For want of a 1

From the Red Line
Published in
11 min readMar 18


Let’s avoid crossed lines.

The impending launch of the R151 trains has gotten train enthusiasts talking, and not necessarily in a good way. This will likely be one of the most controversial posts on the blog. Still, I’ve always spoken up for what I think is right, otherwise why write?

When delivered, much to the shock and horror of the enthusiast community, it was revealed that the LTA added a “1" to the front of the car numbers, in clear defiance of what is allegedly tradition across the NSEWL network. Then SGTrains Spotters followed suit, despite polls indicating they should not.

I believe there are good reasons for this, in what appears to be a minority view. While I usually think these matters beneath the blog, I feel that I must explain things here and not in the short-form of instant messaging. He who does not learn from history is doomed to repeat it, after all, and it’s in the best interest of the community not to have to go through another such transition and then have to deal with those left behind. We’ve already gone through two, as we will see.

source LTA

Customer facing

Much will not change for many of the laymen and women reading this blog. If you ever need to interact with the SMRT SnapRep service, lost and found, or worse the police, often what is of interest is the car number. This is usually prominently displayed at the ends of train cars, and in some cases, on the emergency communication buttons by the sides of the train doors.

It’s quite possible to guarantee that this will be unique across the entire MRT network, much like the license plate of a vehicle needs to be unique across all of Singapore. It will not change with the delivery of the R151 trains, nor will it change with any future train order. Car numbers are need to know for everyone riding the train.

This wasn’t always the case. The NEL engineers decided it would be wise to prepend a “7” to the start of the NEL train car numbers. Otherwise, they used the same numbering scheme as the NSEWL, which meant that had the “7” been ignored, confusion would arise — as it probably did for the BPLRT, which used 101–119 in a similar range as NSEWL C151 trainsets.

PM Goh visiting the NEL in 2003 (source: NAS)

But yet, this “7” was ignored.

Well, this is a Youtube video title with the need to be concise, and it’s clear that the NSEWL goes nowhere near Sengkang. But it’s not hard to imagine that in those days of SMS, reporting only “019/020” in a shorthand SMS might have caused issues especially had the locations been reported as Dhoby Ghaut or Outram Park, interchange stations with the existing NSEWL.

Eventually, the community settled around adding the “7”, but as we see, willischong Video Productions was slow to get the message.

The dust eventually settled

willischong might also have been dissuaded from the third naming convention in the above screenshot, by the announcement and arrival of the C151C trains, which would take up the 700 series of train numbering. 709/710 today is a totally different train on a totally different line, and if you were to say something like that today in a chatgroup or other instant-message forum without being clear what you were referring to, you’d be setting yourself up for trouble.

For the sake of brevity, and since as enthusiasts we are generally expected to know what belongs where, some do omit very valuable context that requires the reader to reverse engineer. But I would argue that this context cannot be removed, especially in these days of mass communication which means information can spread far and wide with little opportunity for clarification or correction; just like my apparent role in all this.

You have that right, but you might just be making more trouble for yourself by leaving things out to begin with. Just as an extra 0 tells apart 709/710 and 7009/7010, what’s an extra 1, or an F?

“It’s not my problem”

Sure, confusion can be dispelled. But does it make sense to confuse others and have to spend your own time clearing doubts, when clear communication can be done in the first place? In any case, you can’t ask a website or a social media post. Many train enthusiasts may aspire to be rail engineers working on the railways. To that I say, communication is key. It was a lack of proper communication which caused the unfortunate death of two trainees at Pasir Ris.

The key here is the Circle Line, where new trains are also likely to be entering service within the same time frame of the R151 trains. I don’t expect there to be a similar level of interest — in fact I don’t expect there to be much interest at all — in the new Circle Line trains when they do get tested on the line itself and also eventually go into service, compared to the NSEWL.

The argument I’m making here is that the Circle Line has already taken up the 800 series of train set numbering. This is the likely reason why the LTA chose to add the 1 for the R151 trains, such that it would not overlap with the CCL trains’ namespace, instead occupying its own in the 1800 and 1900 series. While it can be assumed that the last trainset numbers may probably continue on perhaps to 1999/2000 and onwards, SGTrains Spotters has at time of writing refused to list as such, presumably because using 2001/2002 and onwards would also present conflicts with the TEL trains, which occupy that namespace.

Or they could go back to 001, but without the 1 there may be confusion with our legacy trains. Or even if they find some other range, but that may be kicking the can down the road especially when the Kawasaki-Sifang C151As need to be replaced.

In fact, this issue might even exist now. Does “2006" refer to car 2006 on the NSEWL, or train 2006 on the TEL? Does “113/114” refer to the NSEWL C151, or a pair of BPLRT APM 100 cars?

So let’s get it right the first time as an enthusiast community. Much like how the “7” was used to differentiate NEL trains from NSEWL ones, the “1” can be used to tell apart NSEWL trains from CCL trains. Whilst for now things like run numbers can still be used to infer context, that won’t always be the case, especially as the CCL receives much-needed service upgrades that result in significantly more trains deployed. This may be more than enough for run numbers to overlap significantly with that of the NSEWL.

We have split off from such traditions before. The transition to CBTC signalling meant that the old concept of “Coded Manual (CM Mode)” is now irrelevant. Under CBTC, the new name for manual operation of trains under signalling system control is now called “ATPM Mode”. And the community has made this transition before old terms get too entrenched.

Whilst previous lines took reference from the contract number series (NEL 7, CCL 8, DTL 9, TEL 2), the JRL adopts the “J” series of contract numbers and numerically starts afresh from 1, leaving its numbers up in the air. As seen here by SGTrains as well as the construction blogs, and at the JRL train mockup, JRL vehicles will use “3” as their identifier prefix.

source SGTrains

The rest of us

Of course, some of the means to dispel confusion is easily controlled by the LTA, and this also matters to other people as well. I’m talking about platform and exit numbering — which we didn’t really care much about at first but which has also received new attention as part of the TEL signage revamp.

With DTL3 and TEL, a notable change is that platforms are now numbered sequentially across a whole station complex (Tampines being two stations). Exits were already numbered sequentially to begin with, starting from when we first had to build complex stations at Outram Park and Dhoby Ghaut during the NEL construction, but we never really thought much about platform numbering at least until very recently. Dhoby Ghaut, in particular, has three pairs of Platforms A and B — one each for the NSL, NEL, and the CCL.

With TEL3 we now have more three-line interchanges at Outram Park and Marina Bay, and I must assume that the complexity of getting around these station complexes means that one must be able to uniquely identify every exact point in the station complex. Likewise at Stevens, where all four platforms can be accessed from the TEL lifts. It does no favours to have overlapping platforms. In any case, it’s never too late to renumber, much like what was done at Chinatown, Outram Park, and Marina Bay. If we have to re-record announcements and change other systems components, then we’ll have to pay it.

It wasn’t so difficult before. In the days before the NEL, there was colour-based platform directions, presumably to help the illiterate navigate the system. Presumably Singapore was then the only metro system in the world to assign different colours to different directions. If you were at Toa Payoh and you wanted to go to Bishan, you’d take the yellow line. Going to Orchard? Take the red line.

However, with the NEL, it was quickly deemed unviable as the colours required would quickly and exponentially multiply as the network expanded. The decision was made, likely at the suggestion of consultants Lloyd Northover, to use a system of numbering end destinations instead, a system that still persists today. At Toa Payoh, take train number 4 to Bishan, and train number 5 would get you to Orchard. This served as a fair replacement of colour lines, but may also have pulled double duty as platform numbers.

In block letters

Until 2011.

Between 2011 and 2020, while platform letters continued to be deemphasised elsewhere, they played center stage at Jurong East. You now have to know whether the next NSL train departs from Platform A or Platforms D/E, after all. And as part of this, Jurong East station received a unique signage system and next-train displays with platform letters much more clearly denoted than at other stations. This may be changed when the JRL comes, but I still expect the style to mostly remain, considering what we’ve been seeing on the TEL with big letter blocks, much like those at Jurong East, denoting platforms.

Yet, a historical anachronism that the train enthusiast community continues to cling on to, is the old platform assignment at Jurong East. This is similar to Ang Mo Kio and CCL Paya Lebar in that whilst platforms A and B still remain as the main line tracks, the middle platform was denoted as C/D. No, we have to go even further back, to the MRTC era when numbers were used to operationally denote platforms.

The completion of the JEMP meant that a fourth track was now needed at Jurong East station. 4 was taken since it belongs to the middle platform, so they called it Platform 5, and the new eastbound platform, Platform 6. Using this for train enthusiast lingo still works now, but what about at Tanah Merah, where eastbound services will be relocated to the new platform come 2025?

(source: LTA)

Interestingly, as seen in this diagram, the LTA uses “CAL”, not “CGL”, to refer to the Changi Airport Line. If we’re all going to slavishly adhere to internal terms, maybe not only should SGTrains update their bot, the LTA ought to update the system map to say the same thing then.

Of course, things change across generations, and a newer generation of train enthusiasts may choose to use more public-facing terms such as those used on signs.

Change is coming

This sort of structural change has also taken place in the bus spotting community before. For around 40 years SBS might have been able to solely rely on 4-digit license plate numbers to identify each of their buses. After all, the SBS prefix was reserved for SBS buses. Trans-Island also got their own TIB-series prefixes.

Enter the Bus Contracting Model, which not only introduced SG-prefix license plates for buses, but now also meant that buses nominally purchased by SBST could end up with SMRT, and vice versa. Suddenly, the prefix is now important internally as well. Here’s an example, a particularly more egregrious example since not only do both buses belong to the same operator, they’re also run out of the same depot.

source as of 18 March 2023

Maintaining records with only numbers would be impossible since one would have to use the prefix or suffix to tell them apart. And bus enthusiasts, for much part, have had to cope with this change. Many I’ve seen now use the checksum suffix to identify buses in addition to the 4-digit plate number. Perhaps that’s what they’ve always done, or it was a necessary adaptation to deal with the sea change of the BCM.

The same should apply for trains. While it may be acceptable for systems at an individual line level to drop the prefix and refer to trains solely by a line-specific vehicle number, it is quickly clear why this is not acceptable at a network-wide perspective.

LTA Report on 14 Oct 2020 power failure
REAMS exhibit at SITCE 2019 (picture credit to @380)

Sure, the Circle Line uses “PV” to refer to trains. But so does the separately managed Thomson-East Coast Line, or as SGTrains points out, the SBST-run lines as well. Using these terms make sense to employees at a line level, but not to train enthusiasts at a network level, as without context chaos can quickly arise. It might have been acceptable for enthusiasts to use just the trainset number at that time, but not any more. It’s safe to say that it’s from this period that the current CCL, DTL (and TEL) numbering schemes have been used by enthusiasts.

Change is coming.

The current generation may overwhelmingly choose to omit the “1” from R151 trains just because SMRT NSEWL’s systems do it, but I believe this to be extremely shortsighted and not mindful of the history of the enthusiast community in general. The MRT is a network, and far bigger than just the NSEWL. Likewise, just because SMRT NSEWL does it that way doesn’t mean it’s a good idea from a network wide point of view.

Train enthusiasts, you are here. (source: LTA)



From the Red Line

Sometimes I am who I am, but sometimes I am not who I am not.