New M-9 cars enter passenger service

The LIRR’s new M-9 electric multiple unit cars entered revenue service for the first time Wednesday, September 11.  The first new passenger cars to enter service on the LIRR since 2002 made their initial trip on train 1621, the 6:50a from Huntington to Hunterspoint Avenue on Wednesday.  The cars are being manufactured by Kawasaki Rail Car (KRC) under a contract first awarded in 2013.  The first set of 14 pilot cars were manufactured in Japan, but the balance of the order will be built at Kawasaki’s Lincoln, Nebraska plant (with final assembly taking place in Yonkers before delivery to Long Island).

Passengers onboard the first revenue service run of the LIRR's new M-9 cars (Photo: The LIRR Today)

The M-9 fleet, consisting of a base order of 92 cars and a consolidated 110 car option (for a grand total of 202 cars for the LIRR) is costing $734.2 million, which averages out to $3.63 million per car.  The entire procurement has been significantly delayed—when the MTA first committed funding for the new cars as part of the 2013 amendment to their 2010-2014 Capital Program, the agency said that the base order cars would be completed (i.e. built, delivered, accepted, and in revenue service) by January 2017.  But a host of delays at nearly every step of the way from both the MTA and Kawasaki pushed the completion date back again and again.  Most recently, issues with testing delayed the entry into revenue service of the cars by roughly 3 months (the railroad originally said the cars would enter service in May).

8 cars initially entered service, with that figure expected to rise gradually over time until the entire 92 base order car is in service around May 2020 (for a total delay of 3 years, 4 months), and the 110 option cars delivered and in passenger service by March 2021.  While the testing issues initially delayed the start of passenger service this summer, KRC appears to have continued assembling new cars.  At July’s MTA Board meeting, LIRR President Phillip Eng said that at that time 41 cars were in New York, either ready or were awaiting final assembly in Yonkers.  Customarily, the initial pilot cars are put through the testing process and passenger service is launched, then there is a lull while the project team sees how the cars respond in the wild.  Often times additional design changes result from the first few weeks of passenger service, and those are incorporated into the main production run before that goes too far.  It’s unclear if the same thing will happen here...if the LIRR went straight on with deliveries, they would have to accept roughly 10 per month to make that March 2021 date.

Below are some initial impressions of the LIRR's new EMU fleet...

New train smell

New M-9 cars at Huntington for their first run (Photo: The LIRR Today)

Many passengers found something new as they approached the platform to find their train would be one of the ones with the new cars, and certainly a decent chunk of them noticed that they were riding on something new.  The initial set of cars has been making a handful of trips each day, visiting Huntington, Hempstead, and Babylon on the first day, and Brooklyn, Far Rockaway, and Ronkonkoma on Thursday.

I rode the cars on their first day in revenue service, and the trains were neat, clean, and bright with no odd noises, broken doors, or missing stickers.  The ride quality was just ok...with the LIRR’s track quality being what it is, the cars still bounced around a fair bit.  The M-9 cars would definitely fail the coin balance test.  While we will see how they perform over time, my initial impression is the ride quality is roughly the same as the M-7 fleet (or just slightly better), and not nearly as smooth as the C-3 cars.

Acceleration and deceleration was also just so-so.  The train seemed to be as slow as the M-7's getting out of stations, and not nearly as quick as, say, NJT’s Arrow III cars, which rocket up to top speed in roughly 60 seconds.  Poor acceleration seems to continue to be an issue with the MTA’s EMU rail fleets, with stop penalties ranging 30 seconds or more than modern EMU’s⁠—which can add a lot of time to schedules over the course of an entire run.

Fewer noises

The new cars are pretty quiet.  Onboard the whirring of the HVAC system is comparable to the M-7's, and probably is what it is with such little space above the ceilings, but the M-9's are a fair bit quieter overall.  The door opening and closing mechanisms are nearly silent (no loud whirring like you hear on the M-7's), and the propulsion is barely noticeable inside the car (while the M-7's make quite a lot of noise every time the stop and start).

While nothing will likely ever beat the stealth of the M-1 fleet (or even the M-3's), which can slither into and out of a station with the HVAC fans making the only noticeable noise, the cars do appear to move markedly quieter from the outside as well, which is good news for those living near stations.

Seating squeeze

The M-9 cars have closed-loop armrests (Photo: LIRR)

The seats on the car are more or less exactly the same as the M-7 fleet, they just switched the accent color to light blue.  Comfort, angle, pitch, etc. all seem very similar to the M-7's, so there was minimal improvement overall.  One change that was well-noticed was the substitution of a closed-loop armrest.  The railroads have had to bear the fury of riders for many years as the armrests on the M-7 cars have a tendency to slip into people’s pants pockets and rip them open.  Years later, figuring it would be cheaper to just reimburse people who pursued claims for damaged clothing instead of replacing the armrests with a better design, people still have to watch out for their pants pockets to this day.

The seat design has not really changed since the M-7 cars
(Photo: The LIRR Today)

The M-9 cars have 217 seats per pair: 114 in the A-car (without the bathroom), and 103 in the B car (with the bathroom).  The railroad has said the new fleet comes with 4-6 extra seats per pair—which is not really right.  It’s 4 extra seats per car (217 - 213 = 4), not 6 (unless you’re a group of toddlers).  And even then, I think that’s a big stretch...

The number of actual full height/width seats is exactly the same on the M-9's as they were on the M-7's (195 full seats).  They squeezed out those additional 4 seats by reshuffling the arrangement of the fold-down flop seats in the accessible seating areas.  These seats are barely seats...they are a fold down bench that sits lower than the typical seats and have little cushioning on the bottom, and very little at the back.  It’s not unusual for the cushions to disappear altogether or for the seats to get stuck in the upright position, making the seats completely unusable.  The seats also have to be vacated in the event someone with a wheelchair needs to board and occupy that space.  They are also narrower than typical seats, especially the ones they turned sideways (where your back is running lengthwise to the train), which are significantly narrower than typical seats.

(Photo: The LIRR Today)

Flipping one set of fold-down seats per car to run lengthwise and extending another set of fold-down seats from two to “three” people wide buys four additional seats per car.  The railroad also jammed in another set of fold down seats near the F-end of the A cars, which I think is definitely cheating.  By time the seat folds down, there’s practically no room left for people to put their legs, nevermind two sets of legs, as there is another set of full seats directly across.

(Photo: The LIRR Today)

The M-9's also include a significant reduction in the quality of the two jump seats per car.  These are the "two” seats in the cab area opposite the engineer’s cab.  While some prefer the jump seats, as they are more secluded and separate from the rest of the car (with the equipment and electrical lockers separating them from the other seats), I also question their ability to count as seats.  They are not really wide enough for two adults to sit in the first place, and then they are often locked behind the cab doors and as a result are inaccessible to passengers.  Some crews will even remove the seat cushion and hide it away to discourage people from sitting there, and it’s also not unusual to find these seats broken or locked in the up position.

The jump seats on the M-9's are made of hard plastic and are set lower to the floor.  The back also sits very low, making sitting in these seats very awkward and uncomfortable.  Whether this was done to actively discourage passengers from sitting here, or for some other silly reason, I think it’s a big stretch to count these as real seats.

View the M-9 seating chart here.

Seat counts by equipment type (2-car pair)

	M-1	M-3	M-3 w/ PTC	M-7	M-7A	M-9
End seats	175	173	169	144	138	142
Middle/facing seats	63	59	58	51	57	53
Fold-down seats	2	2	2	14	14	18
Jump seats	0	0	0	4	4	4
Total seats	240	234	229	213	213	217

Overall, the seating story on the M-9's are either no better, or even slightly worse, than the M-7's.  There are actually fewer end seats per pair (not middle or facing seats) than there were on the M-7's, and the cars still have significantly fewer seats per pair than the M-1's and M-3's.

Poor communication facilities

The new M-9 cars include outdated dot matrix displays (The LIRR Today)

Overall, the communications facilities on the new M-9’s are rather poor, and very outdated.  There is very little improvement in the functionality of the Automated Station Identification (ASI) system over the M-7 fleet, and the functionality of the system continues to be extremely limited.  Not more than 30 seconds after I sat down on the first trip did the conductor come over the PA with a non-emergency announcement, and the conductor also preempted the ASI system at the first two stops with an announcement inaccessible to people with disabilities.  ADA requirements for effective communication require business to make communication accessible to people with disabilities.  Manual train crew announcements that have no real-time text components are inaccessible to people with hearing impairments.

The LIRR appears to have put no effort into automating 100% of its announcements so that audio and text can play automatically and simultaneously, making the announcement content fully accessible to everyone (including those who are hearing or visually impaired).  Thoroughly reviewing and refining all announcement scripts, and then building that logic into an updated ASI system should have been a fairly straightforward task.  Not only would this have brought some very badly needed consistency and reliability to passenger announcements (every crew does things a different way, and many still don’t bother to make any announcements at all), but it would have also made information more accessible.

Announcements made manually by train crews should be extremely limited to emergency situations only…all other announcements should either be pre-defined and automated or be typed into the system and recited using a text to speech engine.  But nearly 30 years after the ADA was signed into law, the needs of people with disabilities continue to be ignored by the MTA.

One if the minor additions that was made was a new announcement saying something to the effect of “Change at Jamaica for the AirTrain to Kennedy Airport”…but even that quickly proved to be worthless.  At the time the train was arriving Jamaica Wednesday, AirTrain JFK service was suspended and substitute bus service was operating to the terminals.  The conductor came on and announced that, which was nice, but was then contradicted less than a minute later by the audio announcement saying to take the AirTrain.  Static announcements like these are pretty unhelpful.  All automated announcement information should be constantly updated to respond to service changes in real-time.  Information on service status of connecting lines (e.g. if a subway GO means a particular line is not stopping at a transfer point or the AirTrain is not running), connecting train and track information, and any other changes that might come up during the trip should be considered by and incorporated into the announcement logic, that way passengers are always being reliably given the most up-to-date information.  This is entirely possible with modern technology…or even 2002 technology.  The M-7 cars all have cell cards that make it possible for them to stay connected to the internet, where the systems could update and account for these changes on-the-fly.  But instead of investing a little effort up front to standardize, the LIRR seems content to leave communication up to conductors to do their jobs—or just not.  And, in any case, the railroads' conductors could never communicate as reliably and accurately as a robust automated system could.

Passenger information displays on modern railcars have the ability to show
much more detailed information (and are more flexible). (Photo: Siemens)

The PA system seems to function and sound more or less the same as the M-7 fleet (there is an additional speaker in the vestibules).  The displays also represent little to no improvement over the M-7’s.  They are still dot matrix LED displays, and the text might even be a little smaller than it is now.  They pale in comparison to what’s been seen in modern rolling stock orders…the Class 700 Desiro City cars ordered around the same time feature wide LCD screens that can show a wide variety of content, including train loading information, restroom locations, and the status of service on connecting Tube lines…none of which can be done on the LIRR’s brand new $3.62 million cars.

(Photo: The LIRR Today)

The cars feature destination signs on the front of the train, which is nice, but like the M-8 cars, the text is very small, limiting their utility as a proper headsign.

The M-9 cars also indicate “Car __ of __” on the LED displays, a badly needed feature.  While the MTA boasted in their press release the M-9’s were the first cars in the region to have this feature, it’s really an indictment of how far behind the times we are.  I first saw this feature while riding a train in the UK more than 15 years ago, and we are just starting to see it here.

Signage on this train in the UK leaves no question about where you need
to go to get off of the train...and information is localized to each car.
(Photo: Televic Rail)

Communication of stopping locations continues to be a tremendous issue for the LIRR.  Just last night on my train home, three people missed their stop at Lynbrook because they had no way of knowing what car they were in prior to arrival (and the conductor didn’t say anything when checking their tickets), and we were delayed at least 60 seconds additional at Amtiyville waiting for people to walk forward (riders must be in the first 6 cars to exit at those stations due to platform construction).

(Photo: The LIRR Today)

The M-9 cars show the “Car __ of __” information on both the interior and exterior side displays.  I’ve gone back and forth on this since I first saw this in testing, and I tend to think showing it on the exterior side displays leans towards more harm than good.  Dwell time management is critical at all stations (and it’s something LIRR’s not particularly good at).  When a train pulls into a station, the top priority needs to be getting everyone who needs to get on boarded quickly, the doors closed, and to start moving out.

During these critical times, it's important passengers are only given relevant information, and I struggle to see what value showing this information on the outside of the train adds.  What decision does it inform?  If you know you need to go to St. Albans and that you must be in the fist 6 cars, but the train comes to a stop and you're in front of car 10...what are you going to do?  Walk along the platform and delay the train until you board the right car?  That's not something to encourage...instead, people should just get into the train and then sort themselves into the right car once the train is on the move again.  Trying to do it at this point seems liable to further protract dwell times (and since this information preempts the destination—which is the piece of information that's actually useful in these cases—there's a higher chance people board the wrong train because they couldn't easily verify the destination).  At terminals (like when the train is sitting at NY-Penn Station boarding) is a different use case, so doing this might be fine there, but not at intermediate stops where the goal is to keep moving.

Splotch watch

(Photo: The LIRR Today)

Given how window splotching is starting to become a significant problem on the M-7 fleet, the opacity and cleanliness of the new M-9 train windows will be an important thing to keep an eye on.

Even on the first day of service, there are already some signs of window occlusion, with the effects more pronounced on the south side of the cars (the same side that the M-7 splotching started to become very noticeable).  This could just be ordinary dirt, but the window quality will be something worth keeping a close eye on with these initial cars.

No quantum leap

In addition to some of the things already mentioned, there are a couple of other changes and improvements made in the M-9 design over prior cars.  The LIRR says the new cars are providing "a full slate of amenities", including:

• Closed loop seat armrests
• Electrical outlets on both sides of car in each row of seats
• 4-6 Additional seats per married pair (217 seats/pair) (actually only 4 additional seats per pair, and even that's a stretch).
• Suspension system flip seats (prevents slamming)
• Slightly increased width of seats
• Increased window tint (reduces sun glare)
• Energy efficient LED lighting
• Consistent car number display (“Car 2 of 6”)
• Four 32” multimedia screens in each car (similar to, but different from the video ad screens being retrofitted on the M-7 fleet as part of the Live Media Concession).
• Text-to-Speech service updates (this is not a new feature, the M-7's have always had this, the LIRR just never uses it)
• Front of car exterior destination signs (visible as train approaches station)
• Electrically-powered B-End door: features a press pad customers and crew can press to activate the opening of the door – when the pad is pressed, the door and the door in the adjoining car automatically open. (this feature has been standard on most rolling stock, especially those with end vestibules, for decades...MNR's Shoreliner cars have had this feature since the 1990's)
• Quieter side doors which operate with less impact force and are easier to open manually in an emergency
• Improved toilet room doors
• Additional coat hooks in toilet rooms
• Additional quality treatment for toilet room air
• Improved design for cleanliness and touchless experience in toilet rooms
• High Velocity “no touch” toilet room hand drier
• Scratch-proof mirror in toilet room

...and that's about it.  All around, a rather minor list of improvements over the M-7 fleet, which was designed 20 years ago.  The new cars, which cost $3.63 million per car, are 23% more expensive than the per car unit price for the M-7 fleet (adjusted for inflation effects)...the new M-9 cars are definitely not 23% nicer than the M-7's for the commensurate 23% increase in cost...

Overall, the LIRR's M-9 cars represent no quantum leap in car design or passenger amenities.  These cars are basically a slightly newer version of 20-year old M-7 cars with a new color scheme and a few more minor features.  The MTA seems to put in some effort into completely rethinking the car designs only every 40 years or so...  The M-3's were basically a copy-and-paste job of the M-1 fleet, with a couple minor modifications, and now the M-9 fleet is not much more than a copy-and-paste job of the M-7's, with a couple of minor modifications.  Time will tell if other upgrades 'under the hood' help the M-9's prove to be more reliable than the M-7 fleet, or stand the test of wear and tear better than the M-7's.

The M-9's would have been a top-line railcar design 15 or 20 years ago, but today they pale in comparison to the design of any modern MU practically anywhere else in the world outside of North America.

There is only so much that can be done with the regulatory framework and design criteria the MTA set, granted...the LIRR specified these cars to outdated/archaic FRA crashworthiness regulations—not the new amendments that entered into force this year, bringing American crash requirements more or less into line with Euronorms—and the design criteria calls for seating as many passengers as possible.  There are only so many ways to optimize a 85 ft long metal box that has to seat 210+ passengers...but this is where the MTA's out of whack fleet planning priorities rear their head.  There was no reason to spend all of this money buying hundreds of cars built to archaic regulations and outdated design criteria.  Had LIRR better maintained or refurbished their M-3 fleet (like Metro-North did), they could have bought some time until the new passenger equipment safety standards took effect, and then purchased an off-the-shelf modern EMU design from Europe...(and still be able to have everything here in time for East Side Access opening).  (All the while, the LIRR's diesel fleet continues to languish).

Designing for maximum seating is probably not the wisest move in this day in age for regional railroads.  Today's modern regional rail systems do a double duty of sorts, transporting people from the suburbs, and supporting intra-city ridership through integrated fares and schedules.  Raw numbers of seats are no longer the top priority, but rather finding an appropriate balance between seating and standing space to optimize overall capacity and rider comfort.  Designs also need to seek ways to minimize loading and unloading delays, which can be significant on crowded LIRR trains today.

Doing something like ditching the middle seats (which have never been popular) might seem unconscionable because it means fewer seats, but it might be better overall...  Instead of wasting all that space on an awkward seat that rarely gets used, cutting it and expanding aisles provides more room for standees for close-in stations, and makes passenger flow throughout the car significantly easier (no more tripping over a dozen people when trying to elbow from the middle of the train to the door at your stop).  Making doorways wider (or even adding doors) to increase the rate at which people can board and alight can improve dwell times, and allow the railroad to put more trains on the tracks.

These are all things becoming increasingly common on modern regional rail systems across the world over the last several decades, but are absent from our brand new cars.  A 23% cost increase for a design that's effectively 30 years out of date the moment the cars hit the rails is a lot of money for little gain.

Newer car designs go a step further and are fully articulated...passengers
can pass freely between cars, and thin glass doors serve as a sound barrier
(Photo: The LIRR Today)

The new M-9 cars are even nearly a generation behind other cars being built or in service right here in the United States (and being a generation behind the US rail industry is really saying something...).  The automated B-end door on the M-9's may seem like the cat's meow, but Metro-North's Shoreliners have had these on their end doors since the 1990's, and Amtrak's Amfleet/Viewliner/Superliner cars had these years, if not decades, before that.  The trend nowadays is to have trainsets that are completely articulated, with the entire train either open form end to end, or only thin glass doors between compartments to serve as a sound barrier.  The trains being used on Brightline/Virgin Trains USA in Florida feature the latter design, for example.  The new M-9 cars feature outlets, which is great, but they are off to the side along the walls and are only one per row (so if you're not sitting at the window seat, you have to reach over someone to plug in your stuff).  Common practice today is to locate the outlets underneath the seats, between two seats, so you don't have to reach over or dangle your cable across someone else.

A 10++ year old design is new?? The LIRR is a joke. Like they just upgraded us from a pager to a flip phone. Always 10 or more years behind the times and always will be at this rate. At least they are consistent with disappointing us and not much else.

— Paul D Speh (@pdspeh69) September 10, 2019

Someone on Twitter likened the limited extent of "upgrades" on the new M-9 fleet to upgrading from a pager to a flip phone...and I think the change is probably even less incremental than that...  It's more like upgrading from a brick phone to a flip phone.  A couple new features, a slightly more refined design, but overall little additional functionality.

The new cars are nice, but overall lacking in design and amenities for the extremely steep $3.63 million pricetag (nearly twice what off-the-shelf European EMU's typically sell for, about $2 million (once adjusted for train length and currency effects)).  While we are, regretfully, locked into this design now (though the MTA should seek reductions in the number of cars as part of any delay settlement/liquidated damages claim), the railroads should be trying extra hard to not repeat the same mistakes with the upcoming M-9A procurement...the contract for which may be awarded as soon as this month.  With the new FRA regulations now in effect, there's simply no reason for the MTA to turn their nose up at international best practices any longer.