The Tri-Rail GP49 mod, and where it went wrong


I kind of have a sixth sense thing going on when it comes to engineering. Everything around me gets pretty severely disassembled, at least in my mind. The first time I saw one of these, my sixth sense was NOPEing hard. I was baffled. Why did they do this?

So this is a General Motors Electro-Motive Division (EMD) model GP49. It pulls or pushes things on rails. They were made for the Alaska Railroad and contain a radar guided wheel slip detection/control system for use in icy conditions.

Here, they’ve been made to do a very inappropriate job, and all of us are suffering from the resulting failures.

What’s wrong here? Everything.

First, these units have two cooling fans for the engine, one for the dynamic brakes (resistor grid used as an energy sink, using the motors as generators). Normally, a locomotive of its size would have three.

Above: radiator air intakes. The fans are on the roof below them.

Below: air intakes to cool field installed head end power generator.

Passenger train cars use head end power (HEP) supplied by the locomotive to power lights, air conditioning, cooking equipment for dining cars, etc. This is usually a 480 volt ac 3 phase circuit. The GP49 did not come with a HEP generator so a separate engine/generator pack was added below the radiators and these air vents were popped in to wash cold air over it and keep it from roasting.

Only problem, oops— this allows a lot of air to bypass straight to the exhaust fans without passing through the radiator.

The hot engine alarms and shutdown kick in. A lot.

Currently it is not uncommon for Tri-Rail to run two of these units together, one with only the main engine active, and one only supplying HEP. This reduces the heat load and usually prevents the whole thing catching a fever.


Now you may have seen me mention a radar.

Wheel speed sensor on an F40PH. Sand nozzle at left.

On smarter diesel electric locomotives, there are protection systems in place to deal with wheel slippage. A speed sensor is usually provided on each axle. Analog or digital electronics will test the wheel speed and compare. If one or more axles are spinning too fast, indicating a slip event in progress, corrective logic will activate, reducing the amount of current available to the traction motors by cutting the main generator field current, and possibly applying sand to the rails to make things grip again.

With the GP49, the logic goes a step further. A radar detector actually measures the speed of the train moving over clips and plates on the track.

The idea behind this is that a wheel slip event affecting all wheels can be detected and corrected.

The only problem here is that this system is a little too paranoid. It won’t let the kilowatts flow until it sees the wheels all turning at the right speed to match the radar reading, and it only allows for very slow takeoff.

As of late, either due to updates to the control systems or (convenient?) equipment failure, this seems to be disappearing on the GP49s and they actually…. move.

The overheating problem is still there and very bad, though. Yuck.

How could it be corrected? This would be a tricky one. My best ideas would be to add the third fan, or otherwise increase airflow. Perhaps add a baffle under the radiator and open up the area below to more air for the generator? This one’s gonna be tricky due to the narrow body of the locomotive.

Or just completely reverse the hep modification and place the generator in a separate power wagon? I seem to recall Amtrak having done that.

Train pee. Eww.

8 thoughts on “The Tri-Rail GP49 mod, and where it went wrong”

  1. Used equipment , purchased by bean counters that only see the initial cost as a bargain. Failure to consider the specific modifications to adapt to the Alaskan environment and how the Florida climate fails to match that of Alaska. Government should NOT run anything. The lowest bidder purchasing scheme needs a serious dose of common sense.

  2. Tri-Rail has received an order of new Brookville BL36PH locomotives (though most are just sitting in the yard) and should be phasing out the older ones soon. These are the first loco’s custom-built for Tri-Rail so hopefully there will be less problems and delays in the future! Why they haven’t replaced all of the old ones by now is beyond me…

      1. Sometimes the price of things (locomotives included) gets to be so cheap you can’t afford it.

        This headache is the price being paid for someone else buying sight unseen a device that have no idea about other than the dollar signs.

  3. I kindof stumbled across this a year later and mean absolutely no offense but I thought I should point out – for those of your readers that may be interested in such things and those who stumble in after me – that some of the information in this entry is incorrect.

    These locomotives were not designed for Alaska use; They come from Norfolk Southern Railroad, having been built for their predecessor Southern Railway as GP39X models.

    Super Series wheel slip control was not exclusively designed for icy conditions, but for low speeds where maximum adhesion to the railhead was desired. Previous wheelslip systems only employed a detection of current through the individual electric motors on each axle. The radar system allowed a precise application of more or less power to each axle.

    The engine used on this locomotive is a 12 cylinder engine and is typically cooled using 2 fans to draw air through the intake vents on the side, through radiator cores, and out the top through the fans. A 3rd fan was added to cool only the HEP unit on these.

    The radiator cores are set in a V shape configuration behind the intake vents at the top of the hood. This area is separated by a baffle from the rest of the space below it which houses other things like an air compressor and general equipment like an engine oil filter and water tanks.

    I cannot comment on whether or not these suffer from temperature problems but I can say without a doubt that the reasoning in the post is not the cause.

    1. Thank you! I didn’t know that was incorrect; oddly enough a Tri-Rail conductor had explained a lot of this to me at the time. Then again, he makes the trains go, he doesn’t service or order ’em…

      I suspect the issue was that they did the hep modification initially without adding the third fan, probably by mounting the genset radiator outside the v shaped main radiator. The third fans appeared only recently.

      Being Florida, we get a lot of wet days and the super series slip control probably does a fair amount to combat hydroplaning on the rail head.

      With the hep unit getting cooled by an extra fan, the frequent overheating seems to have ceased.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.