Weather and the Railroad

A couple of years ago, during the first good snowstorm we had while I worked for the railroad, I was asked if the railroad took snow days the way some local businesses do occasionally. I think I laughed out loud. The question was an indirect way of asking if I would have some time off as a result of the snowstorm, which would not happen. More recently I was reading an article online about service disruptions to Chicago's commuter rail system during the "polar vortex" that a large part of the Midwest experienced last month. The article was informative but gave only very basic information, but the commentary by other readers is what intrigued me. One particularly upset reader commented that Metra needed to learn how to provide adequate service and stop using the weather as an excuse every time something goes wrong. Now, I do not use Metra, nor am I involved with their public relations department, but the weather can and often does play a significant role in railroading, although the consequences of the weather may be a little less obvious than other forms of transportation.

Since it is February, and still the middle of winter for half the world, we will talk about winter weather. For most forms of transportation, snow and ice can wreak havoc on roads, runways, and basically any surface over which a vehicle operates, by making that surface quite slippery. The same is true for trains, but since trains are on rails, which serve as a fixed guide way, the results are quite different. While cars might slide off the road, trains will generally stay on the rails, although stopping and starting distances will be longer. When snow gets above the rails, it also can decrease the effectiveness of the brake systems. As the wheels travel over snow covered rails, they warm up, and snow sticks to them. Most trains rely on a brake shoe contacting the wheel tread to slow the train. When released, the brake show is very close to the wheel, though not touching it. The snow that the wheels pick up gets stuck to the brakes as the wheels rotate, and it can build up and create a thick layer of snow and ice between the brake show and the wheel. When the brakes are applied in these conditions, the snow and ice must melt off before the brakes become effective. Typically in these conditions, and engineer will set the brakes periodically while running to help curb the build up of snow and ice on the brakes, but some buildup still occurs.

Snow and ice can also affect the brake system in a very different way if any is allowed to enter the air hoses. A large enough amount can create a blockage in the brake pipe itself, interrupting the normal flow of compressed air through the brake system. This situation prevents the normal application and release of the brakes throughout the train. The portion of the train behind the blockage would not respond to any of the engineer's brake operations. If this is a small portion of the train, it may go unnoticed, but if a significant tonnage is behind the blockage, the results could be catastrophic. The chances of such a blockage occurring can be reduced dramatically by blowing snow out of air hoses before coupling them when the train is first assembled. Doing proper brake tests and inspections also helps to catch this condition before it becomes a major problem.

Brake pipe blockages are one problem caused by snow and ice entering the brake system, but even a much smaller amount of snow and ice can cause problems. It is possible that an amount of snow and ice would enter the brake pipe but would be insufficient to block the brake pipe. However, it could get into the control valves on a car and compromise the brakes there. Each car on the train has a control valve, which is basically the brain for that car's brake system. It responds to changes in brake pipe pressure and applies or releases the brakes on that car accordingly. See our air brake page for a more detailed description of air brake system functionality. Inside the control valve is a pair of slide valves, which each work to apply or release the brakes. The slide valves have holes to allow air to flow from one part of the system to another, as necessary for the application and release of the brakes. These slide valves are fairly small, only a couple inches in diameter and a few inches long. The holes are much smaller, some only the size of the head of a pin. It takes very little snow to plug one of those holes and prevent proper air flow. Depending which hole or holes get plugged, the result could be anything from a brake that responds slowly to one that does not apply, or one that does not release. In a long train, a single car not applying is actually not much of a problem usually, however a brake that refuses to release can become serious. If a car is dragged with the brakes applied, the wheels heat up. If the problem goes unchecked for long enough, the wheels can heat up enough to weaken or even destroy them, which can lead to a derailment. If the brakes remain applied hard enough, the wheels may not even turn when the car is moved. As they slide, the rails will wear flat spots in the wheels. Again, if this problem goes unchecked for long enough, it can also result in derailment. One of these days I'll write about the derailment I had that was caused by brakes stuck applied.

Snow and ice are not the only challenges to the railroad during the winter. Actually, the temperature can be a pretty significant challenge as well, particularly extreme temperatures. Before we get to that, let me explain a characteristic of the air brake system though.

On a train, compressed air is pumped into the brake pipe to charge the air brake system on each car and release the brakes. The brake pipe runs the entire length of the train. On the way, the air passes through steel pipe that is attached to the frame of each car and through flexible rubber air hoses that connect the brake pipes on each car. There are countless couplings, fittings, gaskets, and valves that make up the brake pipe. Each valve, fitting, gasket, and coupling is an opportunity for air to leak out of the system. Leakage is normal, up to a point. Since significant leakage can affect the ability of the brakes to function properly, it is measured before a train leaves its initial terminal. There are limits to how much leakage is acceptable. On long trains, the effect of the leakage can be more noticeable. On a train that exceeds a mile in length, it is fairly common to have a pound of two less air on the rear compared to the head end. A couple of pounds makes no noticeable difference in brake system functionality, and it is nearly impossible to avoid. In the summer, when the weather is warm, leakage is minimal, because everything seals up properly. However, as the temperature drops, all those fittings, couplings, hoses, and gaskets tend to shrink up, and as they do so, more leakage occurs. Rubber gaskets become very hard and do not seal couplings nearly as well as in warm weather. The more extreme the cold, the more extreme the leakage.

In addition to brake pipe leakage, when air leaves the compressor on the locomotive, it is fairly warm. As it is pumped into the main reservoirs and into the brake pipe, it cools off and becomes more dense. Essentially, a volume of air at room temperature takes up much less space when cooled zero degrees. However, if it must fill the same volume, such as a brake pipe, it still fills it, just at a lower pressure. The combination of colder air and increased brake system leakage means that train length gets limited. In the cold it takes longer for the air brake system to charge, and longer for the brakes to release. It is simply impossible to build the same long trains and expect the brakes to work properly when it is extremely cold outside. Just as a comparison, in the summer in Alaska, Alaska Railroad often runs trains that exceed 8,000 feet in length. Temperatures are warm and it takes less than a minute to apply or release the brakes on a train of that size. In the winter, when temperatures in Fairbanks can dip to -60 or lower, it becomes difficult to operate a train that is 2,500 feet long. The brakes respond much slower, and charging the system, or filling it with air, can take hours instead of minutes.

While trains might not be at much risk of sliding off their intended route of travel during snowy or cold conditions, they face other challenges. The nature of an air brake system is such that it operates better in warmer weather. Many of the measures the railroad takes during the winter are preventive. Whether newspaper readers in major cities want to believe it or not, winter weather has a very real and significant effect on railroading, and is a perfectly legitimate reason for making certain service changes.

Second First Trip

At 11:00pm on Friday night, I went on duty for my first trip as a student conductor. Well, it was my second first trip, but the first in Alaska. I would be working the overnight express freight train from Anchorage to Fairbanks. Usually the crew for that train goes on duty at 7:30pm, but the engineer had not been rested, so the trip began behind schedule. We did our paperwork, had a brief about some of the slide zones out there, and then got out the door and on the train.

It was a surprisingly long train for this time of year, at 8,300 feet, but we have also had surprisingly warm weather for this time of year. Usually January is the coldest month of the year, and that cold usually restricts train length to about 3,000 feet. With daytime temperatures well above freezing, and mild nights, the railroad has been able to run much longer trains than usual.

Our trip started off uneventfully enough. We had no speed restricted cars or any cars that required special handling. There were no temporary speed restrictions to think about over the 358 miles, so all we had to remember were several slide zones, but those were five hours away from Anchorage. We all chatted and tried to keep each other awake, and the conductor quizzed me on different rules and operating practices. There was no southbound freight train that night, and we met the coal train within the first 90 minutes, so we had the railroad to ourselves.

As we traveled north, recent rain had made some bluffs near the tracks unstable. This caused the railroad to place some slide zones in service. Those slowed us town a bit once we were north of Talkeetna, but we found everything to be fine in all of the slide zones. Once we started getting into the mountains north of Talkeetna, the snow started to get a bit deeper. There were a few spots where a little of it had slid down onto the tracks, but it was nothing to cause alarm. We crossed the Hurricane Gulch bridge just after 5:00am, and then started down the other side of Chulitna Hill. Just a couple miles later, the trip started to get more eventful.

As we came up on milepost 287, my conductor turned to the engineer and asked if he, in his opinion, thought that using the emergency braking capability of the FRED, on the rear end of the train, would stop a train faster in an emergency than just using the emergency brake valve on the locomotive. As soon as he got the question out, the train went into emergency without any warning. Talk about timing! We came to a stop and the engineer attempted to recover the train from the emergency application.

After a couple attempts, it became clear that the brake pipe pressure was not being restored through the train, and my conductor and I would be going for a walk. We got suited up to go walking through the snow, and grabbed some tools, fusees, air hose gaskets, and air hose retention wires. Then we got off the engine and began walking back to find the problem.

The snow in this area was pretty deep, although the track area had been cleared down to the height of the rail head. On one side of the tracks, the hill dropped off into a river, and on the other side of the tracks, a hill came up from the tracks. We decided to start out on the river side, just so we would be out of the way on the off chance something slid down the hill towards the train. The snow on the river side was above the knees, with a layer of ice on the top. This made walking difficult, because the ice was thick enough that it could not be easily pushed away by walking, but it was not thick enough to walk on top of it. We decided to take out chances on the hill side of the tracks, but that idea was short lived. When we crossed over, we found the snow there to be between waist and chest deep, and it had the same layer of ice on top. So after walking one car length through that, we crossed back to the river side.

After trudging through the snow for an hour, we found the cause of the emergency brake application. A pair of air hoses had separated between two cars. We coupled them back together and the engineer was able to recover the brake pipe pressure. It looked like the air hoses were hanging pretty close to the snow between the rails, so we tightened up the retention wires to get them up off the ground a little more. Once the engineer notified us that he had brake pipe pressure back, we walked back to the locomotive. Despite it taking an hour to get back to the problem, it was only about 20 cars back. The snow and ice had really slowed us down! We were happy it was only 20 cars back, and not 110!

Once we got back to the engine, we got underway again. As we approached the Honolulu siding, the train went into emergency again. This time, the train stopped on a bridge, which had no walkways on it. In January that would not normally be much of a problem, because you could walk across the river below. This January however, it has been so warm that the river was flowing, so walking across it would not be an option. Before we started walking however, the engineer attempted to recover from the emergency application. As we were trying to figure out how we would get across the bridge, he announced he was getting air pressure back, and no one would have to go for a walk. That was a relief! Once he had the brake pipe recharged we got underway again.

For the next few hours, the trip went back to being uneventful. As we got into Denali National Park, it began to get light out. This time of year, the park is pretty quiet, but there were some moose and caribou out and about. We left the park and came along the Nenana River, and headed into Healy Canyon. Just as we started into the canyon, the train went into emergency once again. Again, the engineer attempted to recover the air, but this time he was not successful. Once again, we suited up, geared up, and started walking. This time it only took us about seven minutes to find the problem, and it was the same two air hoses. There was very little snow on the tracks in Healy Canyon, and that made walking much easier!

Since the problem was the same two air hoses, we decided to fix them in such a way that this would not happen a third time! We got some wire off the locomotive, and I had a roll of duct tape. We wired the glad hands together and then tested them to make sure they would not move. Once we were satisfied they were secure, we wrapped the whole mess with duct tape to make sure it would not go anywhere! We let the dispatcher know about the modifications we had made, so that the yard crew in Fairbanks would be aware that they would have to manually uncouple those hoses. By the time we were finished, it was daylight out and the sun was lighting up the canyon. We continued north, although by this time we only had about 90 minutes left to work. The dispatcher told us to go as far as Healy, where another crew would take over, since Fairbanks was still four hours away.

We arrived in Healy just about the same time as our relief crew, and from there we were driven the rest of the way to Fairbanks. We went straight to the hotel to get rested for the return trip. By the time we got to the hotel, it was 1:15pm. Our return trip southbound would normally have gone on duty at 7:30pm, but we were obviously not going to be rested by then!

At 1:30am Sunday, we went on duty for the southbound trip. Once again, the train was going to be late from the start. Just as with the previous trip, the beginning was pretty uneventful. We started with a brief on some of the slide zones we had, although there were fewer than the previous trip. Just like before, we had no temporary speed restrictions and no cars requiring special handling. The train was quite a bit smaller too, only about 5,600 feet long. We got underway and had a pretty unremarkable trip until we got to Healy Canyon.

Healy Canyon is a tight, winding canyon, through which the Nenana River flows. The railroad sits on a ledge several hundred feet above the west side of the river. On one side of the tracks, there is a steep drop to the river, and on the other side is a steep, rock wall that reaches up to the mountains. Because of numerous tight curves, the speed limit through the canyon is 15 mph. It takes almost an hour to cover the distance between Healy and Denali Park, despite being only 12 miles apart. As we were coming through the canyon, we came around a curve and found a boulder sitting squarely between the rails. It was still about a quarter of a mile away when we found it, so the engineer stopped the train short of it. All three of us got out to see if we could move it. It was not huge, about two feet long and a foot across, but it was deceptively heavy! Using a couple bars in the tool closet in the engine, we were able to use some leverage to get it out of the track area and out of our way. Once we cleared the boulder, we were on our way again.

The rest of the trip went pretty smoothly. We had an unremarkable trip the rest of the way to Anchorage, and we made it all the way to Anchorage before we ran out of time to work, although barely. We pulled the train into the yard at 1:25pm, and from there a yard crew took over to break it up.

Winter Railroading

Several weeks ago, I talked about some of the challenges of railroading in the fall.  Since the weather has gotten colder since then, I thought I would share a few of the adventures we have had now that winter has arrived.  It always seems that winter time brings more break downs, equipment failures, and other problems.  It also seems that when things do go wrong, everything happens at once.

For the last few weeks, I have been working as a Conductor, since traffic levels have dropped and they do not need me as an Engineer.  On my first trip working as a Conductor again, we had a pretty normal trip up to Forsyth, but coming back things got a little more interesting.  We got called on duty before the train had actually arrived in Forsyth.  When it did finally arrive, the crew told us that the second locomotive had died on them about a half hour earlier, and they were unable to restart it.  The temperature outside was about 14 degrees, so if we could not get it running again, we would have to drain the cooling water out of it.  Locomotives on BNSF do not have any antifreeze in the water, so when it is cold out, and the engine is not running, the locomotive has to be drained to prevent it from freezing.

When we got on the train, we immediately went back to look at that second locomotive, and see if we could get it running.  We also called BNSF Mechanical on the radio to alert them of the dead engine.  In the cab, the computer said the engine had been shut down due to low cooling water, so that was the first thing we checked. The water appeared to be full.  Actually, it was higher than the full mark, and the person at BNSF Mechanical told us that overfilled was as bad as being low, and we would need to drain some of the water out of it, until it got below the full mark.  So, that is exactly what we did.  I am not sure how much water a locomotive holds, but it is a lot.  It took a while to get it below the full mark, and when it did finally get down to that level, the computer still said the water was low.  We called mechanical back and reported that we had drained some water but the engine still would not start.  He suggested we cycle the computer control breaker, to shut down and restart the computer, which is exactly what they tell you to do every time they cannot figure out what else to do.  When the computer came back on, the low water warning was still there, and the engine still would not start.  We tried cycling the breaker once more, at the instruction of BNSF Mechanical, with no luck.  Since the engine would not start, and by this time the water had already cooled to a temperature below the lowest mark on the thermometer, we decided we had better drain the engine.  We drained the engine, tagged the locomotive as being bad ordered, and then departed eastbound.

For the next couple hours, the trip went pretty smoothly, although the temperature continued to drop outside.  We had a meet at Marsh, about 20 miles west of Glendive, and that would be the last westbound train for quite a while.  The plan was to send us into Glendive after that meet.  We stopped at the west switch, so I could push the button and line us into the siding.  Normally it takes about 30 seconds after the button is pushed for the switch to line and the signal to change.  By that time, I was back on the locomotive, but nothing had happened.  We figured the switch must be frozen, because the last detector had said it was minus seven outside.  The switches are equipped with propane heaters, but they only come on when moisture is detected.  Since it had been dry most of the day, we figured the heater was probably off.  I grabbed a bottle of water, to dump on the heater and turn it on, and went out to line the switch by hand.  It had filled up with snow somehow, probably from the wind, and since the heater had been on earlier, but then shut off, causing all the snow that it had melted to refreeze and coat the entire switch in a layer of ice, beneath the snow that had blown in.

While I was chipping ice off the switch, something weird happened.  The Engineer kept flashing the headlights at me.  Normally, at night, that would be to let the Conductor know that we had gotten the signal into the siding, but since I had the switch on hand, and the points were floating while I swept it out, it was impossible to have the signal.  I was a bit confused, but got the switch cleaned out and lined for the siding, and then gave the Engineer a hand signal to proceed.  He stopped on the switch, and I put it back to power and locked it all up, and hopped back on.  When I opened the door to the cab, it smelled like something was burning in there.  He had the headlights off, but since our meet was sitting right there, that was not unusual.  As it turned out, they were not off by choice, and that had to do with the flashing earlier.  He had tried to turn the headlights on so that I would have more light while cleaning the switch.  When he did that, the headlight switch shorted out, which resulted in the burning smell that filled the cab with enough strength to make anyone gag.    He demonstrated this to me several times.  Turning the headlights on made them come on for about a second, then go out, with an accompanying crackling sound from the control stand.  We used a couple of flashlights to see ahead of us in the siding, and let the dispatcher know that we had no headlights, and until we could get some, we could not leave Marsh.  Seeing as the second unit was dead and completely shut down, we could not move that up front and use it, because the headlights would not have worked on it either.

We decided the idea situation would be to move the distributed power locomotive, on the rear, up to the front, and just use that as the new leader, because it was the only locomotive we had with everything working.  However, being as it was seven below zero outside, cooling off, and windy, neither one of us wanted to walk back to that locomotive, which was over a mile away.  The dispatcher agreed that walking was not a good idea and told us he would see about getting someone out there to give us a ride.  The only problem with that is that Marsh is 45 minutes from Glendive, by car, on dirt roads, and since it had snowed the night before, it was somewhat questionable whether Marsh was even accessible by road.  Eventually the dispatcher told us that one of our Trainmasters was on his way out, from Miles City.

Nearly five hours later, we saw headlights on a vehicle, coming our way.  For the last four hours, we had been wondering where the Trainmaster was, and if he could even get to us.  Turned out he had a story of his own.  He had originally tried to get to Marsh, by way of Fallon, and found snowdrifts as high as the truck he was driving.  From there, he turned around, drove to Glendive, and then tried to backtrack to Marsh.  He told us that he got a few miles out of town, to where the road turns to dirt, and the low tire pressure light came on.  He got out to check on the tires and said he could hear air leaking, so he turned around and drove back to Glendive.  When he got to town, he left the truck and borrowed the yard van and driver, and headed back out to us.  When they arrived, we hopped in the van for the ride back to the DP locomotive, and briefed the Trainmaster on what had happened so far.

When we got to the DP, we found that there was no FRED on it.  Technically, trains operating with distributed power on the rear, do not need a FRED, because the DP locomotive can monitor the brake pipe pressure and serve as a marker.  However, when a DP can no longer run on the rear, such as in our situation, there is nothing to monitor the brake pipe or serve as a marker, unless there is a FRED on the train.  At any rate, we needed headlights more than we needed a working engine on the rear, so we took the DP off the train anyway.  I managed to find a red flag on the DP locomotive, so I stuck that in the coupler, to serve as a marker.  We would be limited to 30mph without a FRED or device to monitor the brake pipe, but we only had 20 miles to go.

When we got the DP off the train, I had to clean out the west switch again, because the heater had turned back off and everything had frozen.  I have no idea how cold it was by this point, but it felt a lot colder than the first time I had cleaned that switch out!  Once we got out, we ran alongside the train on the main track, and then put the DP on the head end.  Fortunately, before the Trainmaster had arrived, I had gone and checked out the east switch, so it was thawed out and working properly for us when we got down there with the DP.  We put the DP on, did our air test, and headed east, finally!  When we got to Colgate, which is the last siding before Glendive, we had only a few minutes before we hit out 12 hour limit.  As we went over the west switch, we got a call from a dog catch crew, who was on their way out to meet us, there at Colgate, so we stopped at the crossing there and waited just a couple minutes for them.  I guess BNSF decided they did not want to fines associated with making a crew work over 12 hours!  Dog catch crews are significantly cheaper!

We hopped in the van and that ended an eventful day for us.  Most crews were getting from Forsyth to Glendive in well under six hours, and it took us over 12, so we were quite happy to be home.  And the thermometer on the bank in Glendive said -12 degrees out, so it certainly had cooled off more since the first time I cleaned out the west switch at Marsh!