Life on the Rails

[Savonarola]

Heh, it would be a good idea to collect these stories in one place - they're a great read. 

Thanks, Malthus, I took tons of notes about Amtrak and never wrote anything.  I was going to try to get those out first and then maybe I'll try to hunt down the rest of my stories.

[Savonarola]

For some reason this thread had eluded me, after seeing it brought back I've read it for the last week or so, google maps and Spanish wiki at hand and everything, and it has been a most interesting read. Thanks Sav!

Glad you liked it.   

[mongers]

For some reason this thread had eluded me, after seeing it brought back I've read it for the last week or so, google maps and Spanish wiki at hand and everything, and it has been a most interesting read. Thanks Sav!

Glad you liked it.   

Sav good to see your thread hasn't been derailed by a truckload of sugar water. 

[Savonarola]

Rail is a "Fail safe" system; for a train that means that if it loses a system it will drop down in speed until it reaches a safe speed that it can run without that system.  If it loses what is designated as a "Vital" system the train will stop entirely; a stopped train is considered the safest condition for the train to be in.

In terms of a crossing the safest state is an active crossing; that is with gate down and lights flashing.  If you've ever been stuck at a crossing where the gates are down but no train is coming it's because the crossing has gone into an error state; it's received a bad signal, or something has gone wrong with the power, or there's a short on the track, or something else has set it off.

This causes problems; if people see a gate down and they don't believe a train is coming they'll run the gate.  This is incredibly dangerous.  Passenger trains can travel at 180 KMPH in North America and even quicker in Europe and Asia; they can be on you in a matter of seconds, and the train always wins.

One way to stop people from crossing tracks is to use the four post crossing; that is two crossing gates on each side.  The problem with that is if someone does find themselves caught when the gate goes down they aren't able to get out.  For this reason in North America we almost always you the two post crossing; that is one long gate at each side of the road.  This allows someone to simply go around the gate.  (There's also a psychological factor to this; if someone is trapped in between train gates, they tend to panic and usually ram the gate a full speed.  Gates are actually quite flimsy and can be gone through at a slow speed; going through at high speed is likely to cause an accident.)

Yesterday's accident between the Amtrak train and the truck in Virginia is something that can be avoided.  We can put radar at the crossing shelter and detect if there's something on the track.  The problem is that a passenger train is going so quickly at full speed that it requires a full two minutes to stop.  So, once again, psychology comes into play, if people know they have two minutes before the train comes when the gate comes down they're more likely to run the gate then if they only have 30 seconds.  In turn they're more likely to get into an accident by running the gate.  That's considered more likely than having a car stuck on the tracks; so the system isn't implemented anywhere.

[Iormlund]

Rail is a "Fail safe" system; for a train that means that if it loses a system it will drop down in speed until it reaches a safe speed that it can run without that system.  If it loses what is designated as a "Vital" system the train will stop entirely; a stopped train is considered the safest condition for the train to be in.

In terms of a crossing the safest state is an active crossing; that is with gate down and lights flashing.  If you've ever been stuck at a crossing where the gates are down but no train is coming it's because the crossing has gone into an error state; it's received a bad signal, or something has gone wrong with the power, or there's a short on the track, or something else has set it off.

This is how most safety systems work. You've got redundancies for every safety-critical input or output. For example, those big red emergency buttons have two sets of wires running through them, and they are normally closed (so a cut wire will trigger the circuit because the CPU will cease to see the signal). Two different CPUs compute every instruction of the program and then compare the results. If any error is detected the outputs default to a safe state.

This means the most likely cause of accidents is human error/negligence.

[KRonn]

We have train crashes in the US, usually trains going to fast for a turn, on trains that don't have the automated systems to check/alter the speed, or however it works. Some trains do have these systems and trains are being updated but it's taking time.

[Savonarola]

This is how most safety systems work. You've got redundancies for every safety-critical input or output. For example, those big red emergency buttons have two sets of wires running through them, and they are normally closed (so a cut wire will trigger the circuit because the CPU will cease to see the signal). Two different CPUs compute every instruction of the program and then compare the results. If any error is detected the outputs default to a safe state.

This means the most likely cause of accidents is human error/negligence.

That's interesting.  I didn't know anything about safety systems before coming to rail.  The only occupational hazard I had to deal with in cellular was radio hazard.  In the United States the American National Standards Institute (ANSI) sets the safety standards for just about everything.  For radio frequency it's set at 2% of the power needed to cause a burn for non-occupational and 5% for occupational (that is people who are trained to work on radio systems.)  Since radio frequency is poorly understood by the general public the Federal Communications Commission (FCC) has halved those numbers.  Since radio frequency is non-cumulative (for thermal effects; I'll leave it up to the internet to tell you if it causes cancer) I spent a lot of time putting up radio hazard signs and cordoning off areas that were completely safe for people to be in.

[Savonarola]

We have train crashes in the US, usually trains going to fast for a turn, on trains that don't have the automated systems to check/alter the speed, or however it works. Some trains do have these systems and trains are being updated but it's taking time.

Positive Train Control System (PTC); that's what I work on.  All Class I freight carriers are required to have a PTC system installed by the end of this year.  (2020 is probably more realistic; it was originally supposed to be 2015, but Congress extended it to 2018 and since no one is close to being ready they'll probably extend it again.)

A PTC system will apply a penalty brake if the train is traveling too fast in its zone; and will stop a train if its gone outside a block which it has authority to be in.  Alstom's ITCS (Incrementeal Train Control System) will also activate the gates which is why our system allows trains to operate at 110 MPH.  Gates usually operate by a track circuit (the train bridges the track, creates a short and the gates come down.)  With radio a train can alert crossings that it's coming from further away.  Radio (as an example of what I was talking about before) is a non-vital system.  If a train loses radio connection it's slowed to 79 MPH; the maximum speed for track circuitry.

[Admiral Yi]

So what's the right thing to do if you're stuck at a crossing for half an hour with no train in sight?

[Savonarola]

So what's the right thing to do if you're stuck at a crossing for half an hour with no train in sight?

I would find another crossing.

If you're in an area where you know there's only freight then running the gate isn't such a big deal; you can see a freight train coming from a long way away.  Passenger, though, comes up on you a lot quicker than most people realize.

[mongers]

So what's the right thing to do if you're stuck at a crossing for half an hour with no train in sight?

I would find another crossing.

If you're in an area where you know there's only freight then running the gate isn't such a big deal; you can see a freight train coming from a long way away.  Passenger, though, comes up on you a lot quicker than most people realize.

Yeah, that is something that needs to be emphasised.

You can see it at one of my local stations, Southampton; quite a few slow v.long container freight trains out of the docks vs fast non-stopping cross country trains that are through the station footprint in a handful of seconds.   

And there you've the advantage of buildings/objects to gauge a trains speed by, unlike in a flat rural environment.

[Iormlund]

I didn't know anything about safety systems before coming to rail.  The only occupational hazard I had to deal with in cellular was radio hazard.

Safety is a big deal in the automotive industry. On the one hand you've got worker safety. Lots of forklifts, logistic trains, sharp metal bits, presses, hydro and pneumatic actuators, heat sources, fumes, robots and (my specialty) high-powered lasers.
We had a fatal accident recently in one of our US plants. Someone apparently decided that it was a good idea to open a window to a machine without adding it to the safety circuit. A guy crawled through it to do some maintenance and they found him crushed inside.

Product safety is also paramount, of course, and we spend quite a bit of money in it. To make cars is complicated enough. To make safe enough cars is even more complicated. To ensure the cars are safe enough is a daily challenge and all of my processes are safety-critical.

[Savonarola]

Not Half the Fun

There really is a line on the Australian Visa Application which asks: "Have you ever been convicted of a crime."  It's a yes-no checkbox, you really can't write down "I didn't know that was still a requirement to get in."  (There's also a question about where you are applying from, one of the possible answers is "USSR."  News must travel a little slow to Oz.)

I had started working at GE on a project for Fortescue Metals Group (FMG) in the Pilbara region of Australia.  Since I was a contractor when the project was going on I didn't travel to Australia during the build phase; instead I had done the radio coverage analysis from the office.  The system got built; but it has had problems on and off.  We recently figured out one of them; right after we installed our data radio FMG had installed a Terrestrial Trunked Radio (TETRA) system on their train at a close frequency right next to our radio.  The result was that our radios were being blown out by the TETRA radios, causing us to drop data messages.

The solution to this sort of problem is to install a filter.  A filter, in radio engineering, is a device which passes the desired signal and attenuates (or filters out) the undesired signal.  I had got a couple filters and verified this worked in the lab.  The next step was to verify it in the field.  Our home office for this project is in Perth; there are only six engineers there and the one communication engineer is part time.  Harish; who is a signaling engineer, but low man on the totem pole, so he gets sent to the field most often, asked me if I was willing to go with him to the field to test out this solution.  I'm always up for field work, so I started making travel arrangements.

In order to work in rail in Australia you must have a "Rail card."  A lot of education in Australia is vocational and a lot of industries have these sorts of certifications.  We have no equivalent in the United States and no means to get Australian vocational training here; so FMG was willing to bend that requirement for me.  Harish got me put into their system, but in order to be able to be registered for inductions I had to be declared "Fit for duty."

No one seemed quite sure what fit for duty meant.  Harish finally managed to chase down an Australian rail medical guide.  It was 210 pages long.  I sent it to our company nurse.  She immediately called me back and asked "Isn't there someone there who could go?"

Well there wasn't so I got sent to the occupational medical center, got strapped down, poked, prodded, analyzed, blood analyzed, and urine analyzed.  Of course that day I had a terrible chest cold and was coughing so hard the EKG showed that I was having a stroke.  I had to come back the next day and be retested for that.

All the Australian documentation was in metric.  Neither the medical technician nor the doctor could figure out how to convert imperial to metric or vice versa.  They weren't even sure what they were looking at first; they saw the vision was listed as 6/6 and just pondered over it.  I figured out then that was our equivalent to 20/20 vision (6 m is 20 ft.)

In any event I was declared "Fit for duty" and was on my way.  Perth is so far from Melbourne, Florida that it's almost exactly the same time whether you fly east (Melbourne-Atlanta-Johannesburg-Perth) or west (Melbourne-Atlanta-Los Angeles-Sydney-Perth.)  The Australians told me there had been problems with theft through Johannesburg, so I went west.

I had to bring my own test equipment on the flight.  Since it is delicate and expensive to repair I have to hand carry it.  That always leads to the fear that I'll have to explain to a TSA agent what a Vector Network Analyzer (VNA) is.  I didn't have to do that, but I did have it checked for explosives in both Melbourne and Los Angeles.

I had to go through security twice because I had missed my flight through Los Angeles.  There had been a problem with the flight in Atlanta, and we had to wait on the tarmac for over an hour.  We landed just as the flight to Sydney was taking off.  So I got in line to have my flight changed.  There was a Qantas flight that was just about to leave when we landed.  One irate customer ahead of me was trying to get on the flight.  The Delta reps explained to him that federal law prevented any airline from issuing a ticket for international less than one hour before the flight.  He was upset that they couldn't change federal law for him at a moment's notice; and let them know that.

Delta had a deal with a local Quality Inn.  It was well after midnight in LA by the time I got there.  I didn't have luggage, but they gave me a flimsy toothbrush and cheap razor for toiletries.  I got e-mails off to the Australian team and called the hotel in Perth before I crashed out to the sounds of jets taking off and landing.

Things looked brighter in the morning; but I still had all day in Los Angeles.  I still had my test equipment with me.  I didn't want to leave it at the hotel or lug it all over Los Angeles; so I spent the entire day at the Delta terminal of LAX.  It's tiny and I knew every detail of every shop by the end of the day.

I did get searched after landing in Sydney.  They were training a new drug sniffing dog and they put the training case next to my suitcases.  The dog kept sniffing my suitcase rather than the training one.  They went through it; fortunately it was a suitcase with clothes, not my carry on with electronics.  I was able to go through without having to explain what a VNA was to the Australian customs officer.

The Australian security agent complimented me at how well I had put all my electronics into bins to be x-rayed before I got on the plane to Perth.  I told her I had done it a few times already.

[KRonn]

We have train crashes in the US, usually trains going to fast for a turn, on trains that don't have the automated systems to check/alter the speed, or however it works. Some trains do have these systems and trains are being updated but it's taking time.

Positive Train Control System (PTC); that's what I work on.  All Class I freight carriers are required to have a PTC system installed by the end of this year.  (2020 is probably more realistic; it was originally supposed to be 2015, but Congress extended it to 2018 and since no one is close to being ready they'll probably extend it again.)

A PTC system will apply a penalty brake if the train is traveling too fast in its zone; and will stop a train if its gone outside a block which it has authority to be in.  Alstom's ITCS (Incrementeal Train Control System) will also activate the gates which is why our system allows trains to operate at 110 MPH.  Gates usually operate by a track circuit (the train bridges the track, creates a short and the gates come down.)  With radio a train can alert crossings that it's coming from further away.  Radio (as an example of what I was talking about before) is a non-vital system.  If a train loses radio connection it's slowed to 79 MPH; the maximum speed for track circuitry.

Good info and you're working on some interesting engineering items.  When we had some bad train accidents a few years ago I was really surprised that some kinds of systems weren't already in place. It seemed crazy that if a train approached an area way over the track speed limit that there was little or nothing to slow it down. As you say, it takes time and money to install the systems.

[grumbler]

Not Half the Fun

There really is a line on the Australian Visa Application which asks: "Have you ever been convicted of a crime."  It's a yes-no checkbox, you really can't write down "I didn't know that was still a requirement to get in."  (There's also a question about where you are applying from, one of the possible answers is "USSR."  News must travel a little slow to Oz.)

I had started working at GE on a project for Fortescue Metals Group (FMG) in the Pilbara region of Australia.  Since I was a contractor when the project was going on I didn't travel to Australia during the build phase; instead I had done the radio coverage analysis from the office.  The system got built; but it has had problems on and off.  We recently figured out one of them; right after we installed our data radio FMG had installed a Terrestrial Trunked Radio (TETRA) system on their train at a close frequency right next to our radio.  The result was that our radios were being blown out by the TETRA radios, causing us to drop data messages.

The solution to this sort of problem is to install a filter.  A filter, in radio engineering, is a device which passes the desired signal and attenuates (or filters out) the undesired signal.  I had got a couple filters and verified this worked in the lab.  The next step was to verify it in the field.  Our home office for this project is in Perth; there are only six engineers there and the one communication engineer is part time.  Harish; who is a signaling engineer, but low man on the totem pole, so he gets sent to the field most often, asked me if I was willing to go with him to the field to test out this solution.  I'm always up for field work, so I started making travel arrangements.

In order to work in rail in Australia you must have a "Rail card."  A lot of education in Australia is vocational and a lot of industries have these sorts of certifications.  We have no equivalent in the United States and no means to get Australian vocational training here; so FMG was willing to bend that requirement for me.  Harish got me put into their system, but in order to be able to be registered for inductions I had to be declared "Fit for duty."

No one seemed quite sure what fit for duty meant.  Harish finally managed to chase down an Australian rail medical guide.  It was 210 pages long.  I sent it to our company nurse.  She immediately called me back and asked "Isn't there someone there who could go?"

Well there wasn't so I got sent to the occupational medical center, got strapped down, poked, prodded, analyzed, blood analyzed, and urine analyzed.  Of course that day I had a terrible chest cold and was coughing so hard the EKG showed that I was having a stroke.  I had to come back the next day and be retested for that.

All the Australian documentation was in metric.  Neither the medical technician nor the doctor could figure out how to convert imperial to metric or vice versa.  They weren't even sure what they were looking at first; they saw the vision was listed as 6/6 and just pondered over it.  I figured out then that was our equivalent to 20/20 vision (6 m is 20 ft.)

In any event I was declared "Fit for duty" and was on my way.  Perth is so far from Melbourne, Florida that it's almost exactly the same time whether you fly east (Melbourne-Atlanta-Johannesburg-Perth) or west (Melbourne-Atlanta-Los Angeles-Sydney-Perth.)  The Australians told me there had been problems with theft through Johannesburg, so I went west.

I had to bring my own test equipment on the flight.  Since it is delicate and expensive to repair I have to hand carry it.  That always leads to the fear that I'll have to explain to a TSA agent what a Vector Network Analyzer (VNA) is.  I didn't have to do that, but I did have it checked for explosives in both Melbourne and Los Angeles.

I had to go through security twice because I had missed my flight through Los Angeles.  There had been a problem with the flight in Atlanta, and we had to wait on the tarmac for over an hour.  We landed just as the flight to Sydney was taking off.  So I got in line to have my flight changed.  There was a Qantas flight that was just about to leave when we landed.  One irate customer ahead of me was trying to get on the flight.  The Delta reps explained to him that federal law prevented any airline from issuing a ticket for international less than one hour before the flight.  He was upset that they couldn't change federal law for him at a moment's notice; and let them know that.

Delta had a deal with a local Quality Inn.  It was well after midnight in LA by the time I got there.  I didn't have luggage, but they gave me a flimsy toothbrush and cheap razor for toiletries.  I got e-mails off to the Australian team and called the hotel in Perth before I crashed out to the sounds of jets taking off and landing.

Things looked brighter in the morning; but I still had all day in Los Angeles.  I still had my test equipment with me.  I didn't want to leave it at the hotel or lug it all over Los Angeles; so I spent the entire day at the Delta terminal of LAX.  It's tiny and I knew every detail of every shop by the end of the day.

I did get searched after landing in Sydney.  They were training a new drug sniffing dog and they put the training case next to my suitcases.  The dog kept sniffing my suitcase rather than the training one.  They went through it; fortunately it was a suitcase with clothes, not my carry on with electronics.  I was able to go through without having to explain what a VNA was to the Australian customs officer.

The Australian security agent complimented me at how well I had put all my electronics into bins to be x-rayed before I got on the plane to Perth.  I told her I had done it a few times already.

I really enjoy these stories.  Your writing is excellent.