Railfan Technology > Train telemetry and railfan radio

One of the railfan arguments made in favor of using a scanner or dual-band amateur radio for railfanning is to monitor "train telemetry" on the UHF 450 mHz band.  Note first that "train telemetry" is not ATCS or PTC; it is the data UHF radio transmissions from locomotives, distributed power units, end-of-train devices, and remote control locomotive packs.  So, just how useful are these radio transmissions to monitor?

First, what are the channels?  I will discuss the six channels that will be of most use to most railfans--I am omitting the remote-control pack channels because remote control locomotives used for switching are generally easily located by other other means.  So, here are the 6 train telemetry channel frequencies with a brief description:
452.93750 mHz      HOTT   Head of Train Telemetry
457.93750 mHz      EOTT   End of Train Telemetry (the EOT or FRED device)
452.92500 mHz      DPU1   Distributed Power Unit Channel 1
452.95000 mHz      DPU2   Distributed Power Unit Channel 2
457.92500 mHz      DPU3   Distributed Power Unit Channel 3
457.95000 mHz      DPU4   Distributed Power Unit Channel 4

All of these channels utilize low power transmitters.  UHF signals propagate best with line-of-sight, so they are generally not readable if there is any physical obstruction (hill, etc.) between the transmitting and listening radio.  As result, the distance over which these signals can be received is generally fairly short.  In my testing, a portable radio (the Uniden BC-125AT, in my case) will generally only reliably receive telemetry signals for about a maximum distance of 2 to 4 miles on flat ground with no obstructions.  Often, I will physically start to hear the whistle or train noise of an approaching train before I receive a telemetry transmission on the radio.  Reception is considerably better on a mobile radio.  In my case, I have both a commercial VHF mobile radio and an amateur dual-band mobile radio in my railfanning vehicle.  I leave the amateur radio scanning the UHF telemetry channels.  In the same environment as the BC-125AT portable scanner, it will generally receive the telemetry transmissions reliably for up to 4 to 6 miles away.  

So, how useful are these telemetry transmissions?  Well, they transmit no railfan radio decipherable data--all a railfan will hear on the radio is a momentary audio "blip" on the channel.  What the blip indicates is that there is a train somewhere within the reception range of the radio.  If the radio has a good signal strength meter, looking at the meter during the transmissions may tell you how close the train is to your location.  Monitoring the signal strength of the several blips in a row may tell you if the train is moving toward you or away from you.  My experience is that, these days,  EOTT and HOTT transmissions are relatively infrequent from trains simply traveling along a mainline.  The good news is that, if a train is utilizing distributed power, there will generally be quite a few transmissions, as the controlling locomotive will send a data "packet" (the "blip") to the DP units anytime the engineer changes a throttle, brake setting, etc. that affects the DP locomotives.  My general recommendation is that monitoring train telemetry with a portable radio is, at best, only somewhat useful, simply because the range that the portable can "hear" the telemetry transmissions is so short.  With a mobile radio, monitoring the train telemetry is more useful.  It can be very useful when one is moving down the road, trying to "sniff out" the location of a train.

Well, then, how to monitor both VHF voice transmission and UHF train telemetry.  If one is not concerned with monitoring NXDN digital radio transmissions, the best best for a portable radio to monitor VHF and UHF is probably the Uniden BC-125AT scanner that I've discussed at length in other posts.  A dual band amateur portable might also be used, but many of these can not monitor both VHF and UHF channels simultaneously.  Also, relatively few amateur radios will tune the VHF analog "splinter channels" created from narrow-banding--this also discussed in many other posts of mine in this forum.  Much the same applies with mobile radios.  Either a multi-band scanner or a dual-band amateur radio is an option, with most all of the same monitoring limitations of portable radios.  The other mobile option is the one that I use and alluded to earlier.  Since I am a licensed amateur radio operator, I already had a dual-band mobile in my railfan vehicle, so, some years back, I simply added a second radio--a commercial NXDN-capable VHF mobile radio--to my vehicle.  I use the commercial radio to monitor the VHF AAR railroad channels, and the amateur dual-band to monitor the UHF telemetry channels.  In many ways, that it the best setup.

I would love to hear about other people's experience monitoring train telemetry.  It certainly may come even more in vogue as voice radio transmissions continue to decrease in the railroad industry.


 



Edited 1 time(s). Last edit at 09/18/21 08:14 by WW.

WW Wrote:
-------------------------------------------------------
> I would love to hear about other people's
> experience monitoring train telemetry.  It
> certainly may come even more in vogue as voice
> radio transmissions continue to decrease in the
> railroad industry.

I've been "hunting trains by FRED" (monitoring train telemetry) for several years now. I know there are others who do this, because another railfan mentioned this strategy many years ago, and I adopted it.

Line of sight or not, I've been able to anticipate incoming trains that I can't see, for example on Donner Pass (UP's Roseville Sub), where I get "5 minute warning" from the "chirps" of EOTD and DPUs. That seems consistent with your mention of 2-4 mile range.

Many of my reports here on TO are result of hunting trains by FRED, at Iowa Interstate, ex-IC (CN) Dubuque Sub, UP's Phoenix Sub and Caliente Sub. Quick TO search says I've been doing this for at least 8 years. I've even compared scanner performance, and my dirt cheap Uniden BC72XLT has consistenly given me "5 minute warnings" while a GRE (Radio Shack Pro 92) with the same antenna was useless, where the train was already visible when I heard the "chirps" and this was at Arizona's open desert (Gila Sub).

Mere hearing the "chirp" isn't enough to guarantee an intercept. Natural intelligence (brain) is required to figure out what to do next. I've done this long enough that I can make guesses on direction and distance (clarity and frequency of "chirps"), and even then I can be wrong.

One obvious location where the "chirps" are confusing is near a yard. Territory with many trains is another obvious circumstance where "chirps" can be confusing. For example, I'd listen to EOTD away from Dubuque Iowa to hunt the less-frequent CN trains, but turned the channel off near Dubuque due to the more frequent BNSF trains on the Aurora Sub.

I'm sure others might want to contribute to this topic, so I'll stop here.

Thanks to you both for this information. I'll have to start practicing this technique, with which I was vaguely familiar, but never really used.

Not only are voice radio contacts becoming fewer, but ATCS Monitor will probably gradually be going away, too. Railroads which used radio-based setups for CTC signal and track occupancy data (most of them except for most of CSX, which uses satellite-based comms) are moving to new frequencies for those datastreams as a result of FCC reassignment action. In doing so they are combining the signalling data with Positive Train Control (PTC) datastreams, and encrypting it. The mix with the PTC datastream of itself means that the data will likely become unmonitorable. But encryption presents an absolutely insurmountable hurdle. It is flat-out illegal to decrypt encrypted communications. Period. Hard stop. Don't even think about it.

BNSF has already completed such projects on the Chillicothe Sub west of Chicago to Ft. Madison, Iowa (the extremely busy former Santa Fe main line). Just within the last few weeks it has become completely non-monitorable for traffic. It was always one of the prime routes in the Midwest to monitor. And as a result, the ATCS Monitor group has decommissioned monitoring servers along that line that dedicated railfans had established over the years. They had forwarded data to a central aggregator server, which sent the aggregated datastream out onto the internet for our ATCS Monitor software on computers, or smartphones via the TrainMon5 app, to be able to view on a track layout diagram. This shift is a process that we can probably expect to continue, and perhaps rapidly. If you are a member of the ATCS Monitor group, as many of us are, there is plenty of discussion of this entire issue, complete with considerable technical detail, on the ATCS Monitor group on the groups.io website.

So, enjoy voice radio, ATCS Monitor and TrainMon5 while you can. Time to start practicing EOT-chirp reading, lining up buddies for text alerts and social media posts, and brush up on the old fashioned skills of signal reading....and just plain old planning a shot, and sitting...and waiting. Tech-driven "run and gun" railfanning is a-changing.

Thanks again for this post and discussion. Looking forward to more comments on EOT "chirp" reading from other members.

MC

Years ago, I stuck my toe in the waters of monitoring train telemetry.  At the time, I lived in an area with very high train frequency, so the telemetry channels just chirped all the time, so I gave up on it then.  Later yet, I lived in area with relatively few trains, but that was "dark territory" dispatched with track warrants issued over the radio.  I didn't need telemetry monitoring because I could almost always get a fair idea where trains were from all of the TWC communication.  Now, I'm in an area with relatively frequent trains, but, as is now the norm, relatively little radio voice traffic.  So, telemetry monitoring is now helpful again.

Another poster mentioned observing signals as a clue for train action.  Unfortunately, that's also getting more difficult as more and more railroads have adopted "light on approach"  signal systems where signals only illuminate if a train is within a few signal blocks of the signal location.

WW Wrote:
-------------------------------------------------------
> Another poster mentioned observing signals as a
> clue for train action.  Unfortunately, that's
> also getting more difficult as more and more
> railroads have adopted "light on approach" 
> signal systems where signals only illuminate if a
> train is within a few signal blocks of the signal
> location.

I've already adjusted to this, and now, I find it "easier", not more difficult. Dark signal = I hunt until I find a lit signal, when I'm confident a train is nearby. Sometimes, chasing after a missed intercept is worthwhile if I find a lit signal within a short time.

Most of my recent reports (UP 1979 on the Fresno Sub, the Z trains on the Yuma Sub) relied on hunting by signals. That strategy was very helpful when I nabbed UP 1996 leading the ZMQLC last December, where I ended up with less than 15 minutes margin to catch my flight. "Hunting by signal" allowed me to take a freeway exit ahead of the train's arrival with minimal wait, instead of overshooting the intercept, or exiting too early and waiting too long (and miss my flight).

My experience may not apply to others, since I primarily hunt (drive around to look for trains) instead of staying at a location for an extended period.

^"Road hunting" for trains can be effective if the highway or road network closely parallels the tracks.  It's much more dicey in areas where that is not the case.  I have both types reasonably close to where I live and it is really easy to miss trains on the lines that don't conform to the road network.  I've been railfanning for about 50 years (and also spent some time in the railroad industry), so I've accumulated a sizable "bag of tricks" to successfully find trains to photograph, but these days I'm having to use most all of those tricks to make for a successful hunt for trains.  On many lines, the "like shooting fish in a barrel days" are over.

WW Wrote:
-------------------------------------------------------
> ^"Road hunting" for trains can be effective if the
> highway or road network closely parallels the
> tracks.  It's much more dicey in areas where that
> is not the case.

Yup. But I still road hunt in dicey areas. I pulled a U turn without seeing the train on the CN Dubuque Sub between Galena and Scales Mound (Illinois), by listening to the EOTD, and nabbed the train at Scales Mound (eastbound M338 manifest). Lit signals at Scales Mount reinforced my "gamble." For most of that stretch, the tracks are not visible from the nearest highway.

There are several sections along Donner Pass that are not visible from I-80, but I've pinpointed trains with combination of tactics, including EOTD (telemetry) and signals. Example of that is my report of UP 1988 leading a westbound manifest last year (July 2020).

> On many lines, the "like shooting fish in a barrel
> days" are over.

I've railfanned with others with ATCS setup, and it IS nice. In one instance, ATCS indicated a "ghost train" which we waited for but never showed up. However, we had enough instincts to react to it quickly enough, and moved on westward to nab our target, an ammo train with CSX units in the consist. That was back in 2017.

Lesson is that rather than tried to aim for certainty, one needs to deal with uncertainty. I don't rely on ATCS servers when I hunt trains alone, and since I don't have a Smartphone, I don't rely on real-time info (like texts from friends) either.

Question - Are other frequencies in use for the head end train telemetry.  Today was trackside next to a train with DPUs, and of course rear end device, and didnt receive anything on the head end to eot device transmit frequency (452 Mhz). 

Thanks.

I was told (but have not been able to verifty) that some EOTs may be tied to Postive Train Control and use those frequencies.  I don't often hear EOTs on trains with distributed power.  

So actually there is a piece of software out there that can decode EOT signals, it's called SoftEOT. I believe it's kind of related to ATCSMon, nevertheless it has its own Groups.IO group like ATCSMon. I've never used it myself though.

Thanks for update, Ill have to try the DPU freqs and see, was only listening to EOT/HOT.