Dear Mr Anderson

[bratkinson]

I don't know about other regular Amtrak passengers, but as for me, I get frustrated with the numerous inconsistencies I find on Amtrak. In the past 10 years or so, (I've been riding Amtrak fairly often since '75), I've encountered everything from 5 or 6 'different' business class accomodations to great/cold/angry/terrible/invisible OBS crewmembers and even missing blankets/pillows/towels/bars of soap in the shower in sleepers.

Today, I had enough. While waiting to board Acela 2172 at WAS in the ClubAcela, I decided to write Mr Anderson about one of my pet peeves I've witnessed since the Acela started running...”Where's Waldo?....er....Acela First Class car?” Using Open Office Writer (does everything Microsoft Word does for free!), I put together a fairly lengthy email to Mr Anderson. After proofing it several times on the train, I'm now ready to send it. So, I did a quick Google and found Amtrak Customer Service Email and I was off to the races.

NOT SO FAST, COWBOY!!!

After filling in all the required info to send the email and getting to the 'enter text' box, there was a light grey message indicating 'only send one paragraph as breaks cause processing errors' or something like that. ARE YOU KIDDING ME?????

So, I 'deformatted'/'deparagraphed' my email and put the following all caps preface in the front:

PARAGRAPH FORMATTING DESTROYED PER YOUR EMAIL SITE INDICATING -DO NOT USE PARAGRAPHS- PLEASE REPARAGRAPH THIS AT EACH DOUBLE DASH – IDEALLY, PLEASE FORWARD THIS TO MR ANDERSON –

Below is the 'real' properly formatted email I sent. I wonder what the response will be.

Dear Mr. Anderson

As a regular Amtrak passenger, I'm writing to you about getting a 'pet peeve' of mine resolved. In particular, why aren't all Acela trains facing the same direction? As bi-directional trains, the First Class car should be consistently at the Boston or Washington end of the train.

I have little doubt that Amtrak staff sometimes goes above and beyond to ensure all NEC regional trains as well as all LD trains are consistently ordered from front to back. Whether trains get wyed or a trip around a reverse loop, the business class car on NEC regional trains is always at the rear, and on LD trains, sleepers and other cars consistently positioned on each train. So why not Acela?

Recognizing that Acela passengers pay the highest per-seat-mile fares, it would make sense that the First Class car of each Acela would consistently be on the eastern end of the train. Although I rarely travel on the Acela due to cost, until 2 weeks ago, it's been years since I last rode one. I used an upgrade coupon to ride First Class several times. I remember at least twice the First Class car was not at the same end as my previous Acela trip. I cannot imagine the frustration regular Acela First Class passengers must feel when, after waiting on the 'usual' location on the platform for their First Class car to arrive, they find out it's at the other end of the train!

The introduction of individually reserved seating has, in my opinion, caused considerable stress among Amtraks' premier passengers. When FC seats were unreserved, one could easily choose a seat facing either direction. Now that seats are assigned, and although the Amtrak web site showed the direction of travel when choosing a seat, there's no reasonable assurance that when the train originates or arrives that it will actually be faced in the expected direction! Even more frustrating is that the Amtrak web site itself indicates the Acelas are not consistently faced. https://www.amtrak.com/first-class-seat

“The Acela Express First class car is always either the first or the last car on the six-car train and is configured with both forward- and rear-facing seats. Some of the single seats are rotated by the train crew and are always forward facing.

We will do our best to predict and display your direction of travel when you view your train's seat diagram on Amtrak.com or the app. A schedule change, equipment change or other unforeseen circumstance may cause the planned location of the First class car to change before departure. You can change your seat at any time after booking, even while you are en route after your train has departed. The Conductor can also assist you with changing your seat onboard, subject to availability.”

If that were true, while riding train 141 today, why did I observe Acela 2175 at Philadelphia with the First Class car in the rear and Acela 2164 at Baltimore with the First Class car in the rear? One of those trains needs to be turned!

Instead of showing a 'last car/first car' seating chart, it should instead be marked 'Odd numbered trains'/'Even numbered trains' with parenthetical indication of first/last car position and/or final destination city. In short, if Amtrak can always turn regional and LD trains, why can't they NOT TURN Acelas?

As a retired, leisure travel only passenger, ordinarily, I would not be writing this email except for the fact that I've booked several Acela trips to put me 'over the top' to reach AGR Select Executive status for the first time ever. I have little doubt that other Acela First Class passengers, many likely to be Select Executive passengers as well, have the same First Class positioning frustrations as do I.

Respectfully,

Bruce R Atkinson

 

[Acela150]

You do realize how many Trainsets are used each day correct? And how they're used in service?

This will never happen. Acela was designed with two Power Cars on each end to ensure a quick turn in WAS and BOS.

You clearly have no idea how the Acela rotation is used to be complaining about it.

 

[Dakota 400]

I understand your message. Inconsistency in product delivery in some of the travel marketplace today is not exclusive to Amtrak. As a Holland America Line frequent guest, their inconsistency of the guest experience has become what one ought to expect. Such varies by ship to ship.

We travelers, today, have to decide whether the experience that we receive, if less than expected, is a "glass that is half full or a glass that is half empty".

Should we have to make such a decision? Surely not! But, unfortunately, that does seem to be the reality of 2018.

 

[cpotisch]

I've gotta say that in my opinion, Acela train sets being oriented unpredictably is one of the less glaring issues I've had with Amtrak of late...

 

[Acela150]

I've gotta say, in my opinion Acela train sets being oriented unpredictably is one of the less glaring issues I've had with Amtrak of late...

I do have to say, that of all things to complain to Richard Anderson about. It's the only thing that can't be changed. Complain about the new dining or something along the lines there of.

 

[cpotisch]

I've gotta say, in my opinion Acela train sets being oriented unpredictably is one of the less glaring issues I've had with Amtrak of late...

I do have to say, that of all things to complain to Richard Anderson about. It's the only thing that can't be changed. Complain about the new dining or something along the lines there of.

Thank you! It really isn't that particularly important as these things go, and can't be changed. I like to think that poor food service and the loss of checked baggage at so many stations is a bigger issue than facing the wrong way in First Class on one route no more than 50% of the time.

I imagine that the CEO of Amtrak will have bigger (not necessarily better) things on his mind than Acela consist orientation, and no, that is not a cue for someone to jump in and say "like destroying Amtrak?" or something like that.

 

[Triley]

The problem is... The trains actually do anywhere from a 1 to 2hr turn at their respective turnaround points, which also includes the time to detrain and board passengers, service the train, and strip/load any food supplies that require to do so. Many Acelas turn in DC with a roughly 2hr layover but if one set gets shopped, usually everything gets pushed up to become a 1hr layover, if they're on time.

The Regionals are able to be looped or wyed because they sit 3 or more hours at their turnaround points. If the Acelas were to do the same, there would have to be a cut in frequency.

Anyway, for regular Acela riders, they know what side first class is generally on, as it truly is consistent, provided there are no mechanical issues. First class should be on the trailing end of the trainset, from the equipment's initial departure for that day. For example, 2172 should be on the leading end, since that equipment originated in Boston as 2155, where it was on the trailing end.

 

[bratkinson]

Apparently my post was misunderstood. I'm requesting they NOT turn the trains around, thereby keeping the First Class car at the Boston end of the train (or WAS end) on all trains. THAT'S what they are designed to do...NOT be turned around!

As for 'how many trainsets' and 'how they are used' goes, as I see it (If I read the timetable correctly) the only Acela service is WAS->NYP or BOS, and BOS->NYP or WAS. 'Straight line' railroading in my opinion. There's no need to ever turn the trains. For those trains not cleaned & turned at the platform, the trip to the yard should be 'out and back' only...no loops or wyes. And if a trainset is rotated out for maintenance or repairs, it should be a requirement for the shop crew to ensure the repaired and/or replacement train is properly oriented rather than 'however it turns out'.

Are there 'bigger fish to fry' in Mr Andersons' lap? Of course there are. But with a single memo or directive, the Acela orientation problem is quickly and easily solved. Crew problems and 'contemporary dining' letters likely flood in every day, and the solution(s) are neither quickly resolved, nor cheaply implemented. Orienting the Acelas is, at most, a handful of shop or operating crew hours to initially get them all pointed the same, and almost non-existent crew hours to keep them that way.

 

[Ryan]

I couldn’t get through the massive wall of text to understand that. Chances are, nobody at Amtrak will either.

Your elegant plan falls apart when a defect in a cab necessitates a trip around the loop to put the functional cab in front, or any other operational situation that makes a trip around the best path forward.

 

[bratkinson]

I couldn’t get through the massive wall of text to understand that. Chances are, nobody at Amtrak will either.

Your elegant plan falls apart when a defect in a cab necessitates a trip around the loop to put the functional cab in front, or any other operational situation that makes a trip around the best path forward.

Granted, there are times due to rarely occurring situations that things get screwed up. But in each case, after the passengers deboard at the terminal station, the train should be replaced with one facing properly, or repaired and turned to face properly. Letting a train run 'wrong end forward' for more than a day is inexcusable in my opinion. But then, I'm a recovering perfectionist. As a mainframe computer consultant/analyst/programmer for over 30 years, if someones' paycheck got screwed up or the books didn't balance in the accounting system, it got fixed IMMEDIATELY. The unfortunate reality these days is that everyone seems content with 'so-so' performance, buggy software, and deadlines not met. That was NOT how I did my job!

 

[west point]

Believe that the configuration of BOS South Hampton (?) yard can cause train sets to swap ends randomly. However there is a good reason for not keeping the first class end always pointing on way , say toward BOS. Wheel wear and equipment wear can be different for what rail it runs on due to rail curvature, bridges, tunnels, etc.

 

[CCC1007]

Not to mention that the servicing of the sets in NYC at sunnyside yard necessitates that the train take a loop to enter the yard or cross all inbound traffic also makes it hard to keep the first class car on a specific end of the train.

 

[Palmetto]

Believe that the configuration of BOS South Hampton (?) yard can cause train sets to swap ends randomly. However there is a good reason for not keeping the first class end always pointing on way , say toward BOS. Wheel wear and equipment wear can be different for what rail it runs on due to rail curvature, bridges, tunnels, etc.

There is a loop at Southampton, and a loop's [or a wye's] function is to keep the head end arriving at the terminal in the same place when it leaves the terminal--at the head end.

 

[Skyline]

Mr. Atkinson,

As you have been riding Amtrak exactly a year longer than me, I think we might possibly be in agreement that we LOVED the heritage cars Amtrak inherited precisely because of the diversity. Very few trains' consists were, um, consistent. It was an exciting time to board a train with our car spotter books and learn the origins of the cars on that day's train. And explore every nook and cranny to find unique hideaways, talk to old-timers about the glory days, etc.

Frankly, since words like Amfleet, Turbo, Superliner, and Viewliner entered our vocabulary I think the experience has suffered. Sure, the heritage fleet was unpredictable; important elements were prone to malfunction sometimes. But many of the trips I took in the 1970s were flawless as well.

My point is: Why whine about inconsistency? Diversity rules! Too bad we don't still have more of that.

 

[cpotisch]

Mr. Atkinson,

As you have been riding Amtrak exactly a year longer than me, I think we might possibly be in agreement that we LOVED the heritage cars Amtrak inherited precisely because of the diversity. Very few trains' consists were, um, consistent. It was an exciting time to board a train with our car spotter books and learn the origins of the cars on that day's train. And explore every nook and cranny to find unique hideaways, talk to old-timers about the glory days, etc.

Frankly, since words like Amfleet, Turbo, Superliner, and Viewliner entered our vocabulary I think the experience has suffered. Sure, the heritage fleet was unpredictable; important elements were prone to malfunction sometimes. But many of the trips I took in the 1970s were flawless as well.

My point is: Why whine about inconsistency? Diversity rules! Too bad we don't still have more of that.

I think that there's a big difference between consist inconsistency/diversity and service inconsistency. And I think that having 17 different varieties of Business Class or being guaranteed a certain accommodation and getting a different one ends up being a bigger issue for most passengers than whether or not you get variety in the consists themselves.

 

[Devil's Advocate]

I couldn’t get through the massive wall of text to understand that. Chances are, nobody at Amtrak will either. Your elegant plan falls apart when a defect in a cab necessitates a trip around the loop to put the functional cab in front, or any other operational situation that makes a trip around the best path forward.

Granted, there are times due to rarely occurring situations that things get screwed up. But in each case, after the passengers deboard at the terminal station, the train should be replaced with one facing properly, or repaired and turned to face properly. Letting a train run 'wrong end forward' for more than a day is inexcusable in my opinion. But then, I'm a recovering perfectionist. As a mainframe computer consultant/analyst/programmer for over 30 years, if someones' paycheck got screwed up or the books didn't balance in the accounting system, it got fixed IMMEDIATELY. The unfortunate reality these days is that everyone seems content with 'so-so' performance, buggy software, and deadlines not met. That was NOT how I did my job!

Although this particular issue doesn't really impact me I do wonder how the people defending Amtrak would explain how other passenger railroads can put their premium cars in the same place at the same time nearly every single trip while Amtrak struggles in vain just to keep their cabs functional.

 

[jis]

Leaving aside the defending or attacking Amtrak aspects of the discussion for the moment, let us focus on the technical issue.

This problem is peculiar to the way in which single ended power assumption is baked solidly into the track layout in New York. Unfortunately, there is no easy way to get a train from Penn Station into Sunnyside Yard without running it around the loop. That is exactly what you don't want to do if you want to operate a "power at both end" train efficiently. Trying to get the Acelas into Sunnyside will involve essentially halting some of the inbound service from the east, both LIRR and Amtrak while the Acela crosses across the westbound tracks to get into the yard. While this could conceivably be handled during low traffic times, it would be extremely disruptive at high traffic times when trains are arriving and departing literally within sight of the tail lamp of the preceding train. Breaking the flow will reduce effective capacity that would not be acceptable, merely to get the cars in the right order.

The only way to avoid running a train to Sunnyside is to turn it in the station, and unfortunately, Amtrak's normal operating procedures does not apparently allows such in a reasonably short time.

Now given that one is stuck with turning New York terminating trains, the other option is to turn all trains at every terminating point. This means increasing the turning time in Boston and Washington uniformly to two, perhaps three hours which is what it might take to wye/loop the train after pulling it out of the station and then bringing it back into the station. Both removing from the station and deadheading back into the station generally are lower priority moves that trains in service which adds to the time needed. This would be instead of turning trains in the station which can be done in less than an hour, and would essentially require two or three more Acela sets to maintain the level of service that is currently run.

The other alternative is to spend something in the order of $50 to $100 million dollars minimally to modify the track layout at F interlocking to allow deadhead moves from Penn Station into Sunnyside fro the F interlocking end (west end) of Sunnyside, bypassing the loop, and avoiding any conflicting moves. This ill involve building at least one somewhat complicated underpass. If that were available, then Acelas (and other double headed trains could be operated most efficiently and also guarantee same location of cars in consists, with same orientation of consist all the time.

So, that in a nutshell is the issue, and hopefully answers DA's question too.

The choice at present is between running fewer scheduled runs while guaranteeing cars in the same location all the time vs. running more frequent service while suffering through not having the cars at the same location all the time. To me at least the choice seems rather obvious, but of course others may disagree.

If the explanation is not clear enough, please let me know and I will try again to do it better.

 

[Devil's Advocate]

(1) The only way to avoid running a train to Sunnyside is to turn it in the station, and unfortunately, Amtrak's normal operating procedures does not apparently allows such in a reasonably short time.

(2) Now given that one is stuck with turning New York terminating trains, the other option is to turn all trains at every terminating point. [SIZE=11pt]This means increasing the turning time in Boston and Washington uniformly to two, perhaps three hours which is what it might take to wye/loop the train after pulling it out of the station and then bringing it back into the station. [/SIZE]

(3) The other alternative is to spend something in the order of $50 to $100 million dollars minimally to modify the track layout at F interlocking to allow deadhead moves from Penn Station into Sunnyside fro the F interlocking end (west end) of Sunnyside, bypassing the loop, and avoiding any conflicting moves.

For me the main issue is predictability. If the premium car is (almost) always at the front or back for a given station and direction then that would seem reasonable to me considering the other complications already in play. I don't fault Amtrak for not building new tracks and I don't expect them to start manually turning their trains at every terminus, but I do wonder why they can't know that a premium car will be at the front or the back in a dependable enough fashion to put up semi-static signage. If the trains are currently being turned at Sunnyside and simply reversing everywhere else how are they getting out of sync in the first place? I don't doubt that sometimes it's due to exceptional circumstances like inoperative cabs, but I'm guessing it's primarily due to lack of attention to detail when cycling sets in and out of revenue service for maintenance or schedule recovery reasons. Those sound like factors that should be within Amtrak's reasonable control.

 

[bratkinson]

I was unaware of the 'forced' trip through the loop at Sunnyside. However, unless my timetable reading skills are failing me, there is only ONE Acela that originates at NYP Mondays-Saturdays (2190 @6:15AM M-F and 2290 @8:00AM Sat, per current 'Northeast Corridor 2' timetable on Amtrak.com) and I am unable to find ANY Acelas that terminate there! Maybe they deadhead one in? Regardless, would it be that difficult to run originating Acelas around the loop a 2nd time before entering NYP to depart to BOS?

And as for wheel wear and other 'same direction all the time' issues, it should be noted that the train travel direction is reversed every time the train originates. However, hitting the same switch, the same direction, day after day will ultimately cause issues. So, perhaps once every 3 years, turn each Acela so the first class car is now at the WAS end of the train. Of course, this event would be duly 'advertised' 30 days ahead of time with emails to all passengers that have traveled on an Acela in the past 12 months or so as well as to all currently booked Acela passengers, AND popups when booking Acela trips 30 days before the reversal and 60 days afterwards. Regular travelers on NEC regional trains will recall that about 2 years ago, all NEC regionals were turned to put the business class and quiet cars last on the train, instead of at front. This change was 'haralded' with emails and a flurry of railfan website chatter.

edit: Devils Advocate, I 100% agree with your post 2 minutes before mine!

 

[jis]

(1) The only way to avoid running a train to Sunnyside is to turn it in the station, and unfortunately, Amtrak's normal operating procedures does not apparently allows such in a reasonably short time.

(2) Now given that one is stuck with turning New York terminating trains, the other option is to turn all trains at every terminating point. [SIZE=11pt]This means increasing the turning time in Boston and Washington uniformly to two, perhaps three hours which is what it might take to wye/loop the train after pulling it out of the station and then bringing it back into the station. [/SIZE]

(3) The other alternative is to spend something in the order of $50 to $100 million dollars minimally to modify the track layout at F interlocking to allow deadhead moves from Penn Station into Sunnyside fro the F interlocking end (west end) of Sunnyside, bypassing the loop, and avoiding any conflicting moves.

For me the main issue is predictability. If the premium car is (almost) always at the front or back for a given station and direction then that would seem reasonable to me considering the other complications already in play. I don't fault Amtrak for not building new tracks and I don't expect them to start manually turning their trains at every terminus, but I do wonder why they can't know that a premium car will be at the front or the back in a dependable enough fashion to put up semi-static signage. If the trains are currently being turned at Sunnyside and simply reversing everywhere else how are they getting out of sync in the first place? I don't doubt that sometimes it's due to exceptional circumstances like inoperative cabs, but I'm guessing it's primarily due to lack of attention to detail when cycling sets in and out of revenue service for maintenance or schedule recovery reasons. Those sound like factors that should be within Amtrak's reasonable control.

Frankly I don't think they particularly try to keep track of what is actually happening and just operate transactionally rather than holistically based on a policy, as they should.

I was hoping that as they go into reserved assigned seats they'd pay more attention, but apparently not.

It still remains true that once things get out of synch it is quite non-trivial tog et them back in synch. One possibility is when a train that was not supposed to be turned gets turned, spend the extra time to turn it twice to set things straight, rather than let the unplanned situation propagate. But there is still a cost of doing that.

Withe current timetables running almost all Acelas through New York, they should be able to get a better handle on this, if they bothered to do so.

Unfortunately, I have to agree with you that there are way too many Amtrak apologists who will come up with corner cases that should not happen often to justify the total lack of a policy. You cannot plan a schedule based on the assumption that some train will derail en route somewhere, and yet that is always a possibility for some unfortunate run on a particular day. There should be an IROPS plan for handling the situation when a train gets flipped due to some unplanned incident, but that possibility should not prevent one from having a normal ops plan that makes sense.

 

[Acela150]

However, unless my timetable reading skills are failing me, there is only ONE Acela that originates at NYP Mondays-Saturdays (2190 @6:15AM M-F and 2290 @8:00AM Sat, per current 'Northeast Corridor 2' timetable on Amtrak.com) and I am unable to find ANY Acelas that terminate there!

You're absolutely blowing my mind right now. And not in the good way either. Your timetable reading skills are almost non existent. Considering you missed an entire timetable that covers only WAS-NYP.

There are 6 trains that start in NYP and go to WAS and 8 trains that start in WAS and end their trips in NYP.

 

[Trogdor]

If the trains are currently being turned at Sunnyside and simply reversing everywhere else how are they getting out of sync in the first place?

Set 1 departs WAS heading to BOS with the First Class car in front (First on the Boston end)

Set 2 departs WAS heading to NYP with the First Class car in front (First on the Boston end)

Set 1 turns back and heads to WAS with the First Class car in back (First on the Boston end)

Set 2 loops at Sunnyside, heads to WAS with the First Class car in front (First on the Washington end)

Now it's out of sync.

 

[Acela150]

I have to double check things but I believe that all turns in NYP have to go through Sunneyside. I'm at work right now and don't have the timetable and consist guide near me.

 

[jis]

However, unless my timetable reading skills are failing me, there is only ONE Acela that originates at NYP Mondays-Saturdays (2190 @6:15AM M-F and 2290 @8:00AM Sat, per current 'Northeast Corridor 2' timetable on Amtrak.com) and I am unable to find ANY Acelas that terminate there!

You're absolutely blowing my mind right now. And not in the good way either. Your timetable reading skills are almost non existent. Considering you missed an entire timetable that covers only WAS-NYP.

There are 6 trains that start in NYP and go to WAS and 8 trains that start in WAS and end their trips in NYP.

Those trains precisely are the problem. They have to either go around the loop to get into Sunnyside or they have to be turned in Penn Station, which currently is not practical.

The WAS - BOS trains are not the problem.

The fact that New York turns then get turned into Boston trains at Washington propagate the problem to the Boston runs. That is where the issue of more efficient consist usage and quick turns vs. predictable car location becomes the tradeoff, and with limited resource of Acela consists the choice of more service rather than consistent car location is quite obvious to me at least. It is not a good idea to split an already small fleet into two separate subfleets if one wishes to provide most service. So currently the only reasonable choice would seem to be to consistently notify car locations on platforms instead of being able to have the same car at the same location all the time. Just IMHO of course.

Now if the consist links for the NYP turns can be consistently maintained then one could put together a plan allowing notification of location of cars.

Indian Railways has Car Designator electronic signs for each car on the platform, and they light up with the car designator before the train arrives identifying location of each car. Several stations on the NEC have those too, so even if train orientation cannot be guaranteed car locations can be identified on the platform before the train arrives.

Even the much vaunted DB on many occasions gets their trains flipped (specially bidirectional ICE consists) so the car location on the paper charts do not match what arrives. So things do go wrong elsewhere too.

 

[Acela150]

However, unless my timetable reading skills are failing me, there is only ONE Acela that originates at NYP Mondays-Saturdays (2190 @6:15AM M-F and 2290 @8:00AM Sat, per current 'Northeast Corridor 2' timetable on Amtrak.com) and I am unable to find ANY Acelas that terminate there!

You're absolutely blowing my mind right now. And not in the good way either. Your timetable reading skills are almost non existent. Considering you missed an entire timetable that covers only WAS-NYP.

There are 6 trains that start in NYP and go to WAS and 8 trains that start in WAS and end their trips in NYP.

Those trains precisely are the problem. They have to either go around the loop to get into Sunnyside or they have to be turned in Penn Station, which currently is not practical.

Other then track space being an issue at NYP, I believe that Trains can not drop Pantographs within the overbuild of Penn Station.

A select few keystone trains will turn in NYP to go back to HAR. I believe that those trains will use tracks 1-4 since they are stub tracks. But like everything NYP, it's subject to track availability.