drive more current than that, then you will begin to drive the substance out of a superconducting state.
Is it a resistance? Well, no, distinctly not - it's not linear in voltage, for one. It could be thought of as an 'effective' resistance, but it's not resistance.
Incidentally, it's very dangerous to simply show "Look, it's infinity!" and use that as a disproof - several things in nature are extremely curious and are literally infinity: take for instance a superfluid, which has, literally, zero viscosity, or an electron, which has (as far as we know...!) zero volume. Dividing by either of those things would tend towards infinity - the fact that it does not actually get to infinity simply means that another process begins to dominate and damp the previous one.
Oops, that's Y Ba_2 Cu_3 O_6 superconductors, and I may also get to play with NbSe_2 superconductors, too. The cyclotron is pretty cool, too.
Good point. Maybe they should switch to DC for these lines. Then there would be effectively no inductance losses, either.
The energy savings is in power loss. I suspect space/weight savings are secondary. Superconductivity means no resistive power loss, whereas normal transmission means usually lose you 10% or so.
As for the cost of cooling the nitrogen, that's trivial. LN2 is as cheap as soda pop.
This summer, I'll be working at the local particle accelerator doing beta-NMR and muon spin rotation experiments on high-temperature superconductors... Should be lots of fun! We aren't studying that particular kind, though (I think just the Yt-Ba-CuO ones).
Ohm's Law only applies to what are called Ohmic resistors. Some metals are generally Ohmic; others are generally Ohmic but only at particular temperatures. Some substances are not Ohmic at all, such as the YBCO superconductors I worked with as a research aide at the Texas Center for Superconductivity.
For comparison: Ohm's Law generally applies to copper, no matter what the temperature is. Ohm's Law stops applying to aluminum once you cool it to about 4K.
Remember: Ohm's Law is a macro-scale observation, and superconductivity is a quantum-scale event. At the quantum level, all sorts of strange things happen that are totally contrary to our macro-scale observations. The Einstein-Bose Condensate is a great example, as is superfluidity in liquid helium. (Anyone who is not utterly shocked and amazed by superfluidity apparently hasn't seen superfluids before.)
Lack of resistance (up to a point), and greater over-all power capacity.. The article says 150 times, meaning if we had 50 or so separatly maxed out copper cables running at a constant voltage, we could combine them into 1 (with the same voltage or current (or combination of the two)). Traditionally if you wanted greater power capacity, you'd do what you needed to the wire to allow higher and higher voltages to minimize the current. Higher voltage usually requires greater separation between lines (since there is significantly enhanced potential for shorts). So by not _having_ to up the voltage, you can keep lines closer together.
So the benifits are power efficiency, and that you need less total physical stuff to get the power downtown.
-Michael
-Michael
...to replace 9 major copper power distribution cables with 3 smaller ones...
So, we'll only need to have 3 cables break before Detroit will lose power...
There's something to be said for redundancy and multiple paths...
-- You can't idiot-proof anything, because they're always coming out with better idiots.
Even with only 100A of current, there still is a huge amount of loss in the line. I very recently graduated with my EE in power systems as well. A tour of Detroit Edison not long ago confirmed this.
The higher, the fewer.
Trading a thick heavy but otherwise low-maintenance copper cable for a thin light but very high-maintenance superconducting one?
I'm trying to picture this setup in my mind. As best I can figure, there is an underground conduit that has a single cable running through it, that they then pump full of liquid nitrogen.
They say it can carry electricity with virtually no resistance, but consider the electricity to cycle the liquid nitrogen and cool it down when it evaporates?
Since it's all underground, I don't see the space saving aspects of reducing nine wires to three.
Can anyone explain the key advantage to this new system? Is copper becoming that scarce/rare that they can't just throw down three more copper cables to increase capacity?
- JoeShmoe
-- I wonder which will go down in history as the bigger failure: the War on Drugs or the War on Filesharing
Resistance R: p = 1.678E-8 Ohms/m for copper
l = 10000m for instance
A = 1.0 cm2 = 1E-4 m2
R = p l/A = 1.7 Ohms
Inductive reactance XL:
f = 50 Hz
û = 50kV
î = 100A
h = 10 m (height above ground)
r = 0.0056 m (wire radius)
u0 = 1.25664E-6 N/A^2
ur = 0.99990 (copper)
L = l * (u0*ur)/(2 pi) * cosh^-1 h/r = 0.016 H
XL = 2*pi*f*L = 5.0 Ohms
Comparing R = 1.7 Ohms and XL = 5.0 Ohms proves your point.
The most practical energy storage system in use right now is pumped storage hydroelectric.
This is used with a hydroelectric generator plant. When demand is low, it will use excess power to pump water uphill into a reservoir; then, during peak demand times, it uses water from the reservoir to generate electricity. Here's a link to one in Oklahoma.
steveha
lf(1): it's like ls(1) but sorts filenames by extension, tersely
I doubt there is any substantial weight savings in the superconducting cabling system. While the superconductor is substantially lighter than the copper, the cooling jacket (we're probably talking a vacuum insulated LN2 jacket) is probably quite heavy.
There are also some technology/safety issues related to the operation of superconductors. A superconducting line carrying a large amount of current can do some pretty catastrophic things if the temperature rises above the critical superconducting temperature. The transition from no resistance to substantive resistance can turn the wire into a nice big heater element inside an LN2 cooled system. Explosive vaporization of the superconducting element has happened in laboratories before. The other problem to be alert for is critical current. Superconductors are only superconducting up to a critical current level. Attempts to pump more than the critical current through the wire will result in it transitioning from superconducting to normal conductivity with the same results as above.
I'm sure the engineers who have designed the system have taken this into account. But the deployment of a crygogenically cooled power distribution system is far from a trivial exercise.
BTW, I've been told that power distribution systems consume almost half of the power generated just in getting the power from the plant to our homes/offices. Also while the superconducting lines can save a lot of energy, it takes a lot of energy to make LN2.
Nah, that's not how LN2 is normally dealt with.
Even inside a good vacuum insulator, the stuff IS going to boil off. Trying to maintain a sealed system will just create a bomb.
Undoubtedly, the stuff will vent to the air around it, and an appropriate amount of circulation will be ensured. I suspect there are fans down there already, so they probably don't need to change anything significantly.
"People who do stupid things with hazardous materials often die." -- Jim Davidson on alt.folklore.urban
Having read a good few responses to this article, I have come to the conclusion that I often do when reading slashdot.
A little ignorance goes a long way.
Come on people, don't post just for the sake of it.
oojah
Do you have any better hostages?
The first link is slide from a Brookhaven talk. Not much useful info here, and the picture doesn't match what the other links describe. The entire slide show is fairly interesting, though.
The second link is PDF whitepaper discussing the commercial production of such cable. A great read, if you have the time to wade through it.
The third link is an article from the Nov. 18, 2000, issue of "Science News" on the same subject as the Knight-Ridder article. Much more technical details.
Nothing for 6-digit uids?
Looking though some EE books I have, the actual power loww of a 10km wire (1/4 inch diamete) is going to be around 4.7% or so. This is actually using methods in Physics more than in circuit analysis, but should still work.
Also, the frequency is a major contributing factor in such line, being the higher the frequency (8khz as compared to 4khz) you get from around 500 watts power loss to 2000 watts depending how much your line amperage is (RMS).
Lastly, though inductors can cause more power loss than the line resistance themselves, you have a assume you have a high variance of power output from a power station. The less the variance, the less that has to do with induction. But, in the cases of power spikes and such, I think I'd take power loss over my house being pumped with 500 amps (when ~200 is what's running my computers, TVs, toasters, etc).
-Wallace
"I am Jack's complete lack of suprise." -Fight Club
Basically the use of stored hydro is to accomodate changing loads. Most power generation systems (especially nuclear) tend to work better if they work at a constant rate. Stored hydro allows you to use excess power to pump water up the hill, which is then used to generate power in times of increased load. That means that the rate of power generation can be kept at the mean power useage (over a given period) while demand can fluctuate.
Ok so some energy is lost, but then energy is lost in all parts of the power generation and distribution system. It is cheaper and easier to run power generators at a constant rate all the time, especially nuclear.
Paul Leader
Really, there aren't any. The stuff is insanely cheap. Like so cheap, you want to start using it as car fuel and stop drinking milk. I purchase 210L for $35.67, which works out to something around 55 cents a gallon!
Nitrogen is cheap, inert, catastrophic leaks have no effect on the world(unless it's in a closed room and someone can't get out before they suffocate), readily availalble (comprises 79% of air), and would only get cheaper to produce as power plants used more of it.
Keeping cables cool is also very easy since LN2 can be easily run through a pressurized system. There is no need to circulate the LN2 since the addition of heat will make some LN2 boil away. Simply allow the vapor to dissapate and replace any lost fluid.
The biggest problem with this project is what happens if the LN2 system fails for some reason. Fortunately, though, they will have an extremely long heads up on a failure and will be able to shut a cable down with plenty of time to spare.
On a side note, the cables use silver because it allows for proper grain growth and flexibility. Otherwise you couldn't make a cable out of the material. A big squarish chunk of it, sure, but not something long, thin, and reasonably flexible like a cable. Science News did an article on it a couple months ago.
www.eissq.com/BandP.html Ball and Plate System. Amuse your friends. Crush your enemies.
Erm, but pure inductance (or a combination of inductance and capacitance, which any transmission line has) is lossless, isn't it?
I think that the main problem with the inductance of long lines is the I and V getting out of phase, which results in less true power being delivered to the load. In addition, heavy industrial machinery is largely inductive by nature, adding to the problem. The power company monitor changes in phase when connecting stuff up, and add power-factor correction capacitors where neccessary.
The problem with saltwater may be true, in that the AC flowing in the line induces eddy currents in the partially conductive saltwater, which then will heat up, ie contributes a loss. I would think that the higher the power and line length the worse the problem.
-- Sig Sig Sputnik
Willy
Going from AC to DC then back to AC isn't the most efficient way of doing things. It is however still done. For example, power is distributed from the mainland to Vancouver island via underwater DC power lines. I believe DC is used here because of the increased effect of inductance with the lines going under salt water.
Using superconducters is great, really, it is... But just because there is basically zero resistance in those superconducters it doesn't mean that all of our problems will be solved. Line losses due to resistance aren't the main loss when it comes to distributing power. There are also losses with the generators, transformers, AC/DC/AC converters and most importandly - inductance. It's a start, not a solution...
Willy
First of all, most of the losses are due to inductance, not resistance (this assumes you're using HV lines - 500kV is typical.) And at 500kV there isn't that much current flowing. 50MWatts just requires 100Amps - very reasonable.
I wish I still had my college books, I could tell you exactly what the losses would be. (I graduated in power systems electronics - this is what we did.) Unfortunately I don't - but I assure you that resistive losses are not the main source of loss from a high voltage power distribution system.
Willy
Ah, a Darwin Award Applicant! Somehow I don't think these things will be 120 volts. A warm spot in the cable could be interesting when the resistance goes up and it isn't superconducting anymore. Third strike is Liquid Nitrogen does not contain Oxygen. I wonder which is first, the Arc, the Explosion, or the Asphyxiation :-)
The truth shall set you free!
Well, it all depends on where in space you are. An object that both absorbs and emits perfectly, put at a distance from the Sun equal to that of Earth, will stabilize at a temperature of about 280 K or 7 C. If it's shielded from the Sun but exposed to inter-planetary and inter-stellar radiation, it reaches about 5 K or -268 C. If it were far from all stars and galaxies, it would come into equilibrium with the microwave background at about 2.7 K. Now, the last is not feasible at this time.
Depending on how it is constructed, you'd have to use a lot less energy to keep it cool then a similar system on the ground, but there you go...
Kierthos
Mr. Hu is not a ninja.
And I can say, with great precision, that the cables will be stolen and sold at a pawnshop.
It's not about increasing power production, but about efficient short term storage.