Railway Signalling

I’ve long been interested in railways. Not because I’m a “foamer” (UK parlance — apparently some people foam at the mouth when they get the chance to watch passenger trains move, or so the railway employees would have it) or a “railfan” (the US term — Is that supposed to be like “sportsfan”? I mean, just because I want to take a photo that has a train in it doesn’t make me a weirdo, does it? Apparently), but for the same reason that engineers tend to interested in almost everything: how does it all work?

Model of a searchlight railway signal
Not bad for a model railroad!

One part of real-world railways that is fascinating is the signalling necessary to make operations safe and efficient. It’s beguiling to an engineer in no small part because, by design, you can’t infer the behaviour of the entire system just watching the signals that go by as you’re on a train: automated signalling isn’t just about local conditions, but about the relationships between track conditions and the locations of trains across vast distances. The relevant Wikipedia pages have never been much help, either. As an unrequited model railroader, I’ve seen plenty of articles about modelling signals, and even descriptions of CTC machines and Train Orders, but still precious little about how signalling systems, as a whole, work. So I’ve long been curious.

A few days ago I came across a fantastic reference by one Carsten Lundsten about how signalling is done in North America. I’ve been engrossed. It appears the site was written somewhat for a European audience, but as far as I can tell it’s pretty informative for a Canadian and American one, too.

Rather than blathering on about which rule number a signal represents or what speed limits are, these documents concentrate on how signalling systems protect trains and how they have improved over time to provide greater automation and flexibility. If you’re interested, definitely start with Basics of North American Signaling and Safety principles.

Absolute Block Control, double track, through to Centralized Traffic Control, double track
Want to know what this means?

Be sure to make your way through to the page about Absolute Permissive Block signalling — by far and away the best explanation for APB and how APB is different than ABS I’ve ever seen, and I’ve been casually researching this for years.

Opposing signals clearning in section of main track under control of an Absolute Permissive Block signalling system

Enjoy, all ye “train geeks”.


Model photos from Richard Stallard’s site about his Marbelup Valley railway.
Signal plant diagrams from Carsten Lundsten’s site, as above.

Poisoning DNS perhaps a bad idea

This is insane. I’m sitting at a café in Sydney using their hotspot. Went to search for something, and I kept getting strange looking “site not found” pages. Huh? Thy were working a few hours ago. So I started digging.

The café’s upstream ISP is “Optus”, one of the major Australian carriers. To my astonishment I found that Optus’s DNS servers are interfering with Google searches, stealing their DNS lookups and serving results pages on their own (shitty quality) branded search instead! Try https:? No connection; and Google+ wouldn’t load either.

Obviously as soon as realized what’s going on I immediately changed DNS servers to something reliable. Before I did I found a tiny “about this page” link at the bottom of the heinous Optus search results page, where I was told how great this was for me, but how I could opt out of their “default” search engine if I wanted to but was warned this was an “advanced setting”.

Seriously, what do Optus think they’re doing? From a commercial standpoint, do they really think that their captive audience matters to anyone advertising on the web? Of course not, but in the mean time they’re certainly going to alienate customers who just maybe actually do want to use (in this case) Google sites.

There’s a bigger issue, though. Unaltered answers to DNS queries is a backbone of net neutrality. That’s our problem, but once carriers start poisoning nameservers in their own favour it will be but a blink before everyone is doing it to each other and lookups will become worthless. While I’m sure the morons in Marketing who thought that sabotaging DNS queries would be a good idea won’t be worried about the wreckage that will cause for everyone else, such a war wouldn’t be good for any of the companies involved, either. And meanwhile, if they really want everyone to learn how to install an app to “fix” the internet…

Of course, this is only a taste of what we’ll be in for when the communications minister finally gets his compulsory Great Firewall of Australia censorship in place, but one thing at a time. If you’re looking for internet access down here, clearly Optus or anything that uses their network should be blacklisted.


DART data

A couple weeks ago an earthquake in south-western South Island, New Zealand, triggered a tsunami. The alert text put out by the Australian Bureau of Meteorology was:

    An undersea earthquake of magnitude 7.9 at Latitude 45.960S
    Longitude 166.470E occurred at 07:22 pm EST on Wednesday 15 July
    2009 near OFF W. COAST OF S. ISLAND, N.Z.. Sea level
    observations have confirmed a tsunami has been generated.

“Alert” is actually a bit of a misnomer; it was in the weather forecast. I only accidentally noticed the warning when I happened to check GNOME’s trusty little weather-applet wondering whether it was likely to be sunny the next day. So it’s great that I saw the alert there, but, if there’s a warning then I think gnome-applets needs to freak out a little to get the user’s attention. Maybe a different current-weather icon in the panel? Or better yet, a libnotify popup (“run for the hills” is a little more important than “your battery is low”)?

Anyway, as all this was going on, I was curious about the obvious question: “how do you detect such a thing a tsunami wave in the deep ocean?” One of the things about such energy is that although real tsunami become disasterous (ie huge) when they hit shallow water, out in the deep ocean they are not much more than a tiny ripple. Hard to notice amongst swell and chop.

Various national agencies have tsunami detection instruments floating around out on the high seas. These graphs show the data as gathered from buoy 55015 on the evening of 15 July:


which doesn’t appear to mean much, until you realize that the monitoring systems switch to transmitting high-resolution data when they detect an event; the longer time base data around the event above looked as follows, and suddenly it becomes clear that they noticed something abnormal:

Snapshots taken from the National Data Buoy Centre real time data page as presented on 15 July 2009; if you dig around you can find historical data there too.

It would appear that general sinusoidal trend line on the above is normal variation; sure enough if you look at today’s 15-minute data,


it’s pretty clear that normal measurement is tidal variation. Huh.

Column height

Anyone who has spent time out on the water knows full well that the surface is insanely variable; that got me wondering how they measure the waves going by (more to the point, how do they notice a tsunami wave a couple centimetres in size out amongst the randomness of wind driven chop and ocean swell?).

Well, it turns out that these at-sea buoys are not taking soundings from the surface as they bob around. The instruments are placed on the sea floor (!) and measure the hight of the water column as inferred by the water pressure.

Image from the NOAA National Data Buoy Centre page about DART buoys.

Note to super-tanker drivers: please don’t hit these. Thanks.

Never cry wolf

Thankfully, this earthquake and the associated ocean wave it generated did not result in a destructive tsunami when it reached the coastlines in the region. The fact that the authorities raised a high-profile alert for what turned out to be a non-event, however, raises the risk that the next time there is a tsunami wave detected the warning will be ignored by the news media.

The essential message remains, though: detecting waves is one thing (and impressive!). Forecasting destructive potential is another, and that part is still a very grey area. Hopefully people will accept this.


The Great Storm of 1703

Eddystone Lighthouse

Looking into disaster scenarios and doing actuarial and engineering forecasts of potential impacts, I came across a fascinating paper about the impact of “The Great Storm of 1703“¹ on southern England and the Channel coast. Obviously this predates modern meteorology, so what makes it interesting is how they modeled what the wind forces likely were.

Given that there is every likelihood that such conditions can arise again, the forecast of the economic impact (specifically, claims in excess of available reinsurance) means such and event would likely have a catastrophic effect on the financial system if it had already been weakened by other difficulties — kind of like as it is at present.


¹ A 7 page retrospective published in .pdf form by a firm named Risk Management Solutions at their website. The fate of the Eddystone Lighthouse, pictured above, is also interesting.

Innovative uses of bobcats

Ordinarily Rail car movers (aka HiRail trucks) are either enormous ugly yellow painted beasts (which are inevitably maintenance nightmares) or are glorified pickup trucks (which, while great for zooming around to fix a remote signal, that haven’t much hope of actually moving anything). This week, though, I came across an unusual approach to dealing with moving small numbers of cars in an private industrial rail yard that might do the trick. A Canadian company called Brandt Equipment markets a vehicle called the “Rail Yard Boss” which is nothing but a John Deere front loader adapted with rail wheels and a coupler. Photos from their website (I don’t live where there is so much grain, you know?):

Rail Yard Boss, front quarter view

The interesting part is that it uses its hoist to transfer some of the weight from the car its pushing to itself, thereby increasing the friction between its wheels and the track, improving its traction as a result. That’s smart.

Rail Yard Boss pushing a string of empty hoppers

The advantage over a real switcher, of course, is that it can just up the steel wheels, pivot, and drive out of the way.

Rail Yard Boss, front quarter view

On the other hand, given that this thing can only push a max of 5 loaded cars, you’ll probably end up needing a real switcher sooner or later anyway (and what yard doesn’t have a spur as a switching lead to get things out of the way?). Still though, pretty neat.


On Bridges

Doing some background research for a client I came across a surprisingly comprehensive introductory page about the basics of bridge design. This isn’t Civ Eng stuff by any means, but it’s a great overview of the taxonomy of bridges, with some excellent illustrations:

Through truss
Pony plate girder
Pratt truss

along with admirably concise descriptions.

The source website is an exhaustive survey of bridges in and around Pittsburgh, Pennsylvania, USA, maintained by one Bruce Cridlebaugh. Though I am sometimes moved to wonder what drives people to such exhaustive efforts, it’s not like people I hang around with aren’t equally obsessed over obscure interests. :)

As websites go, it’s actually surprisingly well laid out. The intro page has a graphic explaining to visitors how to use their site:

How to use pghbridges.com in the form of a series of step by step images

“…To choose from a comprehensive listing of structures, (1) go to List by Location, (2) choose a map area, then (3) select from the list for structures in that area…”

Granted, enthusiasm to use this technique oneself is tempered by the obvious hassle of having to create such a graphic in the first place (not to mention that screenshots have a bad habit of getting out of date the moment you alter your stylesheets), but it is nonetheless impressive information density.