Upgrade, or upgrade?
Much ado about something, for once.
The Sengkang-Punggol LRT is getting a significant upgrade, but are they going far enough? If you ask different people, they’ll tell you different things, but that’s the way it is. At the very least, though, I don’t think you’re going to meet anyone who’s going to tell you that this isn’t the right step forward.
But they’ll probably disagree on what should come next, especially since the authorities like to keep their cards close to their chest and not announce anything except the immediate benefits.
Shake up the market
Compared to the Alstom-Bombardier solution in use at Bukit Panjang, the SPLRT is built off the Japanese AGT standards, which means a relatively freer hand in how the system can be built and expanded, up to and potentially including bringing in additional suppliers to give Mitsubishi a run for their money; which may explain the three-year lead time or worse, if design time is also needed. The Hong Kong Airport people mover, for example, operates vehicles built by both Mitsubishi (flat-sided) and Ishikawajima-Harima (round-sided), and apparently a Thales Seltrac signalling system.
And the authorities here intend to take advantage of this, by purchasing 17 2-car trains; not a lot at first glance nor does there appear to be a lot of space to use them, but I think this has potential. In AGT world, 34 cars is actually a considerable amount — it’s double the fleets of some airport people movers, and given that they’re mostly a standard product, the modifications needed to meet LTA requirements could be done with a lesser cost premium given a larger fleet.
Secondly, by 2026 some of the first-generation Crystal Movers could be in bad shape, especially those that did not undergo modifications for 2-car operations. A further add-on order for replacing the 41 first-generation vehicles in the style of 40 more R151 trains would help to keep costs down, especially if we choose not to go with Mitsubishi and another manufacturer has to develop a design for us from scratch.
This may be a sufficiently large order, especially if the intention in the nearer term is to go to full 2-car operations throughout the LRT network. If none of the 17 new 2-car trains are for replacement, and we subsequently buy 41 2-car trains to replace the 41 first-generation vehicles, that’s a total order size of 58 2-car trains, nearly as much as the original Contract 830 CCL fleet. Not to mention a potential three-car service that could be made possible with the new depots, and by making better use of existing space in stations.
Space is not a problem
The most immediate thing one notices is that the LRT depot will practically triple in size, from 3.5 hectares to 11 hectares. Land is not a problem; there is a lot of land available at the MRT depot deck next door, which given our current mindsets is not likely to be used. So why not?
Firstly, if we want to buy that many trains as described above, we’ll need a lot more space. The current LRT depot may be just right for the existing 57-strong fleet, but if we’re talking about more than doubling it ((17*2) + 41 = 75), we’re going to need a lot more space, which might justify the tripling in size of the depot — especially with our habit of building far more than we might need.
This new depot will also include new maintenance facilities — and all facilities could be built to accommodate longer trains and replace the current maintenance facility, but that might be a distant pipe dream, more on it later. More directly, the Crystal Mover system in Miami operates with very similar vehicles to ours, but with two vehicles joined directly much like MRT trains. If this is what the authorities mean by two car trains, there will need to be changes to the workshop anyway to fit these trains, which have some differences from a pair of single-car vehicles. However, without redundant escape doors and equipment lockers, they may be able to carry more people than a pair of single cars.
There may also be a need to modify the platform edge barriers installed a few years ago to accommodate the slightly different door spacing of the intermediate vehicles. Right now, some people say that the platform edge barriers don’t go far enough in providing railway safety, so perhaps such a modification could come with the installation of platform doors, again like in Macau.
Distant possibilities
Unlike the underground stations, where there could be an argument to be entertained that the implementation of longer trains is going to be very difficult (at one’s most optimistic), on the SPLRT you have this:
That’s a lot of space and not a lot of things done with it, which could be enough perhaps for four cars and change (or three with additional equipment rooms). With a new depot, new workshop facility, and a rework of the current one, I don’t think there would be a bad time to also throw in a modification of the signalling design to allow for network-wide three-car train operation.
I think any of the Japanese makers would be willing to sell us three-car vehicles, be it Mitsubishi, IHI, or someone else. Nor would it be a stretch to think that there aren’t the clientele to justify the needed engineering effort to add an additional car to a network ostensibly not designed for it. After all, the alternatives are even more unthinkable, and perhaps far sillier.
Three-car vehicles might allow a continuation of the current operating model where trains can be built up from single cars, providing some flexibility. But will that be needed come 2028 when Punggol gets far more developed and three-car vehicles are run most of the time (if not always)? I’m really not that sure, and there could be some more benefit from switching out to mostly fixed three-car sets (hello Orlando), such as the additional space from the removal of redundant end emergency doors and equipment cubicles, even if the vehicles can’t be made walk-through as seen in Macau.
Capacity wise, it’s going to be similar to some of the AGT systems back in Japan too — the Yurikamome 7300 series, of 6 smaller cars (similar total length, but narrower), is also rated for approximately 300 passengers per train. While overall capacity might be limited with the need to share tracks through the Town Centre stations, three-car LRT vehicles might provide speed and capacity far better than a bus, reducing the need to spend more on parallel services as well.
Raising the dead
That said, sharing tracks through the Town Centre is not a wholly lost cause. LRT vehicles already have much wider doors than their MRT counterparts, something that should probably be retained in new orders, which allows dwell times to be kept low in order to keep frequency up. So long as the signalling allows for it, 90 second headways or even lower, while difficult on the MRT, may not be impossible on the LRT. That gives you three minutes on each loop, the promise of which may be enough to overcome the access penalty up to the road-median LRT platforms.
At Sengkang, apart from the current upgrade to add more lifts and escalators to improve circulation, it can also be considered to reconfigure the LRT platform and mezzanine level in order to provide a Spanish solution like at Choa Chu Kang. While a squeeze, additional LRT platforms could be placed on the opposite side of the track from the current ones for alighting passengers, such that the current platform is only for boarding, or to separate them by service such that West Loop passengers use the new platforms, and East Loop the current ones, or something.
At Punggol this is where things get interesting. Remember the abandoned terminal platform for the cancelled North line?
This is where we may have something to learn from New York’s three track lines. What could be done is to connect the middle track to the West loop in both directions — building out the two crossover provisions just before Sam Kee going counterclockwise, and constructing an additional track to link the middle track to a pocket track near the south side of the loop, as in the below picture.
This would allow one direction of the West loop to be sent through the middle track in the peak direction during peak hours, allowing independent operations of that loop and its corresponding service on the East loop. In such a scenario, it would then become possible to allow both loops to operate at higher frequencies (around 2 minutes or even less?) and provide a higher level of service with more reliability than the delicate operation in Sengkang.
It’s not a lot of concrete and I think sounds less ridiculous than some other things I’ve seen, so it may be worth a try. Punggol may well need it especially with the upcoming PDD, as well as BTO developments around the town. Since there’s almost zero chance the North extension will be built, with the stub tracks at Teck Lee leading right smack into the middle of the future SIT campus and the NELe likely to play the role it was meant for (NE19, anyone?), we can reuse these built infrastructure to improve service. That additional guideway probably isn’t going to be too hard?
The main problem will of course be passenger information with such service patterns and platforms semi-randomly swapping over depending on time of day, but with good signage that might not be an issue. It also wasn’t that long ago when the LRT ran on a unidirectional loop anyway, switching directions at three pm in the afternoon.
Push the envelope
By chance or by design, there’s more than enough opportunity that can be taken to improve the capabilities of the SPLRT. Let’s see how much they’re willing to experiment, but realistically there’s not a lot that can be done in terms of organization. However, plenty of opportunity with infrastructure investment.
But first, to get the basics right is more important. Maybe, though, there could actually be proper countdown clocks telling people the makeup of the next train and when it comes next. If there’s a technology upgrade along the line, this is probably among the top, if not a full-scale resignalling to a moving block system, if one wants the high frequencies in the Town Centre areas.
The lessons learnt from such upgrades can probably serve as a prelude to the true capacity expansion battles to be fought on the MRT network in future, especially on the medium capacity lines. Eventually, I believe it’ll mostly be similar, with increasing pressure on the existing network and a lack of alternative routes for the construction of relief lines. In this aspect, pushing the envelope just for the LRT already acquires enough meaning on its own, if you ask me.
That said, there is only so much that makes sense to do, especially given the LRT’s status as a feeder to the NEL. The NEL, on its part, will need to be able to cope with the additional ridership from such capacity upgrades; that is probably one of the future battles to be fought.
