Slashdot Mirror
The Fiber Age Meets The Power Grid
tulare writes: "According to this story at Wired, a research team is developing a way to replace the steel core inside high-capacity electrical power transmission lines with a fiberoptic core, which apparantly could provide a dual benefit: a 200% increase in emergency transmission capacity along with the ability to "carry several gigabits of data per second." (Per line?) There are a few kinks to work out - like how to splice the data in and out of the lines, but the story talks about an initial rollout date in 2003. Not soon enough to bail Californians out of the current crunch, but considering the benefits (less line sag, greater capacity without building new towers/routes), the effort certainly seems worthwhile." There's some more info from the researchers at this site as well.
The fiber cabling does not carry the power, they support the aluminium cabling which carries the power (previously, steel would support the aluminium lines)
This "new" technology is about 20 years old. Various utilities have been doing it for a long time, mostly using the static wire (the thin wire at the top of the tower that is used primarily for lightning protection). There are problem other than the obvious ones already mentioned, the big one being that when the static wire (or conductor in this case) is hit by lightning, the thermal shock tends to shatter the fiber (unless the conductive element is just melted, in which case the fiber breaks since it isn't strong enough to carry the two ends of the conductor).
For this reason, utilities have preferred to use their ROW to bury the fiber, rather than string it up on the towers.
If the researchers are claiming that an electric utility can achieve 25% better efficiency by exchanging more data, I have bad news there: utilites have been very heavy users of data processing at every level of the operation since the 1920's. The utility I used to work at had many joint projects with IBM in the 1950's and 60's, in fact, due to their heavy volume of transaction processing. A lot of new stuff (like check scanners) in the computing world was driven by utility requirements.
This doesn't even scratch the surface of the utilities' power flow management efforts. So I really doubt there is much in the way of SCADA that hasn't been thought of by now.
sPh
I like it. The Echelon guys would be in for a nasty shock when they try to splice that fiber...
OTOH, the next thing you know, California's internet fees go up to $1900 a month, compared with $30 for the average user.
Kevin Fox
--
Kevin Fox
The NY Times had a pretty good article about this recently too.
Spelling: aluminum, steel. Yeah, I know, you probably think spelling doesn't matter, but misspelling the core terms in what you're talking about makes you look like an idiot.
As the article you obviously didn't read thoroughly enough points out, aluminum is not stronger than steel in the way that matters. Pound for pound, aluminum has a 4-5% higher tensile strength than steel. However, the pound of aluminum will have a much greater volume, which means a wider cable, which means greater stresses from wind etc. and from ice in colder climates. Aluminum is also notoriously brittle, and has a smaller difference between yield vs. ultimate tensile strength. In other words, it will break where steel will stretch, and again the difference becomes even more important at lower temperatures. In conclusion, then, while aluminum does have advantages over steel for some applications, it is inferior to steel as a load-carrier for power lines.
It would actually be interesting to see the same sorts of comparisons between steel and the proposed glass fiber. Some kinds of glass have amazing tensile strength, but it's not clear whether those kinds are compatible with data transmission and glass in general is even more notoriously brittle than aluminum. It's likely to be far more complicated than "X is stronger than Y".
Slashdot - News for Herds. Stuff that Splatters.
After carefully reading the project page link, I think the author over at wired missed the point entirely. FiberOptics are glass -- low tensile strength, while the core of a transmission line's primary purpose is to provide load carrying capacity in tension. Thus, using fiber optic cable for a transmission line core makes no sense.
On the other hand -- recent advances in carbon composites have resulted in some amazingly good fibers with strength/weight ratios that are hard to believe. Replacing the steel core with carbon composite fibers would allow more current carrying aluminium in the same diameter cable.
The splicing question now makes much more sense -- splicing metal cables is a simple mechanical proposition -- not so with joining composites.
So yes, soon, all of California will be addicted to CRAC.
Sometimes, you just can't make this stuff up.
I use to work in the test lab at Burndy Connectors, where we tested connectors for these types of connections. Most of these (99.9%) are compression, or crimp connectors.
... as the link in the original article explains this is one of the many hurdles...
... It could grab a cable and pull both ends with up to 100,000 ft-lbs of force ... and we got up into that range testing these types of connectors. We had big shields to stand behind during these tests ... as the device (cable & connector) under test would/could send stuff flying when it pulled apart.
.2 seconds in duration. Two pulses withing a couple of minutes would raise the temp of the cable to over 100C and turn it black.....
I would think the fiber would have to be quite sturdy to withstand this type of compression
I remember our test (pulling) machine
I remember the crimp had to crimp enough to really grab the steel core (to provide 95% of the cables rated tensile strength). A really good design (connector & crimp tool) could actually exceed the cables rated strength.
Now for some real fun, we use to test grounding grid connectors. Imagine a 10 meter circle of 2500mcm stranded copper cable (about 2.5 inches in diameter; with connectors every 3 meters. We would hook it up to a huge power source (usually a sub-station) and pulse high currrent thru it [I dont remember exactly somewhere around 50,000 amps, but I remember it was in the 5 to 20 megawatt range. The pulses were
-- www.globaltics.net
Political discussion for a new world
California's problem is more of an infrastructure problem based on contradictory laws, many of which seek to avoid the consequences of the laws of nature. Mix that in with corporate and civic oportunism, and there is enough blame to go around to tar and feather everyone.
They have run into the classic "Pick two out of thre problem": Cheap, Reliable, Easy/Fast
They want to have all three, and it isn't there.
Check out the Vinny the Vampire comic strip
"It is a greater offense to steal men's labor, than their clothes"
-
"The nice thing about this is you don't have to replace the whole line, just the lines with sag," Rodriguez said. "For a few hundred thousand dollars you can fix an existing line in a few weeks time so it wouldn't be a bottleneck."
So they're only replacing some of the lines, wouldn't that mean it would be impossible to transmit data, since only some of the lines are being replaced?Still, it provides a roadmap for the future for full conversion, I just think that 2003 is probably way too early to expect data transmission
You crazy man? You piss off supahfly!
Same reason they don't use gold or platinum... Wires used to be copper, though, before economics changed the laws of physics.
Anyone know if aluminium runs hotter (i.e. more transmission loss) than copper? That wouldn't be a problem for utilities, they'd just crank up the meter price to cover it. ;)
If you were blocking sigs, you wouldn't have to read this.
Maybe you should take the time to read the article. There's an optimum cable diameter above which wind and ice becomes a hazard. The can't make it thicker, but they need to get more current carrying aluminium in that diameter as possible.
I do agree with you about the data though, it's more of an "Oh gee, I guess we could do that," sort of consideration.
Anyway, they'd be better paying the up front cost and burying the whole damn lot. Then they could make it as thick as they like, and lay some nice new fibre in there while they're at it. No, wait, that would require a long term viewpoint, like thinking 2, maybe 3 years into the future... ;)
If you were blocking sigs, you wouldn't have to read this.
If you're going to run optical fiber along a power cable, it would make more sense to replace one of the outer aluminum strands with a jacketed bundle of fiber. That puts the fiber right where it's easy to work with, instead of in the structural center of the cable beneath the conductors.
--
Scientists restrict study to entire physical universe; creationist
Imagine a conductor going in and out of your screen, in the middle. Now assume an increasing current through that conductor, going into the screen. Since the magnetic fields from a conductor form circles around it (by the right-hand rule), you'll have a magnetic field going clockwise around the conductor.
You can assume that this current flows only on the very surface of the conductor, but that would imply an arbitrarily small depth and a rather large resistance. This is pretty obviously not the case in reality, so it's worth analyzing the situation to see what really happens. If you have a step-function increase in current along the conductor, you'll have a lot of current flowing in the surface layer, a big magnetic field around the outside, and a smaller current in the bulk of the conductor with a smaller field there. The bigger field tries to penetrate the conductor, along with its associated current. It can't do this all at once; as the field flows into the conductor it sets up eddy currents like smoke rings blowing down a pipe. These currents flow in the forward direction (the direction of the change in current) on the outside and in the reverse direction (against the change in current) on the inside. The eddy currents have to fight the resistance of the wire, and they decay exponentially with time. After a few time constants, the current is flowing pretty much evenly through the whole wire.
As you can see, if the time-constant of the eddy currents in the wire is a lot smaller than the frequency of the power in the wire, skin effect will be pretty small. The construction of the wire has an effect, too. Since the time-constant of the eddy damping is a function of the thickness of each individual piece of conductor, winding strands in thinner shells will reduce the skin effect. The trapezoidal arrangement of conductors in the aluminum-clad-steel wires may be designed for this purpose (or maybe it was just a convenient way to squeeze more aluminum into the cross-section than a single layer of pie-shaped wires would have been; my guess is, a little of both).
--
Scientists restrict study to entire physical universe; creationist