The entire article goes on in great length about how awe-inspiring the
engineering of this cable is, whereas I just see a bunch of complexity that
probably didn't need to be there. The 118 layers of paper tape, and the oil
pumps, and so forth. Surely there was a simpler solution.
http://www.howtospotapsychopath.com/2010/03/10/today-on-surf... Take care that you don't come down with "Engineers' Disease", though - the
tendency for people with a high level of technical knowledge to decide that
their knowledge must be applicable to specialised fields that they don't
actually know a lot about.
You are a programmer, not an electrical engineer. Wild speculation regarding a field you know little about is not attractive.Allow me to reverse your statement.
"I don't see why programming for multicore processors has to be so complex. Why not just run four copies of the program on all four cores?"
"I don't see why every web site makes you create a new account. Why not log into everything with my facebook account?"
"I don't see why the legal system has to be so complex. Why not just have the judge decide everything, and get rid of the lawyers?"
This is actually a good idea.
Message passing is more robust than the currently prevailing shared memory + thread spaghetti practice. Many/most problems being attacked with the latter approach can be reformulated to use message passing and looser coupling.
Actually for many web sites this is a pretty good idea too.
http://www.fujikura.com.sg/hv_xlpe.htm
At the the maximum voltage over XLPE was 110kV.
Sticking 2 of these cables side by side would have other interesting effects.
http://www.emfs.info/Sources+of+EMFs/Underground/multiple.ht...
It appears oil and paper systems are still used, but I don't think they use the high pressure oil system like the one in the paper.
http://www.hitachi-cable.com/about/publish/review/__icsFiles...
In this case, the simplest mechanism would probably have been to build above-ground high tension power lines on massive steel towers. This is the most common mode of high voltage electrical transmission, and the cables are easy to access and repair. However, this cable was probably built when the area was already densely populated, and property owners would not have accepted having such an eyesore going through their back yards. So they had to bury it underground, which required heroic engineering to insure that it would only have to be maintained very rarely (designed for an MTBF of 60 years).
There are analogous problems in computer hardware. For example, if you want a computer to control a cardiac pacemaker, you can't just buy a generic server and throw a gigabyte-sized Linux distribution on it. It has to fit in a tiny space, run for years on a small battery and be 100% reliable. Very different from creating a web application, but the extra complexity and expense are necessary due to the problem that's being solved.
astonishing that they don't do this. Their first attempt failed; the second likely will also. And both at tremendous expense when a tried-and-true methodology exists.
As another commenter mentioned, the engineering involved in the story is likely not required very many places on earth. To be in a situation, developing a solution to a massive problem with little to no precedence is an exhilarating and terrifying position to be in. It may sound overly-engineered to you but, at the time, it may have been the most cost effective engineering solution to a problem rarely encountered.
The issue is that this is 230 KV, vs. 230V.
A single excellent commercial electric wok is about 50-100A.
Each individual circuit is about 15-20A.
The really awesome thing now seems to be HVDC transmission -- eliminates capacitance problems and problems with multiple grids.
This. Imagine a global grid connecting alternative energy sources for each optimal locale: Geothermal in Iceland, Solar in North Africa, and of course Thorium in Norway.
Peter Thiel seems to think this scenario is improbable[1] but I don't think it takes into account ABB's recent breakthrough with HVDC switching[2].
[1] http://blakemasters.tumblr.com/post/23787022006/peter-thiels... [2] http://new.abb.com/about/hvdc-grid
This is about the most arrogant, ignorant thing I've read on HN. No. I promise you the electrical engineer who designed that feeder did not "neglect to realize" that a steel pipe is conductive and if the feeder came into contact with it, it would short. There were no doubt a number of considerations and trade-offs involved in the choice of material.
Tell that to the Texas Department of Transportation, where any given highway project will typically run over time and budget.
Sometimes massive infrastructure requires massive repairs, but that doesn't automatically make it not cost-effective.
This is sort of like the electrical engineering equivalent of the Tacoma Narrows Bridge collapse:
https://en.wikipedia.org/wiki/Tacoma_Narrows_Bridge_(1940)#C...
Like most failures, this is a systems failure. Someone engineered the hell out of that cable. But they neglected to realize that it would be drawn through a steel pipe.
"So, what went wrong? Varying load conditions in the three legs of the 3-phase circuit caused tremendously strong and dynamic magnetic field changes. The electromagnetic forces between the three conductors and the steel pipe (gack!) cause the conduit to wiggle around inside the pipe."
The other lesson is that you shouldn't hyper-optimize locally, and instead design for maintainability. I'm not sure what they were thinking when laying a cable that requires experts to be flown in from the east coast to repair. The simplest mechanisms are the ones that are the most easily repaired. The entire article goes on in great length about how awe-inspiring the engineering of this cable is, whereas I just see a bunch of complexity that probably didn't need to be there. The 118 layers of paper tape, and the oil pumps, and so forth. Surely there was a simpler solution. I'm sure after this failure and the ones on the east coast, they are not still laying cables like this.