Friday, March 19, 2021

Fun with Ladders

 Not step or extension, but an array of railroad switches.  So, all my non-railroady friends can return to their regularly scheduled program...

I started building my model railroad 35 years ago.  One of the first parts was a flat classification yard.  I hand laid the track, including nearly all the switches from kits.  They had the points soldered to a solid metal throw bar and relied on the points to route power to the frog.  Standard ground throws did not provide enough throw and force to make reliable contact.

If you add in that lining a route on the yard ladder can require throwing several switches to my propensity to run through misaligned switches, then the chance for failure it rather high.

So, I decided to replace all the ground throws with Tortoise switch machines and use a rotary switch to determine the route through the ladder.  

Simple concept.  But how?  When twin coil switch machines ruled the road, modelers would often use diode matrixes and a capacitive discharge to do this.  How about a diode matrix for Tortoises?

I found this: Where you use the internal contacts on the switch machines to handle some of the logic.

Since I was going to make a big mess drilling holes under the switches for the switch machines, I decided to convert the switches to a more DCC friendly arrangement. I used copper clad ties to replace the original solid metal throwbar and just filed through the copper in the middle of the tie.  I used the second set of contacts on the machines to power the frog.

Since this was going to be a lot of wiring, I decided to use terminal strips to standardize the wiring for each machine and to make it easier to fix mistakes. 

I made two fundamental mistakes.  One was not paying attention to the Tortoise pin-outs for the internal contacts.  The second was getting the "reverse" and "normal" backward on several because of how I mounted them under the layout.  This meant that my great planning to solder most of the switch machine terminal ahead of time on the work bench went for naught and I had to do soldering under the layout.  Yuck.

Because of the way the points were soldered to the throw bar, the Tortoises needed stiffer wire to develop the right combination of force and throw to move the points.  I also had to cut some relief in the ties near the points to reduce the friction throwing the switched.  

The control panel has LEDs to indicate which track the ladder is lined for as well as some push buttons to manually control the mainline - to yard crossover that is powered by a DCC accessory decoder. (NCE "Snap-it")

Here are some pix.

Yard ladder

Converted switch kit

Terminal strips

Control Panel

Some of the switch machines

Panel wiring

Sunday, December 13, 2020

Annotated 2040



Annotated 2040


I have annotated this blog in red to show what I was thinking about when I wrote it and to see how some of the ideas look today.

There have been quite a few advances in technology applied to railroading in the past several decades and several more are in the works or are being talked about.  Each has pretty much been a single application to a single situation.  PTC for signalling. ECB for braking.  DPU for locomotives.  I've always wondered what it might look like if all the technology, and a bit more, worked together.

Additionally, railroad are constantly changing.  The current trend is toward more intermodal freight and less coal.  This will likely require rethinking of how railroads operate and what the purpose of their track network is.

I've taken both these thoughts and welded them into the narrative below.  It's a real snoozer unless you like thinking about technology and railroads...even then...beware.

One project I’d really like to tackle is to try to put some more solid numbers against my fiction.  What would it take to smooth this route to 60 mph, electrify, etc?  How much time could really be saved?  What do the costs and savings look like?  So far it’s been too much and I’m too lazy to tackle it.

So, for now, an annotated version, including “what was I thinking?” and what’s happened since 2014! 

Friday, May 18, 2040. 8:00 AM. Harrisburg Pennsylvania.

It was quiet, except for the faint hum of an occasional electric motor.  The Harrisburg Mega-terminal was busy, though.
(Electric is coming fast.  It’s cheap and simple.  Cheaper energy.  Less

Charlie pulled his car into the parking lot and plugged it in.  Amazingly, he could still charge his car up for free at work - something about offsetting greenhouse gas credits.
(Carbon credits, cap and trade, carbon tax – we clearly needed and still need some sort of push to get to carbon neutral)  He really didn't understand, but free is always good!  Most places these days would bill you through your car's e-VIN - every car had an RFID tag in it these days.

(RFID tag on car:  There has been a lot of public discussion about how to pay for highways when electric cars become the norm and fuel tax revenue dries up.  One idea is to pay a per mile tax – or mileage toll.  A universal RFID, like Easy Pass et. al. could facilitate this.  It could also be used for billing battery charging – like Tesla does.)

The lot hadn't changed much since he hired on in 2010.  It had been repaved a couple times, but the gravel addition that was put on in the early 2020s had since gone back to weeds.  The terminal itself had kept expanding.  At the east end, was a vast, parking and storage lot for containers.  It reached almost to the ancient, but serviceable train shed at the Harrisburg passenger terminal to the east.  On the west end, the parking and storage reached nearly to Rockville Bridge.  In the middle of this were the two mile long loading track, four of them, straddled by 16 loading cranes. Compared to 2010, nearly three times the traffic was moving to and from the terminal these days.
(Triple the traffic = same number of trains, 3x the size)  The terminal superintendent had recently posted a chart in the crew room showing that growth of intermodal freight had kept up a pretty steady 4% growth rate this century (Extrapolated from NS data).  In the past couple of decades, about half of it was from the growing economy and the other half was conversion of carload traffic.

(Conrail had been expanding Harrisburg intermodal terminal at a great rate through the 1990s.  I was involved in some of that expansion, so I have a feel for the area.  The conversion of Rutherford to intermodal was a mistake in my opinion.  Multiple terminals in a terminal area make aggregating traffic difficult and costly.  One big terminal is better.  So, here I have expanded Harrisburg about as far as it can go.  The huge length makes it easier to dispatch a huge train.  The number of cranes, pad tracks etc. I ratcheted up based on the traffic level in 2014.  Exactly what the Chicago terminals might look like in my 2040 scenario, I didn’t think that through)

He walked into the crew room and his "electronic timetable" put him on duty and started downloading the latest updates, bulletins and other required information.
(More and more reporting has been automated over the decades.  It’s logical that crew time would be automated at some point in the future)  It also uploaded the "wheel report" and job profile.  Charlie didn't even give this a second thought - even if the company still used clunky-old local WiFi to do the uploading.  Wireless cell/satellite data was universally available and super-cheap these days.  They could just push the update as soon as it became available.  But, no, he still had to come to the crew room to get it.  Someday, maybe, the railroad would enter the 21st Century, he thought.

(And, here we are in 2020 with 5G building out.)

The Electronic timetables started out as an app for an iPad and the old paper timetables faded into history.  Eventually, a lot of other functions were added until there was no more need for computer terminals and printers.  Strangely, Charlie missed the sound of the printer.  It was just too quiet these days. (do we even need a tablet?  All the same info was on the locomotive....)

(Going paperless has been a thing for quite a while.  Maybe another couple decades?  For reference material, electronic would be an advantage – it can be searched)

Speaking of quiet, the room was nearly empty.  Not like the heyday of 2020 when there were usually a bunch of folk coming and going and telling war stories.  Some days, Charlie was all alone.  Today, Bill was there.  He was going on duty for the "Baltimore coal turn" job.  He wasn't happy.  The job was to take loaded steam coal to Baltimore for export, dump it, and return empty.
(Harrisburg to Baltimore and return would be a real stretch in 2014.  You’d have to really smooth out the process of on duty to depart, train arrival, dumper availability, coordination with Amtrak NEC, etc.  These are the kinds of things that can be worked on to really improve crew productivity.)  Africa was hungry for steam coal and the port was still pretty busy dumping coal onto vessels. (At a management class Conrail ran at Penn State in the 1990s, a futurist told us Africa was going to be the last refuge of cheap labor and last to industrialize.  It still looks like that’s happening but I think I’m way, way off on them importing coal.  Solar is just way too cheap – even now.)  The job almost never made it back in the required 8 hours.  Something usually got fouled up.  Charlie remembered when he hired on, there was a steady stream of coal coming off the Mon Valley - mostly moving in big, "bathtub" gondolas, 135 to a train.  A couple of locomotives on the head end, and a couple DPUs on the rear to Altoona where they would cut off. (A modest improvement from 135 car trains with manned rear end helpers)  Now, those gons were long retired, replaced with some nearly as ancient rapid discharge hoppers that formerly moved Powder River Basin coal from Wyoming to power plants around the country.  There were still some coal fired plants in the country, but coal usage was way down from the peak in 2020. A lot of rapid discharge cars were made surplus in the past 20 years. (I now wonder if there will be any steam coal moving at all in 2040.  Renewables are winning the day with combined cycle gas plant bridging the gap.  2040 is looking more like renewable plus storage, maybe hydrogen and fuel cells as part of the mix.)

The train for the "Baltimore coal turn" had come into Harrisburg directly from the mine, just barely making it in 8 hours.  These trains were dogs.  Typically 360 cars long with a single DPU every 40 cars.  30 year old 4500 HP, AC locomotives.  Still straight diesel - no electric or LNG-hybrids on these trains!  Even though the route from Harrisburg through Pittsburgh to Cleveland and Chicago had been electrified 10 years ago, these trains still ran with ancient diesels because the route from the mine to the mainline and from Harrisburg toward Baltimore was not.
(Port Road Branch wouldn’t have enough traffic to justify electrification, at least early on.  Funny, because PRR hung wires there.  Also, the lines serving the coal mines in their dying days would be poor choices for electrification.)

Invariably, one or more locomotives would quit, or stop "talking" to the head end, or otherwise be cranky.
(Big trains need super reliable equipment in order to get over the road in the same time as a smaller train.  More parts, more part that can break.  Electronics don’t “wear out” so much as just randomly fail.)  On top of that, these trains had a hard time getting to track speed and were constantly being run around by the speedier, electric powered intermodal trains. (Incremental HP should be cheaper with electric locomotives, plus short time rating is available.  Higher balance speeds + faster acceleration = more miles at track speed.  All paid for by equipment utilization and crew productivity)  Further, these trains had old-style "stupid" airbrake.  The original brake valves from when the cars were built in the first part of the century were mostly still in place and the train still carried an ancient end of train device they called FRED.  Charlie had no idea why they called them FRED.  Maybe it was the name of the inventor?  An old head had told him a story about a rather vulgar acronym.  That couldn't be right.  Railroaders didn't use that kind of language. (It’s nice to dream…) Anyway, you only used the air at your own peril.  A brake application was almost always trouble.  You'd either wind up in emergency or some car or another would fail to release.  What a pain.  And, no graduated release! (Graduated release was the one feature spec’d out for AB brake valves circa 1930, but they couldn’t get it to work, so passenger only.  Trivial with ECP)  Stopping smoothly and quickly was out of the question.  Lots of coasting and dynamic brake was the rule.

The "Baltimore coal turn" was the worst of the worst.  Even if you made it down to Baltimore in good shape, dumping the train was an adventure.  The rapid discharge mechanisms failed like crazy.  In each group of 40 cars, there was always at least one car that wouldn't open, or close, or got stuck half way.  Charlie thought some new equipment might not be such a bad idea, but with coal in decline, he suspected the company was just trying to run the wheels off this old crap and then get out of the coal business altogether.  Even now, the metallurgical coal movement that would remain after steam coal finished dying was starting to be converted to container service.  Many of the metallurgical coal mines weren't producing enough volume to require mine spurs these days.  Instead, coal was loaded into 45 foot open top containers which would then be trucked to the nearest intermodal terminal.  The stuff loaded out before it cubed out, so the only thing you had to watch for was overloaded containers. (08-04-2014 update:  They are moving grain in containers...why not coal? (I’m thinking volume would be too low to justify unit train business.  Note that a tenet of PSR is unit trains have to run frequently, otherwise dump the business into the carload network…and don’t get married to the business network concept in the first place.  Containerization of metallurgical coal would fit this scheme)

Good thing the rapid response Mechanical Flying Squad is around.  Literally.  They flew in semi-automated quad-copters that could hover automatically and put down just about anywhere and lend a hand. (Jan 2017 update - I thought this prediction was far-fetched.  Maybe not!) Rarely did it take them more than 20 minutes to get to you and they usually could get you patched up and moving in minutes.  Not like the old days, when he had to sit and wait for his conductor to trudge back to the trouble spot and to do his best.  Without the Flying Squad, they'd never make it over the road. Charlie did miss having a conductor in the cab with him, though.  Eight hours alone can be a long time. (If you are going to go to one man crews, you have to get help out quickly when things go bad.  Either need lots of help scattered along the line, or fast transport.  I wouldn’t bet against flying service trucks at this point in time – drone technology is moving fast.)

Charlie got his electronic timetable out.  It was actually an iPad-flex.
As soon as it saw he was looking at it, the small display powered up.  The battery was nearly fully charged and would last the better part of a week, thanks to WiFi trickle charging and ambient light solar cells. He unfurled the windowshade-like main screen and locked it into place, (Jan 2017 update - ) (Curved screen TVs were just coming out in 2014, why not scrollable?  Note Samsung has folding screen out now.)  looked at the "current job" icon and double-blinked.  He though about the days of dusty mice and smudgy touchscreens and broken phone glass.  Ancient history! (Google glass was just being talked about.  The man-machine interface is a tough one.  Blinking at the screen?  Who knows?  Maybe by 2040, you’ll wear a brain cap ala Arthur C. Clarke sci-fi)  A window popped up.  "TVHARCHE18_2" (NS has been trying to go to service and OD style train symbols nearly forever.  I figured by 2040, they may actually get there) .  He had an intermodal train that was being built on the yard at Harrisburg and was going to Chicago-Englewood on the 18th day of the month, second section.  He looked below the symbol and navigated his way through the job briefing, wheel report, bulletins, safety rule of the day and the menu at the crew hotel at the other end of the line.  Yum.  Lasagna for dinner, and French Toast for breakfast. He could deal with that. (Lots of standard stuff that’s been standard for century and more, plus some new stuff.   The device noted that he did these things....(Inward facing cameras would be ancient history.  AI monitoring of employees would standard stuff by 2040)

He noted that  there was a PTC section out around Conpit Jct, a slow order in Latrobe proper, nothing crazy in the consist.  All "smart DPU gen V equipement"

Charlie had the train as far as Cleveland.  He could normally make it in about 6-1/2 hours, thanks to some major curve straightening, realignment and electrification.  He could hold at least 30 mph all the way up and around the fabled Horseshoe Curve, and 60 mph most everywhere else.  There were a couple of 35 mph curves that were still being worked on, and some slow going through Pittsburgh proper, but that was it.
(Slow sections knock the stuffing out of the running time for really long trains.  If you want to improve equipment utilization, crew productivity and the transportation product, you have to get the average speeds up.  That’s investing in infrastructure)  There were even a stretch of 80 mph  from Alliance to Cleveland that was being tested.  There was some talk trying to bump up speeds elsewhere on the route, but there was some internecine "discussion" going on at high levels about this. always. (During my time at Conrail, I analyzed operating empty coal trains at 50 mph vs. 40 mph.  Trade off was fuel vs equipment.  I think 50 mph won out.  I also analyzed 60 vs 70 mph intermodal trains.  60 mph won because the trains that needed the extra 10 mph from time to time couldn’t do it without coming up with more locomotives and the trains that could do it, didn’t need the somewhat faster running times.  I think 60 mph is a good speed to shoot for.  You don’t have equipment issues and V squared part of train resistance isn’t too terrible)

 There'd even been some talk of a "Big Tunnel" project to by pass the Horseshoe curve.  The main proponents were the HSR passenger crowd, looking shave more time off the Phila-Pittsburgh route. If the tunnel could be constructed with a 1% grade, it could be used for freight, too, but that would cost more than a steeper passenger-only tunnel.  So, far, the project was more of a dream than a plan.
(Japan started new HSR line in the 1960s. Amtrak has been around since 1971.  Any doubt we’ll still be figuring out what passenger rail infrastructure to build where with what money in 2040?)

The new, lightweight stack cars couldn't take the pounding of the older equipment, so the power was distributed every 3000 feet through the train.  On the head end was an 6000 HP, four axle, AC electric locomotive with an aerodynamic cab on each end.  This unit was only two years old, built by GE....  NS had purchased three boosters for every cab.
(Kinda, sorta, back of the envelope swag using RoadRailer specs and limits for the target.)

Today's train is 14,000 feet long.  There's a block of 1500 feet on the end for Pittsburgh and a block of 3500 feet for Toledo.  The rest are for Chicago.

3000' Chicagos
booster loco
3000' Chicagos
booster loco
3000' Chicagos
2000' Toldeos
booster loco
1500' Toldeos
booster loco
1500' Pittsburghs

36,000 HP, 12,000 tons nearly 3 miles long.
(3 HP/ton is a good bump over the Conrail standard of 1.75-2.0 HP/ton for intermodal on this route.  But, might need sharper pencil on the math here.  Need 5 HP/ton to waltz over HSC at 35 mph.  Even if short time rating is 125%, need a total of 8 locomotives.  Six will only get you about 25 mph.  Maybe that’s enough?  Or maybe I need six, 8000 HP locos with 125% short time rating.  That would do it.  Sure beats 15 mph you can manage with 2.0 HP/ton)

The cars in the train were all less than 10 years old.  They were very light weight - about half of what a late 20th century stack car weighed.  And, they were smart.  Each car had computer on it whose function was to "talk" to the cars it was coupled to.  What did they talk about?  Who am I and who else is on the train?  What train is this? Where am I (yes, each car had GPS and 'dead reckoning'). Are my roller bearings happy?  Am I riding okay?  Is the track smooth?  Where am I?  What are my brakes doing and what does the engineer want them to do?  To I have enough air pressure in my reservoir?  Is my handbrake on? How fast am I going? What do those boxes on my back weigh?  Are my wheels round?  How's my battery charge?

It was a virtual trainline powered by "ultra-violet" wireless, the current version of "Bluetooth" micro networking, only much more robust and much faster.
(What’s faster than “blue”?  Ultraviolet.  Wired connectors are never going to work out.  The industry is going to need some contactless.  There has been talk of inductive connection buried in the glad hand or glad hand gasket.  I think some sort of targeted RF communication might be better.  Perhaps some sort of wave guide antenna, receiver combo that keeps the info from stray from anything other than a car on the same track only a few feet away might be the thing – or maybe laser light?)  The battery that kept it all going charged from small solar cells mounted on the car plus a small generator built into one of the roller bearings.  It was extremely rare for one to go flat. (Herzog has trains with “smart” cars for ballast service using solar panels to keep a battery up.  The power for ECP systems was/is a stumbling block.  Feeding power from the head end along the wires used for control is the same problem current air brake systems have.  It’s not a great idea to mix power and control signals in the same conduit.  The electrical connectors are the weak link of current ECP.  The industry needs a “next gen” design.  I remain stunned and amazed they haven’t gotten after this pretty hard.  I think they are making the same mistake as with PTC.  You drive the implementation and get the government to regulate around that.  Otherwise, they’ll regulate it and you’ll have to comply.)

The other big change was in the car structure itself.  Distributed power and a "smart" trainline meant that the forces between the cars could be kept very low, so the cars were designed accordingly using half the steel - and half the buff strength.  This made them cheaper to acquire, which allowed the old fleet to go to scrap in a big, economically justified, hurry.  The newest cars were even lighter - made mostly of aluminum.  Charlie had heard that the expected life of these cars was only 10-15 years.  That seemed strange to him.  Most freight cars in the past lasted 40 or more years, but these new cars were so cheap, they didn't need to last as long.
(DPU plus smart trainline can have a big impact on car design.  Freight cars are strong because we traditionally had to have all the power applied in one place – the head end – to keep crew productivity up.  Have to have high buff and draft strength to accommodate.  Freed from this constraint, cars don’t have to be as strong.  Less strong means lighter and cheaper.  Cheaper means economic life can be shorter.)

The things that hadn't changed were the standard, three piece freight truck and the coupler.
(If you only need to go 60 mph, the three piece truck is good enough.  Better trucks have a lot more complicated suspension and cost a lot more.) What was new, was these new intermodal cars had much lighter draft gear and machined couplers with much less free slack.  There was still some "play" in the train, but there was never any violent slack running in or out like the old days. (DPU allows for this, too.)

The three piece truck still had its issues.  It was prone to hunting at higher speeds, but the difference now was the "smart" car could detect incipient hunting and kill it with a brief, light brake application.  These new cars had been tested at speeds up to 125 mph, although nobody was regularly running faster than 80 mph.
(A total swag on my part.  125 mph with a three piece truck?  80 or 90 maybe.  ATSF managed with the Super C.  More speed might be controllable with more smarts on the car.)

Time to go.  Charlie walked out the door and got a quick ride in a crew shuttle up to the head end.  It was a two mile ride.  The yard crew had built the train entirely by remote control from the tower.
(partly why Charlie was lonely in the crew office)  He climbed on to the locomotive.  It recognized him from his electronic timetable, verified it with the in-cab camera, and greeted him, "Hi, Charlie!" (more automated reporting of people)  Charlie had never really gotten used to that.  It was still a bit unnerving.  Reminded him of HAL in that ancient sci-fi movie "2001 - A Space Odyssey".  Even though the rest of that movie was far and wide of the mark, it nailed the "talking AI machines".  They were everywhere!

The locomotive "shook hands" with the info on his timetable and told him the train was in order.  He radioed the tower and asked for permission to depart.  Kind of weird that this tradition still stood.
(weird because this could be easily automated.  People still need relationships with people, though)  A minute later, his PTC display went green and showed he had a route out of the yard on to main all the way across the bridge to Marysville.  Seconds later, the yard tower said, "TVHARCHE18_2, have a safe trip!"(With PTC, you could have yard tracks “signaled” and with ECP, slow speed stopping is enhanced, so you can wheel right out of yards at good speed)

He notched out rather rapidly at 8:47 AM and the train slowly began to move.  The DPU units were set to "automatic", meaning they would figure out what to do based on what the engineer was trying to do with the train and what was best for smooth train handling.  When accelerating  from a stop, it took tens of seconds for the train's slack to sort itself out, so the train just crept along for a few hundered feet (
Even with those reduced slack couplers and draft gear – Think Tightlock type H couplers – even passenger trains have some free slack).  After that, the train started accelerating rather robustly, reaching 30 mph "yard speed" in a matter of a couple minutes.  Having Electronically Controlled Braking (ECP) made all the difference. (early ECP tests showed stopping distances a low speed cut in half, so slow or restricting speed could theoretically be doubled).   Stopping distances at all speeds were greatly reduced.  No more, "it takes over a mile to stop a train from 60 mph!".  Trains could now stop at the rather uncomfortable rate of 6 mph/sec, bringing a 60 mph train to a halt in 10 seconds and a little over a quarter mile.(6 mph per second requires about 27% adhesion.  AC locomotives can manage >35% in all weather conditions – pulling, so 27% for braking should be safe assumption)

The train cleared the yard limits at 8:55 AM - five minutes early. Charlie could have set the train speed control on "automatic" and let the train follow the maximum authorized speed along the route, but today, he decided to set it at 40 mph manually.  As the head end was approaching the west end of Rockville Bridge, both the rear car and the yard reported the rear end clear of the yard.
(more automated reporting)

The PTC display chimed as it updated the movement authority all the way to Lewistown.  Charlie set the speed control to automatic.  As the rear end of the train exited Rockville bridge and cleared the curves into Marysville, the maximum authorized speed display changed to "60" and flashed a few times to let Charlie know it had changed.  Charlie felt the train accelerate (
There are just some places that aren’t going to be cheap or easy to raise speed.  Crossing the Susquehanna River would be one of them).

He set the auxiliary display to show the status of the DPU units in the train.  The information was fed thorugh the "ultra-violet" trainline back to the head end.  For a while, the DPUs also carried radio equipment to use as a back-up to the trainline data feed, or on trains not fully equipped with "smart" freight cars, but the radios came off 5 years ago.

The display showed all the DPUs running at nearly 7000 HP, 115% of their nominal rating.  Electrification was a wonderful thing!  In short order, the train was approaching 60 mph. Charlie watched as the DPU  power backed off closer to the continuous rating of 6000 HP, and then as it backed off gradually, rear to front as the speed indicator nudged 60 mph.
(Electric locomotives can pack a big oomph in a small package.  Plus, temporary oomph from short time rating.  Lots more miles at track speed!)

Every restrictive curve between Harrisburg and Spruce Creek Tunnel except Jack's Narrows had been re-engineered for steady 60 mph running.(
Another tough spot on the line.  The PRR had really smoothed this route out over the years.  Northern RRs had money to spend in the late 19th and early 20th Century.  Routes were smoothed and straightened a good deal, unlike in the South where “laid on the dirt” still rules in many places to this day)  Charlie settled in an watched the scenery go by.  He never got tired of the hills, farms and windmills along the Juniata River valley and could pick out remnants of the old Mainline Canal here and there.(You can now, so in 20 more years, why not?)

Charlie's train rolled through Denholm, where the PRR had a giant "mainline fueling plant" a century ago.  A huge bridge, with coal chutes slung underneath to fill the tenders of steam engines,  Hopper cars were pushed over the bridge and dumped directly into bins above the chutes.   You had to know where to look to see the remnants of it - just a few rock walls peeping out from the vegetation.  There was still a weigh-in-motion scale on the controlled siding. On it sat a coal train.
(nothing changed from current day)  The engineer was on the ground talking to a couple of mechanical department folk.  Their quad-copter had landed in the wide spot between the two main tracks and the siding.  They were looking in the direction of a hopper about 20 deep that had coal spilling out from under it.  It was amazing where those things could put down, thought Charlie. (Jeez...I make up something like like this and then I see this:  on 1/7/16)

What a mess!  Charlie guessed they would cut away from the train on both sides of the disabled car, then use the DPU units on the back part of the train to back the train out onto the main.  The head end would pull ahead, then back down onto the rear portion, couple up, get an "air test" - actually, the air brake computer did the test all by itself -  and go.  Might take 20 minutes to do all that.  Would have taken 3 hours in the "old days".  PTC, DPU and "the flying squad" made things much easier these days.  Charlie gave them two "toots" on the horn and a wave as his train glided by.
( I think I may have ECP on this train…My Baltimore turn did not.  Inconsistency in the story?)

As the train slid through the curves at McVeytown, the morning "Keystone Service" train flashed by at 90 mph.  On NS, it was known as AMPITHAR18_1.  On Amtrak is was simply train #660.  Amtrak kept a rather simple train symboling system that had its roots in the ancient Timetable and Trainorder train operation days.

In 2030, Pennsylvania had purchased some "off the shelf" Talgo X equipment.  It wasn't the "latest greatest", but it did allow 7" of cant deficiency, meaning it could handle curves set for 60 mph freight service at 90 mph
(I did the math on this).  It was built to European crash standards, but PTC and other technology applications had improved the railroad safety record an order of magnitude in the past two decades, so the FRA decided the old crash standards weren't needed and they adopted the international standard....finally.(Amtrak’s new Acela, too?)  It always took them a while to catch on.

The state of Pennsylvania was "funding" 5 round trips, Pittsburgh to Harrisburg extensions of Amtrak's New York to Harrisburg trains.  Truth was, these trains were more than covering their operating costs since the running time had been shaved over two hours over the past couple of decades.  Pennsylvania just had to come up with the equipment.  Ohio had purchased a few sets of their own to be able to continue a few trains a day out to Cleveland and one to Columbus.  There were rumors Ohio was interested in more.
(Maybe 50 years of talking about expansion would finally yield fruit?  Talgo seemed like a good bet in 2014, with Wisconsin, Oregon and Washington in on it and Talgo building a plant.  Tilt would be very useful on this route.  Seems much less likely now…) The HSR crowd was grumbling about 90 mph not being fast enough and "the rest of the world is embarrassing us - we only have 500 miles of true HSR". "Rumors and grumbling.  The railroad never changes", Charlie thought.(Even Joe Biden won’t get us much HSR.  But, we might gets some HrSR, or at least some expanded frequencies.)

He watched the Amtrak train fly by. At at relative closing speed of 150 mph, it was hard to get a good look at anything, but Charlie saw nothing amiss. He keyed his radio, "TVHARCHE18 to AMPITHAR18.  You look good on the north side, Harry."
(I now wonder if vision systems and some AI might backstop the “rollby”)  Charlie had known Harry Muntz (A good middle Pennsylvania name) since he hired on.  The both got "cut off" in the recession of 2019 and Harry had hired on with Amtrak. Charlie rode it out and had been with NS every day since.  There hadn't been much hiring since then until recently.  One man crews and long, DPU trains reduced the man power needed to run the railroad.  Long crew districts from electrification were pushing man power down to new lows. Only as those a bit older than Charlie started retiring had the railroad started hiring again in earenst. "Roger, that, Charlie. AMPITHAR18 out."  Less than two minutes later, Harry returned the favor.

One thing Charlie missed were the talking defect detectors.  They were gone.  The "smart" freight cars now knew more about their "health" than the detectors ever knew.  If there was a problem, Charlie would know about it.  The locomotive would tell him.
(The locomotive knows all.  It knows where it is, what it’s doing, where it’s going, who’s on board, what it’s pulling, and how all those freight cars and DPUs are doing.)

The lading could "talk" over the "ultra-violet", too, but Charlie was blissfully unaware of this data. It streamed over the locomotive data radio and was distributed to the shippers and receivers of the lading.  They could know the temperatures inside the containers, the ride quality, the location and speed and the estimated time of delivery of their goods.  Almost too much information.
(and not much of a stretch once you have “smart” freight cars and solid trainline data.)

Right behind the Amtrak train came MRBELENO18_1.  A merchandise train from Bellevue Ohio to Enola Pennsylvania.  Two electric locomotives and one hybrid diesel electric.  One at the head of each block.  On the head end was the Baltimore block, then the Morrisville block, and finally an Oak Island block.  The diesel was on the Oak Island block.  The catenary had only been strung as far as Reading, on the way to Philadelphia, on that route, so a non-electric was needed to get to Oak Island.  That this train was using a diesel electric hybrid was a bit strange.  The standard non-electric locomotive these days used LNG for fuel.
(LNG was a hot topic in 2014.  It’s cooled off a bit.  Now we have “battery road slugs”.  So maybe the non-electric standard will be a diesel electric paired with a battery road slug)  Nearly everything was a hybrid, storing the dynamic braking energy and using it to accelerate later.  Once GE got the energy storage technology figured out, even the older units were modified to be hybrids.  It turned out that a combination of super-capacitors and storage batteries did the best job.(Batteries seem to be winning the day, right now)

In the "old days" they just would have traded power at Enola on this train, but now, the philosophy was "power goes with the block".  No more shuttling power around the yard, cutting and coupling all over the place.  DPU enabled the change.  Life was much simpler.
(this is a paradigm shift the RRs haven’t quite grabbed onto yet)

The train would be block swapped in Enola, each block being paired up with a similar blocks on the trains from Chattanooga TN and Decatur IL.  With the DPUs being remotely controlled at the yards, the whole block swapping operation was pretty simple.

There wasn't much car classification at hump yards going on.  In fact, NS was down to just five major hump yards.  Macon, Chattanooga, Enola, Bellevue and Linwood, with Bellevue being the largest, by far.  It was the major hub in the network in the north.  Chattanooga was it in the south.  The other humps were basically regional hubs and the whole network was really a fairly simple to understand hub and spoke network.  It was a bit more circuitous than the "old way", but it made up for it with faster trips, on the average.
(RRs are going to keep needing to find ways to rationalize the merchandise traffic network.  You could have a bunch of small flat yard and a really complicated blocking network, or you can have a few big hubs with lots of spokes.  I chose the latter.)

The traffic that moved on these trains had once been the "bread and butter" of the railroad business.  Now it was almost strictly "boutique" business. Chunks of raw materials for manufacturing.  Stuff that wouldn't fit on the highways.  Low value finished goods.  Most of the higher value stuff was moving in intermodal service service and the economic dividing line between the two kept moving in the favor of intermodal as economy of scale and frequency and speed of service kept driving intermodal costs down and value up.
(and the beat goes on.)

Interestingly, unit trains other than coal and crude oil had pretty much dried up.  The unit train sized "chunks" of traffic now moved in the merchandise "hub and spoke" network.  For example, what was a 75 car grain train was now just two 40 car blocks moving in merchandise trains service, each with their own DPU locomotive.  For example, grain moving from Illinois to Georgia would get picked up at the grain elevator by a passing train and taken to Bellevue.  There it would block swap to a Chattanooga train.  The dwell was only 4 hours, on the average.  At Chattanooga, the same thing would repeat.  Finally, at Macon, each grain block would move as it's own train to the feed mill.
(Take that EHH!  I have minor unit train traffic moving in the merchandise network!  Very PSR like, don’t you think?)

Crude oil trains, unlike coal, had a much more modern fleet of cars.  The flow of crude from the oil fields to refineries was steady and expected to remain that way for quite a while. .
(The view from 2020 shows that oil trains are likely to dry up too.  Oil train oil is at the high end of the cost to produce.  Oil demand may be drying up faster than the low cost producers will deplete their supply, so Bakkan oil demand might not every come back)    The techniques for getting oil out of the ground kept getting better and cheaper.  Charlie's morning "paper" had a nice article about energy.  World consumption of oil was steady.  Coal was declining - slowly.  The balance was being made up by a combination of wind, solar, natural gas and nuclear.  Conservation efforts had kept the growth rate down to "a dull roar".  The CO2 concentration in the atmosphere had started to decline in the past decade, but there was still great concern over global warming.  Charlie wondered, "Whatever happened to blissful ignorance?" (looks like I was blissfully ignorant in 2014.  What I lay out here isn’t near enough to get the results I lay out.  Coal and oil are dying faster than I imagined they could.  Natural gas seems to be the steady one.  Nuclear still having trouble gaining traction)

The trick to keeping the railroad fluid these days was to keep the merchandise trains powered up to go with the flow of the intermodal fleet and keeping the coal trains out of the way of everything.  There were still too few Amtrak trains to cause much trouble, even though their number was growing.  Most days this worked fairly well.
(Ain’t never gonna run like a Swiss watch, or even a Swiss RR, with 3 mile long trains….)

The PTC display had been getting updated regularly with new movement authorities, usually keeping "the virtual brick wall" about 20 miles ahead of the train.  Charlie suspected he was following a merchadise job out of Enola and he wasn't gaining or losing ground to him.  Usually, that guy was behind the TVHARCHE_2, but, apparently not today.  Charlie was bored and decided to find out.  He toggled his auxiliary display away from the DPU screen to "apps".  He selected "track line view" and it popped up with  the "Northern Division" (one of five divisions on the railroad) , "Pennsylvania Middle" display.  He was looking at a dispatchers view of the railroad from Harrisburg through Pittsburgh.  Sure enough, there was a train about 25 miles ahead.  He zoomed in.  It was tagged with MRENOCHI18_2.  Just as he thought.
(No huge technical hurdle providing dispatcher’s track line view to locomotive cab.  This is a “need to know” vs. “have to know” issue.  I believe you get the best performance from people if you can get them to behave as owners rather than wage slaves.  Trusting them with information is one way of doing this.  And, I think it would be cool to do.)

He looked again at the screen.  He was lined to pass the train ahead. MRENOCHI18_2 was to cross over from track one to track two at CP-Hunt and then run into the yard at Atloona to work. He'd have to slow down to 40 mph to make the crossover since it was one of the few interlockings on the division that still had #20 switches instead of those long #26.5, moveable frog jobs.
(60 mph crossover – track speed for freight.  Saves fuel and time.  MOW not fans, though.  More work for them to keep things in shape, but the goal is to make the RR run well, not make the MOW budget small.  Better product, plus much lower costs, means you can spend on MOW, if that’s what it takes.)

That was quite a ways away, though, and there were several eastbounds headed this way before that.  The train dispatching system was generally operated by the "Movement Director" application.  It figured out the smoothest, fastest way to get traffic over the road, protecting commercial schedule times and connections in the process and would dispatch accordingly unless the train dispatcher intervened.
(NS currently has, and mostly uses, the Movement Planner function in its train dispatching system which tries to get all trains over the road as close to on time as possible by minimizing variance from schedule at train end point.  This is an evolutionary step forward from that.)  It had greatly increased dispatcher's territories.  The "Pennsylvania Middle" dispatcher had the railroad all the way from Harrisburg to Conway.  Charlie still marveled at all the information at his fingertips.  Back when he hired on, the company only told you as much as they thought you needed to know.  Getting information from the dispatcher about the trains around yours was nearly impossible. All you needed to know was your territory, the train consist and the signal aspect.  Period.  Then, came the big push to save fuel, and some on-board "smarts" about how to handle the train smoothly to minimize fuel consumption.  Nobody liked that, at first -  engineers were proud of their knowledge of train handling! Slowly attitudes changed.  Later, the train dispatching systems started giving "advice" about meets and passes so you could reduce train speed to arrive at the meet point at just the right time.  This was nice, but often the dispatcher was too busy to pass the info along, or too busy to give you an update when one became available.  Then, came the "Walmart affair."(pure fantasy)

The US economy in the late 'teens was stagnant.  Wages for the lower end of the labor pool were actually shrinking while the upper end continued growing.  The "class warfare" gang were turning up the volume every year.  Walmart continued to be the "poster child" of low end labor, and, pardon the pun, the number one target of organized labor.  "Occupy Wall Street" which started in the late aughts had morphed into "Occupy Walmart".  After a long and very messy period - several Walmarts burned to the ground - labor won the vote and Walmart became a union shop in 2021.  Right on the heals of this, McDonalds was organized in 2022.
(Pandemic has delayed the timeline.  That’s my story and I’m sticking to it!)

Despite the rancor that occurred before the vote, the first contracts at both places were put in place rather smoothly. Wages would roughly double over the first five years and management standard benefits were extended to the hourly employees.  Part time work was also limited.  In exchange, there were almost no division of labor requirements put in the contracts.

Nobody could have guessed what would happen next.

Five years later, Walmart was still the low cost leader, with prices even lower than five years prior and McDonalds was still selling $1 burgers.  How was this even possible?  Everything about the stores was on it's head!  The stores became highly automated.  There was a revolution of applied technology. It was the "perfect storm" of technology and materials developed over the past several decades.  Smart machines, with vision, heuristic "smarts" and "tactile hands" could do everything from cooking and assembling burgers to stocking shelves to loading your car. Stores were rearranged and redesigned to take advantage of the automation technology.  It only took a small handful of workers to keep the machines operating decently and in order and a few more for customer service.  Employment at Walmart and McDonalds was down 70% from the peak.
(I believe the way forward for the US economy is AI and automation.  Enough said.  Unionizing service jobs?  Who knows?)

Interestingly, the mass exodus of jobs from these companies did not increase unemployment.  The creation of an entire industry based in automation of service jobs pushed the economy into a new growth spurt as the US became the leader of the "service automation" industry. (

This changed everything at the unions.  It was the end of "fighting for wages and benefits".  That ship had sailed.  It was the beginning of "make my members valuable" era.  Now, the unions fought for making information and education available.  It took a while for all of this to sink in at the railroad, with their centuries old adversarial relationship with labor, but changes occurred rapidly.  All the unionized employees were now treated more like "owners" and less like "mercenaries".
(The “wages and benefits” slant to unions has failed.  Time for a different approach based on employee value)  There were company paid opportunities for educational sabbaticals and information was shared on a "why not?" basis instead of a "need to know" basis.  Getting to see the "track line view" of the railroad was just one small indication of the sea-change in the service industry.  Life was good!  (There has been huge disruption in the US economy before.  From 90% farm work in pre-Civil War days to 90% non farm work.  The change was slow compared to the rate of change now, but we don’t have 90% unemployment because farms were automated.  The jobs most of us do now were unimaginable in 1870.  Why should current times be different?)

The train proceeded through Mount Union.  The East Broad Top was running today.  Their ancient steam locomotive was all polished up and was being turned on the turntable there.  Out of the turmoil of the late 'teens, came a neo-Victorian era.  The "well to do" became more interested in putting their wealth to charitable work (out of guilt?  who knows?)  and even architecture and fashion were following late 19th Century England.  Things of the 19th Century were in great demand.  The confluence of these two trends occurred on the East Broad Top Railroad.

In his late 80s, Warren Buffet bought his second railroad.  This one, just for fun.  The East Broad Top. A complete, fully functional - or nearly so - 19th Century railroad.  His sprinkled a tiny bit of his enormous wealth into the property had had it running and looking exactly like it was in 1910, just hauling people instead of coal.
(Soooo close!  Not Buffet, but a bunch of well to do retired RR execs!)  Although 1910 was really post-Victorian, it was close enough in most people's minds and the road enjoyed a swell of popularity.  There were at least a couple hundred people riding the train - some dressed in frilly dresses with parasols! (Fashion and culture are strange.  Victorian revival?  Why not?)  Many of them were watching the locomotive turn and were taking pictures.  A few turned their cameras around and shot Charlie's train as it slid by.  "Foamers never change", Charlie muttered to himself, although he had quite a collection of train pictures from when he was younger somewhere "in the ether."

Even photography had changed a great deal over the past couple of decades.  Where people used to joke about being about to put every minute of their life up on YouTube, now, not only was it possible, many people were doing it. Small, wearable cameras could upload video and audio on a near-real time basis. (update 2/26/14:  I had no idea! Don't remember what you were doing on April 13th, 2037 at 2:12 PM? Just pull up the video! This technology had dropped the crime rate for personal crimes down to nearly zero.  Most crime these days was electronic.  The old arguments about data security and personal privacy still raged, however. (Look at all dust-ups over cell phone videos now.  The civil rights era was enabled by TV news shining light into dark places.  Cameras on cell phones and video door bells, etc. shine light everywhere.)

There were cameras mounted all over Charlie's train.  The locomotives had several looking forward and backward outside.  There were even a few freight cars equipped - mostly to monitor the lading.  His cab had one aimed at Charlie himself.  Being watched every second of every day just seemed pretty normal, Charlie thought.  Every step forward with technology brought a host of new, unanticipated questions and issues, it seemed.  Things seemed to happen before anyone had a chance to fully think them through.
(We live in a technological era.  Technology leads, culture and governance lag)

Beyond the personal video cameras, popular amateur photography was nearly completely 3D these days and there was no difference between still and video. It was all video.  If you wanted a single picture, you just selected a frame from the video.  You could create small bas-relief replicas of any scene you captured using a 3D color printer.  It was a popular way to decorate your home and make personal jewelry.

The latest thing was Google WorldView
(What’s better than 3D street view?  WorldView!).  If enough people shared their video with the service, Google could create a virtual reality of the scene, complete with sound.  You could view a 3D version of the event from any view point you wanted.  There were even Google Rooms where you could immerse yourself in the virtual reality based on the images collected, processed and stored - sort of like the "holideck" in the old Star-Trek TV show -  though some thought this was just a passing fad. (Oculus , now.  What next?)

The last of the eastbounds for a while passed and Charlie gave them the once over as they passed.

The PTC display flashed and showed the maximum authorized speed dropping to 50 for Jack's Narrows, along with the braking curve that outlined safe operation.  Considerably before that, Charlie heard the switch gear in locomotive snap and hiss.  He pulled up the DPU screen.  One by one, the locomotives had gone into dyanamic braking, from the rear to the front and were gently reducing the speed, keeping what little slack the train had stretched out.  All the DB energy was directed back up into the catenary.  The speed dropped right to 50 mph right as the train entered the curve, and then, nearly four miles later, as the tail of the train left the curve, the locomotives went back to motoring, one by one, from front to rear.
(Automated train handling should be a thing by 2040.  It almost is now with Leader, 

The train ahead had slowed down to make the crossover at CP- Hunt and the distance between the trains was narrowing.  The PTC display showed authority only as far as MP 201.5 about a mile short of CP-Hunt.
(Actual places currently in ETT)  The "Movement Director" knew what was going on and allowed Charlie's train to decelerate.  First by coasting, then with some gentle dynamic braking, gradually slowing as it approached the end of the movement authority.  When the speed had dipped to about 45 mph, about 2 miles from the interlocking, the PTC display flashed and the movement authority moved all the way to Cresson. Soon the train was passing CP-Hunt and headed back up to 60 mph.

Shortly after leaving Huntingdon, the train started slowing to 40 mph.  This was the only slow section of the route, other than the climb over the Alleghenies, still left.  The rest of the slower curves to the west had all been superelevated or eased to allow steady  60 mph operation
(MOW doesn’t like this either.  More work for them.  But, goal is moving freight not managing dept. budgets).  The head end entered Spruce Creek Tunnel at 40 mph and held steady at that speed as the train wound around the curves along the Little Juniata River through a river-grade gap in the mountains toward Tyrone. (Some of this is 35 mph, currently)  Heading south out of Tyrone, the train quickly got back up to 60 mph.

Charlie glanced back at the mountain ridge to the east.  As far as he could see, the top of the ridge was peppered with windmills, all of which seemed to be turning all the time.  He wondered how many of those windmills it took to power his train.
(Wind power in the US has doubled from 4% of generation to about 8% since 2014.  Those mountain ridges in Pennsylvania should be prime locations for wind turbines.)

Passing through Bellwood, Charlie took a glance at the Bellwood tunnel project.  They were using one of those huge tunnel augers to bore through the mountain into the valley to the east.  There was a line of ancient open top hoppers being loaded with the excavated material.  The new tunnel would allow a straight shot from Spruce Creek Tunnel to Altoona - 60 mph all the way.
(I did at least look at a map to plot this route)  The new route was supposed to open in 2042 and shave another 30 minutes off the running time, as well as provide some more of those "carbon offsets" (Total SWAG at running time and that there would be positive ROI).

The train cruised into Altoona.  Charlie peers deeply into the Juniata shops to see if there is anything interesting lurking at the backshop.  It was getting harder to see now that they had catenary strung up all over the place.  He did spot the nose of one of the "Executive F7s" peaking out of the E&M building.  He heard they were converting it to a hybrid, with super capacitors for energy storage located in the B units and 3000 HP LNG fueled engines in the A units.  It would be the first to solely use super capacitors instead of mostly batteries.  Interesting, Charlie thought.
(and the F units are gone… Would F units make a good test bed?  Technically? Probably not – better to use something more modern so more of the form – fit – function work is valuable.  From a PR standpoint?  Probably a good idea and you don’t take a revenue producing unit out of the fleet.)

The speed started dropping as the train approached "the brick yard curve".  There hadn't been a brick yard there in all the time Charlie had been working.  He wondered about the origin of the name.  Something to look up some day.  Couldn't do anything like that while working!
(Actual place.  Brickyard is no longer there.)

The headend hit the grade and started digging in.  Charlie switched to the DPU screen and watched as the DPUs started powering up as their portion of the train hit the grade.  It was all so smooth.  Charlie had run DPU train manually here in the past and it was tricky.  He marveled at how smoothly the locomotives did it by themselves.

The display flashed and Charlie switched to the bulletin order menu.  The PTC outage was over, so he wouldn't have to take any Track Warrant style movement authorities this trip.  Charlie still had to do some "manual running" to do, however.  Every trip had some to keep train running skills sharp.  Most times it was on pretty easy territory, but a few times a year it was over the mountain. Charlie did not look forward to those times!  It was nerve wracking.

The train cruised up and around the Horseshoe Curve right at the 35 mph limit
(Currently 30 mph in apex of curve, 35 most of the rest of the way.  Most freight never comes close to track speed at present.  Could save huge amount of time.  Speeding up the slowest segments give you the biggest bang for the buck in reducing running times).  It was quite a change from the pre-electrification days.  Charlie had run trains back then that were lucky to break out of the single digits going around the curve, although 12-15 mph was more typical.  Looking at his DPU screen, he saw that all the locomotives were running at 108% of capacity (I only have enough oomph for 25 mph here – see comments back where consist is laid out).  The locomotives knew how long it would take to get to the top of the grade and adjusted their output to match.  Charlie used to be amazed that they could do this, but it was getting to be "old hat" - whatever that meant.(Does anybody know what “old hat” means?)

There were only two tracks on the mountain now, the third track came up with the electrification.  Trains could get up and over the hill so fast, the the extra capacity of the third track was no longer needed.  There were just about the same number of trains a day as 40 year ago, but those trains were longer and faster.
(As long as there is anybody on board a train, the push for longer trains will remain. Should trains ever be totally automated, the push will be in the other direction.)

The headend rounded the apex of the curve and Charlie waved to the 25 or so railfans and families in the park.  Most not taking his picture, waved back.  On display there were the two ancient NS "heritage units".  One was painted "Pennsylvania" and the other "Conrail".  The old GP9 that was there for 50 years had been donated to the state railroad museum about 10 years ago. The display had been upgraded to include a roof that was supposed to keep the locomotives from weathering.  That old GP9 was pretty rusty when they moved it out.
(It’s pretty rusty, right now.  It’s been there since 1987?  The heritage units will be about 30 years old in 2040, and obsolete in my scheme of electrification plus LNG – battery hybrids)

Shortly after that, an eastbound approached going down grade.  It was a long intermodal train similar to Charlie's.  As the locomotives passed Charlie, heard the whine of the invertors, but no dynamic brake cooling fan.  The train was pumping energy back into the catenary and was, in fact, helping get Charlie's train up the mountain.  At the times the the power regeneration was greater than the consumption, there were big sets of batteries and capacitors that stored the energy for later use.
(It is a lot easier to do energy storage wayside than on board.  Electrification makes this possible.  Another possibility would be to do wayside hydrogen generation and then use that in a fuel cell.  Direct electrical energy storage would be more efficient, though)  The company news release said the system paid for itself in a matter of months. The scream of the dynamic brake blower was almost a thing of the past.  Charlie couldn't recall the last time he heard it.  Must have been a couple years now.

Soon, they reached the top of the hill.  There, the tracks split, one on the left going to the "new" tunnel, the one Charlie's train was on, to the old, double track tunnel.  With the electrification and the size of the trains these days, the short steep "slide" section just east of the "new" tunnel was nearly a moot point.  It didn't matter which tunnel they routed you through.  Electrification was a bit tricky here.  The tracks had been shifted to the center of the double track tunnel and the floor undercut just a hair in order to sneak the catenary through.  The "new" tunnel had to have the floor undercut a great deal to fit the wire.  There was some talk about daylighting both tunnels, but cost and local opposition stopped that idea cold.
(This project would be way down the list since there are still more curves to smooth…and only so many people and machines to do the work.)

The train coasted away from the tunnel, the DPUs smoothly going from motoring to slight braking as the headend entered Cresson. On the right, where they used to service helpers, there were a few older diesel locomotives- battery hybrids - that were occasionally used to assist trains in trouble on the hill as well as serve the few local customers that still had direct rail access.  Next to them was the sub station with one of the energy storage systems.

The train gently accelerated to 60 mph leaving Cresson. From Cresson toward Pittsburgh, there were big changes over the past couple decades.  Many curves were eased and had increased superelevation.  The goal was to get the speed to 60 mph the whole way to Pittsburgh.  They were close. So far, there were only a few curves left to ease.  There was ongoing work on one of these now.  Charlie grabbed his iPad-flex and took another look at the bulletin order.

At Bolivar - MP 295, he had to reduce speed to restricting to pass the work crew.  The PTC system should handle it automatically, but Charlie was ultimately responsible.  It was not unknown for there to be times where the locomotive did not know about a work zone and would sail through at track speed.  The manual data entry that defined the work zone was not infallible, after all.  He checked the PTC data on his route by zooming out, scrolling and zooming in, on the PTC screen.  It had the work zone in there properly.
(Single point failure systems are generally bad.  Having people and machines back each other up is better.)

Next, Charlie took a look at where he was to run manually.
(Keeping up skills in an otherwise automated environment is really important.  If/automation fails, you have to know what to do.  The PATCO High Speed line from Phila to NJ has operators run manually off-peak, but use the automation during rush hour when there is less slop for individual variation.  I worry about the current push for “auto-driving” from Tesla  They are “fair weather” systems that work well only under the best conditions, leaving the driver to take control when things get dicey.  Really need them to be the other way around.)  It was departing the work zone all the way to Pitcairn TV, the intermodal yard just east of Pittsburgh.  The toughest part would be negotiating the hill at Latrobe.  Keeping the train forces under control as with the train moving through a sag or draped over the top of a hill was always a bit tricky.  Fortunately, he could see that status of the train forces on his display and adjust the DPU power and braking accordingly.  Charlie thought about the stories the "old heads" told when he hired on about when then guessed wrong and the slack ran in and out with teeth-rattling results.  He couldn't even imagine how they got over the road with a train back then.(Been there, done that.  Riding train on NEC with engineer who recently bumped back from Amtrak to Conrail – and had been up for >24 hrs because of divorce court.  Totally misjudged a sag.  Started notching out fast.  Too, late.  WHAM!)

The train rolled through what was left of Johnstown at a steady 60 mph when the auxiliary display started flashing and sounding an alarm.  The last locomotive had a fault displayed -  "#3 inverter fault".  Charlie silenced the alarm and looked at the DPU screen.  Sure enough, that locomotive was only putting out 75% of the power of the others.  Charlie called the dispatcher, "TVHARCHE18_2 to the Pennsylvania Middle dispatcher".  "Go ahead Charlie," was the reply.  "Rear unit just lost an inverter.  Am I in good shape to proceed?" "We see that. Standby for a minute."  "Roger, that. TVHARCHE18_2 out"

On automatic control, the loss of power by the last locomotive would be taken into account, but it couldn't look ahead to the whole route of the train and see what the impact of the loss would be on the schedule performance of this train in particular, and on the network, in general.  Would the running time be longer?  By how much?  Would it leave locations short on power?  Crews?  Would it cause congestion because the train was not in the right place at the right time?  For these answers, they needed to run the network tactical model.
(Holy grail – at tactical resource planner. Strategic planning tools exist, but real-time tactical, no.  Conrail tried in mid 1990s with American Airlines adapting their gate planner to the RR.  Didn’t work.  Mostly data quality.  Better data and computing should make this possible and integrating them into a single interface, even better.)

The network tactical model took the current status of the network and then simulated the future to see what would happen.  There were three primary dimensions to the model - trains, locomotives and train crews.  The trains portion worked like the movement director and predicted where trains would be when.  The locomotive and crew portions took the train information and then forecast forward the availability of crews and locomotives for the next 72 hours, looking for surplus and shortages and providing actionable feedback to keep things balanced.

For example, if it predicted Charlie's train would arrive late into Chicago, and therefore the power wouldn't be available for the currently planned next departure from Chicago, it could suggest that the rotation of power at Chicago be altered, that extra locomotives start moving from a surplus location to Chicago -and what train to move them on, or it could suggest that the planned departure at Chicago just be set back a little.  It was a great help to management to keep things running smooth.
(Big trains, lots of moving parts, reliability is still going to be an issue.  How to manage and work around the exceptions with minimal resource cushion is key. RRs currently get into congestion problems by trimming resources too much when things are running well only to be short of what they need to get out of the ditch when things go bad.)

The current status and projection of the network was also fed to commercial side of the house, as well, so that customers could be kept informed of precisely when their shipment would arrive.  This was work that was started piece-meal at the start of the century, and even a bit before that, but only in the past decade or so had it all come together.  Big Data, abundant wireless communication, cheap data storage and fast parallel processing made it all possible.
(Just a continuing trend.  See NS’s Rail Pulse initiative of 2020)

"Pennsylvania Middle to the TVHARCHE18_2."  "Go ahead, Bill."  "Charlie, you're okay to go.  That unit stays with the Pitcairn block.  Mechanical has the part headed there already."  "Roger. Thanks, Bill. TVHARCHE18_2 out."  "Pennsylvania Middle dispatcher out"

The train was only a mile away from the work zone when it started to decelerate.  It used to be that, in the interest of saving energy, the proper technique was to try to coast into lower speed zones, avoiding braking.  But now, with regenerative braking and wayside energy storage, there was little energy penalty and the improved running times were worth it.
(A swag.  No math performed here.  Something for a later date for me.)

The train dropped down to 20 mph and Charlie changed the operating mode from "automatic"
(Leader calls this “auto throttle, but this would be a bit more sophisticated with multi-DPUs)  to "manual".  Restricting speed meant he had to be able to stop in half the line of sight.  Even with this 3 mile long trains, the ECP braking reacted almost instantaneously, and full braking could stop the train rather quickly.   Still, Charlie had to be ready.  Nothing eventful occurred.  Charlie noticed they were moving a lot of earth from a blasting zone where they were straightening a curve.  It was nice to see the company investing so heavily in right of way improvements.  Longer trains meant that the time penalty of slower speed zones was amplified.  A one mile long train meant travelling slow for two miles, not counting slowing down and speeding back up.  A three mile long train meant travelling slowly for six miles.  No wonder they were working so hard at straightening curves!

Charlie's display showed he was clear of the work zone and he'd gotten confirmation over the radio from the work crew there.  The train was still passing through the curves that were in the process of being bypasses, so Charlie held the speed to the required 40 mph.  Charlie had the DPUs in "synchronous", for now.  They were following a river grade so train handling was pretty trivial.

They were clear of the work zone.  Charlie switched over to "non-schro" mode.  The DPU display showed the controls for each locomotive.  Since it was a touch screen, Charlie just had to increment the "throttle" on each locomotive by taping the "+" button next to that locomotive.  He worked from front to back, tapping the "+" for each successive unit in order, about 5 seconds apart.  This kept the train forces nicely sorted and the train firmly stretched.  In a couple minutes, they were all humming at full throttle, except that last unit, that was operating at 3/4 capacity and shortly after, they were back at 60 mph track speed.

Soon, the train approached Pitcairn yard.  Charlie was still in manual mode.  He throttled back to idle and applied the blended braking as he was taught.  The train glided into the  five mile long controlled siding that served as the yard lead.  Charlie set the ECP brake control in "park".  This applied the brakes to each car and then activated a motor-driven "handbrake".  The display flashed yellow "PARK" while this was happening and then steady green "PARK" when it completed and "heard back" from each car. The yardmaster called him on the radio and told him to release control of the last locomotive on his train.  He worked the touchscreen and did exactly that.  "It's all yours, Charlie told the trainmaster.

Out of Charlie's sight, a utility man rode up to the rear of the train in a glorified golf cart, got out and cut the train just ahead of the last locomotive.  Despite all changes, an integrated, automated coupler hadn't appeared yet.
(The industry will need one, but it’s a really tough nut to crack.  You can figure ways to interoperate ECP and non ECP equipped cars, couplers are harder.  With all the other changes, I decided the industry would leave this challenge for later. )  The industry still used the "Janey" style coupler and rubber air hose to physically connect the cars of the train. The utilityman was going to use his handheld controller to operate the locomotive and make the set-out for him.  The whole process took less than 10 minutes.

While this was going on, Charlie watched the terminal work.  This was the pilot location for terminal automation.  It was somewhat derisively referred to as "Robo-Terminal".  There basis of it was a smart vision system that coordinated and operated all the machinery of the terminal.  There was a ring of cameras around the periphery of the terminal that kept track of all the hostlers, cranes and rail cars.  It worked like a 2-D PTC system, giving each piece of equipment permission to travel on a defined space in defined direction.  As the equipment moved, the space became open for another equipment route.  Even the men in the terminal wore a device to show them where it was safe to travel and warned them if they were in danger.
(I have actually seen an attempt at a version of this in a CSX intermodal yard in Tampa.  Didn’t have all the bells and whistles, but was a start.) The cranes used visions systems to grab and load the boxes on the train.  There was even an interbox connector robot that would set and and lock the connectors.

The truckers pulled through the gate, but dropped their load just on the other side.  The automated hostler then took the load and either parked it or took it for loading on the train.  On the outbound side, when the trucker arrived for a box, a hostler would get if for him and take it the receiving lane.  There the trucker hitched up, and got rolling.  The equipment was all electric.  Mostly battery powered.  The equipment itself knew when it was time to charge and would automatically route itself to the charging station and plug itself in.
(Super Roomba!)  It looked to Charlie that about 1/3 of the equipment was on charge.  About the only trouble he'd heard about was a disruption during a particularly heavy snow squall, when the cameras couldn't see equipment position and movement. Things ground to a halt in short order.  He wasn't sure what they could do about it, but progress in this area seemed inevitable.

Charlie watched with amazement.  It all looked a bit like "Rollercoaster Tycoon", a computer game from his youth, although a bit less exciting.

The pick-up was a bit trickier.  There was a Toledo block and a Chicago block to pickup and they had to be spliced in.  First, the utility man made the cut just ahead of the locomotive that headed the Toledo block in the train and had Charlie pull ahead a mile and a half.  Then he brought out the pick-up block.  It consisted of a locomotive, 1500' of Chicagos, another locomotive and 2000' of Toledos.   He backed them onto the rest of the train, then had Charlie back down and couple up.  In a matter of minutes, the consist integrity and health was confirmed.  Charlie put all the DPU locomotives back under his control, did the required air brake test to make sure the brakes applied and released on the rear car, and go underway.

Since electrification, they couldn't make the move directly to and from the loading tracks, instead they had built some long support tracks with catenary strung over them and they moved the cars to and from there using battery-powered, hybrid switchers.  A concept that had been in development since the early 2000s was now commonplace in the industry.
(Looks like I missed this one.  They didn’t work out very well.  Maybe all-electric switchers?)

The whole stop at Pitcairn had taken less than 20 minutes - a far cry from the "old days" when this would have taken an hour or more.
(The fastest pick-up, set-off dwell on Conrail in the mid-90s was 45 minutes for some intermodal trains, I believe.  I don’t think NS has anything less than an hour now, and most are 1-1/2 or so.  Airbrake rules make this so time consuming.  You have to make sure you have a solid trainline for commication and enough air pressure to power the brakes.  You can only feed air in so fast without risking all kinds of trouble – false signals.  With ECP, trainline is power only.  Feed it in as fast as you care to, from any point.  No worries about gradient.  Also, “bottling” cars is possible)  The smart trainline and ECP really simplified things!  Pumping air and tying on handbrakes were things of the past.  Good riddance!  If "time is money", then these changes were worth their weight in gold, Charlie thought.

The train rolled out of the controlled siding at 40 mph and was soon up to 60 mph, heading for Pittsburgh.  Although it didn't seem like anything much had changed in the past 25 years, the intermodal world had shifted a great deal.  Commerce around the globe had shifted as cheap labor shifted.  Massive container ships now plied the seas from the coast of Africa carrying consumer goods to all points on the globe.  Africa was the latest and final stop on the global migration of cheap labor.  For the US, this meant traffic a lot of stuff that used to come in through west coast ports, now came through the east coast ports, to be distributed inland.  There was still plenty of traffic at the west coast ports, too, and it all had to be distributed across the nation.  The result was goods flowed in both directions, from coast to coast, and fewer container moved empty. This lowered costs and improved profitability for the railroad, at the same time.
(Two thoughts here.  In the mid-90s, I heard a futurist talk about cheap labor moving around the globe with Africa as the last stop.  As soon as they had transport and communication, they were good to go.  In my life, cheap labor has moved from US to Japan, to Korean, to China, India and Mexico and now, to the rest of the SE Asia.  Africa is the next stop.  China is investing heavily.  Also, in the heyday of the PRR, they made money because they could move finished goods out of NY, Phila and Baltimore in the same box cars they moved the raw material inbound.  The whole thing fell apart when finished goods started moving by truck.)

Traffic volume, increased, too, as the economy grew, plus a bit more.  As the traffic grew, the number of lanes that could be served profitably increased, and traffic was converted from highway to rail.  As the traffic base shifted from bulk to intermodal, the demand on the right of way changed.  Steady speed, not low energy consumption became the driving force.  River grade routes with lots of lower speed curves fell out of favor.  Straighter routes, even with undulating grades, gained favor.  Everywhere, routes were being reshaped for this new world.
(Change or die.)

As the average speeds came up, railroads started competing for even more highway traffic.  In some lanes, they were meeting or beating team drivers.  Further, they were able to compete on shorter and shorter distances.  For example, there was quite a bit of Baltimore and Philadelphia import traffic moving by rail to Pittsburgh - just 300 miles away.  Speed also meant fewer crews and locomotives needed to get a train across the road.  It only took two crews to get from Harrisburg to Chicago and only three to get from Harrisburg to Memphis, down from three and six just a decade ago.
(It is currently four crews NJ to Chicago and 7 from NJ to Memphis.  I based this on 50 mph avg speed for intermodal.  Doable with 60 mph max?  I think so.  Just get rid of the slow stuff!)

 The train rolled through downtown Pittsburgh at 40 mph and then resumed its 60 mph cruise up along the Ohio River. Charlie looked out across what once was sprawling Conway yard.  All that was left was the old hump tower, the car shop that had been converted from the diesel shop built in 2015, and a dozen or so tracks used for local service.  The track were mostly filled with bad-order hopper cars.  Scenes like this always left Charlie a bid depressed. It was soon behind him.
(Conway was kind of the middle of the “X” for Conrail, an important spot in the network.  Both humps had a lot of work to do.  For NS, it’s just a wide spot in the road along the way from Harrisburg to Elkhart.  NS closed one of the humps and with the doubling in size of Bellevue, I figured Conway was doomed to close.  The only way you can make Bellevue pay for itself is to close or reduce other locations.  Conway was a logical choice.  Maybe they were making my “two big hubs” prediction come true!?  Just then, NS announces they are replacing the old roundhouse with a new locomotive shop.  Makes no sense to me… Conway is still a hump yard on NS.)

The train rolled over what was left of the flying junction with the former line toward Youngstown and Lake Erie.  It was all gone, as was the coal traffic to the lake.
(Coal to Ashtabula used to move this way.)

It was just past noon and Charlie was getting hungry.  He usually brought some food from home and put it in the locomotive's cooler, but not today.  There were no good leftovers at home this morning.  So, he decided to order on-line and have it delivered.  Just 15 years ago, this was a "fire-able" offense.  Now, some limited internet use was allowed, but it was strictly monitored and filtered.  The most useful thing he was allowed to do was order "take out.".

He picked a Chipotle that was 20 minutes away in East Palatine, Ohio.  They did a regular business with train crews and he could count on them to meet the train.  As the train rolled into East Palatine at 60 mph, he could make out the delivery van adjacent to the new Market St overpass.  Charlie put on his safety goggles and opened the cab window.  The delivery person was there with his lunch.  He had hoisted the lunch wrapped in it's insulated lunch box.  It was suspended by lightweight plastic rope attached to "Y" shaped rod fashioned out of fiberglass, like a strange fishing pole.  He stuck his arm out and through the loop of rope.  The rope popped off the Y shaped pole as the lunch box was quickly accelerated to 60 mph.  Charlie felt the rope jerk in the crook of his elbow, the reached across and "reeled in" his lunch. "Hooping up orders". 150 year old technology, repurposed!   His burrito and can of coke were in good shape, but he had to let the soda sit for a few minutes before opening...for obvious reasons!
(The whole “meal on the fly” thing is pure silliness.  Still, DoorDash and GrubHub now exist.  They didn’t in 2014, so maybe?)

Charlie was just finishing that Coke, when the train started slowing for Alliance OH.  The junction there required reducing speed to 40 mph to leave the old PRR mainline for the old PRR branch to Cleveland.  NS still used the Conrail names of "Fort Wayne Line" and "Cleveland Line" for these lines
(Why change them?  NS left Conrail’s route names alone in 1999).  Charlie's train was one of the ones set for regular testing at higher speeds on the Cleveland Line.  Seven minutes (actual math done here) after beginning to slow down, the train was accelerating toward 80 mph.

The test consisted of having each freight car gather, store and spin-off more information about the ride quality as well as having the locomotives furnish information about energy consumption.  Charlie knew one of the big concerns was the stability of the trucks and suspension systems on the cars.  The data included accelerometer readings to measure forces on the carbody as well as the frequency and severity of the anti-hunting brake applications.  There wasn't really anything much extra for Charlie to do.  He was to keep a log of what he felt and saw in the cab but he wasn't sure anyone actually ever read it. (Sept 2017 update. Reality closes in...

As the speed indicator was just creeping up on 80 mph, an Amtrak Talgo train flashed by on the opposing track with an audible "whoomp!"  Charlie saw it was Davy Chew
(one my friends from my Mechanical Trainee days last name was Chew) on the mid-day departure out of Cleveland. He guessed it was up to track speed of 110 mph.  He had just under two seconds (more actual math) to glance and inspect the other train.  He keyed the mic, "Everything looks good, Davy!"  A minute later came the reply, "You look good, too, Charlie.  Have a safe trip.  Amtrak #864 out."

The scenery flashed by at 80 mph.  It really didn't look that much different from 60 mph, thought Charlie.  Still, in 40 minutes
(more math) they were slowing for the approach to Cleveland.  The train slowed to 60 mph and swung onto the flying junction to the "New Belt Line".  The new belt followed the general alignment of the old belt line around Cleveland that the New York Central had built, but it had high speed flying junctions at each end and one in the middle, where Charlie was. It was a joint project of the State of Ohio, CSX and NS and it had done quite a bit to smooth traffic and speed trains through Cleveland.  The state got what it wanted, too.  110 mph train operation on shared right of way.  Money had a strange way of "adjusting positions" thought Charlie.  How long ago was it the freight roads say "no way, no how" to anything over 90 mph?(I think this is the kind of horsetrading Amtrak and states are going to have to do with the frt RRs to get more trains on freight road corridors.  Too often elaborate plans get made and communicated to the public before they are anywhere near a deal to make it happen.  Then, guess, what?  The freight RR’s dig in their heels and it all dies.)

At the other end of the New Belt in Berea, the train swung over the Chicago line and then joined it from the right hand side, all without slowing down even one mph.  However, the crew change point was coming up.  It was out on a wide spot on the mainline just this side of Elyria.
(Elevated crew change platforms are a European thing to absolutely minimize the dwell for step on/step off crew changes.  You need this when you have high capacity main lines with high traffic density) The train started slowing as the regenerative/dynamic braking came on from rear to front.  The train glided to a stop at 3:52.  This was three minutes early.  The time by the slow order was reason.  That usually was worth about 5 minutes, although the 80 mph test section typically saved seven or eight, netting out the three minutes to the good. (It’s worth noting here how getting rid of one slower section – running 6 miles at 40 instead of 60 -  is worth nearly as much as 50+ miles of raising track speed from 60 to 80.  Getting rid of the 15 and 20 mph stuff would be huge.)

Charlie had no sooner put his grip down on the high-level crew change platform when outbound engineer Anna Blaha
(a ficticious relative of an actual Conrail Dearborn employee) walked by and said "Hi!" to Charlie.  She was in her mid-twenties, just a three year veteran of the railroad.  They were hiring so many young folk these days it made Charlie feel old. (Hiring seems to always come in waves.  I was among the last hired in the early Conrail era.  I was nearly the youngest in my office for 15 years.)   Charlie's iPad-flex tied him up as soon as he exited the crew platform. (More automated reporting)  He used to have to wait until he got to the lodging to go off duty, but the railroads had negotiated some minor changes about 15 years ago when the rules were changed to fight crew fatigue.  The big change was call times had to come within a four hour window.  When the rule was put in place, you'd have thought it was the end of the world for the railroads, as much as they fought it and moaned about it.  In the end, it was no big deal.  The push to scheduled trains, plus automation made the whole issue moot.  Just about everybody held scheduled jobs these days - "carded jobs" like the old heads he knew used say. (I’m just being optimistic that the RRs will figure out the fatigue issue on their own with a hammer coming down from the FRA.  You just can’t have people working perpetual swing shifts and have them rested and alert all the time.  People need to honor their circadian rhythm.  Naps on duty, alerters et. al. are just window dressing)

The third locomotive on "his" train roared by at about 50 mph as he approached the crew shuttle.  Anna was on her way.  She should be into Chicago in less than 5 hours
(Oops.  Should be 7 hours. 360 miles at 50+ mph), if the terminal was in good shape to take her train.  It usually was - expansion notwithstanding.  There was some talk about, if they could get the speeds up to 80 mph for these intermodal trains, they could make the Cleveland to Chicago pool a "turn". (80 mph running could get time down to 5-1/2 hours – 65 mph avg, so you could “turn”.  The route is straight and flat and fast.) "Faster, cheaper, better" was the mantra.  Charlie wondered if it would ever end. (Supposing 3-1/2 hours from NJ to Harrisburg, an hour at Harrisburg to sort blocks from other eastern origins (Baltimore, Phila, Trenton, multiple NJ terminals), 7 hours to Berea, and 7 more to Chicago.  That’s 940 miles in 18.5 hrs or 50 mph avg.  That gives you a 48 hour cycle on power and equipment, with 5.5 hours dwell on each end.  That knocks the current cycle down by a day or 33%.  Toss on an hour or so for cut-off and grounding, and you start creeping up on team driver competitive times.)

The crew shuttle turned into the 25 year old crew hotel.
(Once upon a time, RRs use RR YMCAs to put up their crews overnight.  The last one I knew on Conrail was in Collinwood.  It became easier and cheaper to just put crews up in hotels that had 24 hour dining, so the RR YMCAs died out.  Eventually, finding hotels with 24 hour dining at or adjacent became problematic, and expensive.  So, NS started building their own crew quarters with attached restaurants.  I imagined one in Cleveland built in 2015) Charlie remembered it when it was brand-new.  It was still in pretty good shape, for as old as it was, just like Charlie liked to think about himself!

The lobby was decorated with pictures and icons of railroading-past.  A steam locomotive, N&W 611, posed on a turntable, a signal tower with the operator on the ground "hooping up" an order, a caboose with the conductor's head hanging out of the cupola window, an EOT hung off the rear coupler of a box car, paper train sheets with penciled stringlines, a signal bridge with three-headed signal masts all at stop except for a lone "high green".

"High green".  Charlie remembered calling signals when he started.  He must have said "high green" 10,000 times!  He hadn't said it once in the last 5 years.  Almost unimaginable.  What was next?  Charlie had no idea. Change had had a "high green" his whole career, he decided.  No reason progress was going to stop now.  Charlie walked past the entrance kiosk that checked him in on the fly and headed to beans.  Lasagna.  Yum.
(Who doesn’t like lasagna?)

(Some day, I’m going to crack open the track charts, ETTs and Google maps and see what might be done on the route, including some “spreadsheet” style train performance calculator action.  Stay tuned….)