Elsewhere in the news:
Relay Room of London Tubes:
Computer problems at Grand Central Terminal
If you put all your eggs in one basket and for some reason the basket can not accept any eggs you are in trouble. If you don’t have another basket to put your eggs in then you have to wait.
Now if the power supply did fail to the computer system it is a failure to properly back up the system properly.
One other item. – This extended cold spell seems to be causing many failures not just of the computer. As our posters in northern climes can testify cold breaks things faster.
Cold can raise hell with things just as much as heat can.
And Blue Streak’s 100% correct about the folly of putting all your eggs in one basket. There’s too much of that going on nowadays, especially “digital” eggs.
Mark my words, one day there’s going to be a “digital Titanic”, and it won’t be pretty.
But firelock when it happens you will not be able to gloat to the rest of us ! !
Do you think it will be a “CME” ?
Last nights incident at MNRR was because a person, a human being, pulled a power supply to change it out, did not look for the back up or was not aware that there was one plus did so at peak transit hours. It was human error, not technical or mechanical. Power remained to the trains which were ordered to run at restricted speed to the next station and stop. It was the power to the dispatchers board(s?) that was disrupted turning the boards off and the signals dark. Evidently MNRR has dispatchers and managers who had the experience and ability to think through the problem to move trains to stations and give commuters the chance to leave the trains for other modes. I can think of another operation which lost its signal system recently and just left all trains sitting at blank signals wherever they were until the signals came back on.
Henry: Did the signals go dark ? All the signal systems I have heard of still operate with a DD failure. The dispatcher’s desk would have to clear signals at CPs. Are intermediate signals automatic on MNRR or do they have to be cleared by setting up a route. ? One thread I read is that signal maintainers went to some CPs to clear trains to the next station much like Amtrak has on its portion of the NEC. ?
Looks like there were quite a few people there! Not good for Metro North.
I think that if the power was cut, those blue screens of death are photoshopped…
When it comes to moving trains -
Protect, then authorize!
When the computer aided signal system crashes - so has the protection.
Of course the next refrain is - run it like they did before computers!
There were various ‘manual tools’ that were in place and used constantly to facilitate the operations prior to computers. Since computers became the critical element, all those manual tool have been done away with…they cannot be recreated and used ‘on the fly’. Without the computer being involved in the lining of signals, protection of the trains (and their occupants) is seriously comprimised.
MNRR issues a detailed explanation of the causes. It also faults itself for starting a replacement of a power supply during rush hours instead of a weekend.
Interesting — it was a loose wire that caused other power supply to crash.
http://www.mta.info/mnr/html/mnr01_23_2014_shutdown.htm
any one think NJ Transit would have been as forth coming ?
No. We know NJT is not capable at the moment to do so if only because it appears to be under the direct control of Governor Christie’s office since the Sandy debacle.
And from all news reports and MNRR website announcements, yes, it appears the signal system went dark thus trains were instructed to continue at reduced speed to the next station. While the overhead and third rail were energized and the trains had power so there was heat and lighting aboard. I am not sure by any of these reports whether or not the switches were operable, though.
Metro North has no wayside block signals, only home signals are left. All block signals are via in-track encoding (DCC ?) with cab indication, and yes , the switches must be thrown by the computer.
ROAR
MNRR has no wayside signals except at interlockings. Signal system is cab signal only. So, that explains the lack of signal in that they are controlled by computer and induction of some kind. Question is, were the home signals dark or just the cab signal system?
I won’t be gloating brother. It’s a prediction I don’t WANT to come true. But the way things are going and people seem to be making themselves more and more dependent on the gadgets, well…
By the way, pardon my ignorance, what’s “CME”?
Read this Wikipedia. Go down to history. Not mentioned is how much problems occurred to the budding telegraph system. IMHO telegraphs were much more robust than solid state electronics of today. Only equipment in a fully shielded faraday cage may be protected. Depending on its speed some protection can be implemented… But the electric grid ? ?. aluminum foil completely around an electronic device might protect it.
http://en.wikipedia.org/wiki/Coronal_mass_ejection
A past national geographic article a long dissertation on them.
In the history section of wiki read the Carrington link. Its scary
Also Analog magazine had a science fact article on it.
Coronal mass ejection (from the Sun). Acts like a slow-rising EMP (that’s electromagnetic pulse); risetime in about the 10-second range. Fast-acting isolation can work nicely, and Faraday-cage approaches can keep the charge/voltage away from the device traces. Problem with modern semiconductors is that some of the internal vias act as antennas, and neither the critical voltage nor critical amperage to cause damage to the microcircuitry is very high.
The principal disaster that was supposed to come from one of these is damage to the power grid, for reasons related to obsolete transformer design and slow, foreign-sourced transformer replacement. (This issue has been covered not only in the press but also in sensationalist thriller literature!). Much of the problem has been quietly solved, in my opinion similar to Y2K, in the years since the trouble was first identified…
Coronel Mass Ejection! So that’s what CME is! Thank you gentlemen, it’s a wasted day if I haven’t learned something new. And up till now if someone mentioned “coronel mass ejection” to me I would have thought it referred to throwing a cigar butt out a car window.
I was only marginal on science in school, that’s why I got into history.
And what could cause a “digital Titanic”? CME possibly, but I’d be more concerned with cyber-attacks. Just when you think you’re safe, along comes some gnome somewhere to prove you’re not.
Look at what happened to Target.
I wonder when, or perhaps if, the backup system was tested. The operation appears sloppy enough that I think it possible no routine testing occurred. A backup system that isn’t tested might just as well not exist!
One system was “disconnected” for service or something, and the people shutting down the other system did not know that fail-over was not working.
ROAR