On Wednesday, the council’s Finance Committee is being asked to give urgent funding approval to purchase 17 new battery powered trains, or Independently Powered Electric Multiple Units (IPEMUs) as they’re referred to officially. The council paper and the full business case notes that the urgency is because a deposit needs to be made by September this year to ensure delivery of these new trains in 2019 which is when the capacity constraints really start to bite. Given how plainly obvious this issue has been for some considerable time, it’s a bit absurd that it now needs to be made so quickly.

Auckland’s 57 new electric trains only finished rolling out two years ago but they’ve been such a success that we’ve been saying for almost as long that we’re going to need more of them. That’s in part because the use of trains has been growing much faster than expected. For example, it was initially forecast that in 2017 we’d have about 17.8 million trips, rising to 20.8 million in 2023 when the CRL opens. We actually had 19.6 million trips this year with growth remaining high at almost 17% annually. As such, even AT’s newer projections of 22.8 million trips in mid-2020 seems extremely light. While there’s still a lot that can be done to improve off-peak services, the main thing causing those low projections is a lack of capacity at peak times.

We already know that until the City Rail Link (CRL) is completed in 2023 we can’t run any more peak services. AT say that options to improve capacity with our current fleet are “close to being exhausted”. The only option really being to reconfigure seating in our existing trains to cater for more people standing. Other operational changes, such as speeding trains up, will improve services but won’t be enough to free up additional trains and therefore capacity. That means that the only other way to improve capacity before CRL is to buy more trains so that all services can run as six-car trains, except those on the Onehunga Line. We need about 15 more trains to be able to do that.

At the same time, we also need to deal with the issue of trains to Pukekohe. Diesel shuttles currently run between Papakura and Pukekohe forcing a transfer for almost all passengers. This is far from ideal from both a passenger and operational perspective. The diesel trains are old and costly to run, they are also due for a significant, and potentially expensive overhaul. In addition, the area between Papakura and Pukekohe is expected to see substantial growth in the coming decades with much of it based roughly around the rail line. Electrifying the line to Pukekohe, along with new stations to support the growth has long been on the cards to help ensure new and existing residents have options to avoid having to sit on the southern motorway to get anywhere. ATAP considers electrification to Pukekohe as a 1st decade project and expects this to cost around of $160 million.

Despite the local MP saying the government had committed to funding the project, it hasn’t appeared to have been any progress on this essential project and it was not expected that it would happen until around 2025.

The IPEMUs are meant to solve these two issues with a single solution. The idea is pretty straight forward, the new IPEMUs will be the same trains as what we have now, but they have batteries added to the roof of the central trailer car to allow the trains to run beyond the wires. Those batteries recharge again when they hook back up to the electrified network at Papakura.

In total, combined with the trains needed to increase capacity on existing services, we’ll need 17 new trains. These would all be used on the Southern Line, allowing for some of the existing trains to be used to lengthen services on other routes.

The costs of these new trains are below and it’s worth noting that Option A only deals with the capacity constraints and doesn’t involve new trains to service

To address the problems and opportunities, AT shortlisted two main options utilising the Better Business Case methodology, one of which has two variants. The options are:

  • Option A: purchase 15 three-car EMUs at $133 million
  • Option B: purchase 17 three-car IPEMUs
    • with CAF (supplier of the current EMUs) batteries at $207 million, or
    • with Korean (LG) batteries at $174 million.

Those CAF supplied batteries aren’t cheap, adding almost 40% to the cost of a train. But they note that savings from running diesel trains ($67m) and from addressing the capacity constraints, which would otherwise require more bus services ($84m) are substantial. The main benefit of going with CAF seems to be to deal with warranty issues while the Korean batteries are not only cheaper, but expected to perform better too.

The business case proposes a direct procurement from CAF – Construcciones Auxiliar de Ferrocarriles – (the provider of the current EMU fleet) conditional on CAF maintaining the prices for the base EMUs, with acceptable and transparent cost additions for the battery technology in the IPEMUs. CAF have offered a fully warranted battery solution that in AT’s opinion is low risk but high cost. AT has sourced alternative battery systems that promise simplified installation, improved range at reduced capital cost. The alternatives are still work in progress until compliance to industry standards has been demonstrated, therefore the CAF IPEMU solution has been treated as the base at $207m and the alternative solution as a sensitivity test at $173m. AT will work with CAF and the alternate battery supplier to determine whether the alternate solution can be successfully engineered and deliver the value anticipated.

Battery powered trains certainly seem like a useful addition to help extend services beyond wires. The biggest issue though, is that they only expected to last 7-8 years, after which time they’ll need to be replaced, or we’ll need extend the wires to Pukekohe. We could of course do both and allow for services to be extended further south, perhaps to Pokeno. It is expected that by the time they need to be replaced, advances in battery tech will give the IPEMUs greater range.

The business case suggests that the Benefit Cost Ratio is 2.8 with a range 2.0 – 4.1. Interestingly the 15 normal EMUs have a higher BCR at 3.2, but it doesn’t address the issue of services to Pukekohe and therefore the cost of diesel shuttles.

The report notes that the first vehicle will be available for service within 27 months, with all in service 36 months from order placement. Initially they would run as normal EMUs to boost capacity until the batteries and systems can be fully tested. One in place, the business case assumes we could reach 30 million trips annually before the CRL opens.

The council agreeing to fund these new trains will push it right up to its debt ceiling, going over that would see its credit rating downgraded and that would lead to more expensive debt costs for not just Auckland but all local council’s in NZ. As such, the funding from the council is conditional on a few things, including:

  • The NZTA committing to fund at least 50%
  • Auckland Transport committing at least $50m from their existing capital expenditure budgets

Overall this seems like a decent idea but I do have a few thoughts about it that I haven’t seen covered.

  1. The council paper notes that there will be periods when the power will be off on parts of the Western Line during CRL construction and that these could allow some services to still run. Presumably that would only work for weekend shutdowns as there wouldn’t be enough battery trains to service both Pukekohe and the Western Line if the power was shut off on a normal weekday.
  2. It’s not clear what would happen to these trains after the CRL is open. Presumably they’ll be designed to run safely through the CRL but current plans see the southern and eastern lines linked together and so there wouldn’t be enough of them. Although it’s worth noting that this purchase of trains is just to get us through to the CRL and we’ll need a lot more after that project is complete. In fact, we’ll probably be needing to start thinking about purchasing more at about the same time as the last of these IPEMUs goes into service.
  3. The business case assumes that Pukekohe will still be electrified in ~2025. This effectively lets the government off the hook for that while imposing higher costs on the council so I’m a bit concerned that this will mean the current plans to electrify to Pukekohe will be subsequently delayed as a result.
  4. As a counter to point 3, even if approved tomorrow, it’s excepted to take at least 4-5 years to extend the wires to Pukekohe (I’m not sure why it will take so long). This decision and additional cost to the council at least allows us to get some of these issues addressed a few years earlier.
Share this


    1. I wonder what’s the range of those units? If they can reach 100km perhaps we could reuse them for Pukekohe-Hamilton? Obviously they would have to recharge before return trip.

    2. I heard they are using the recalled Samsung Note 7 batteries. They have millions of them sitting around.

    3. My back of the envelope calcs: time taken for a trip Pukekohe to the CRL and back out to Manukau – about 1h 48m – say a 2h turnaround, or four hours for a round trip. Four trains therefore required for an hourly service, or eight for a half hourly service. Assume that they’re six-car trains, 16 B-EMUs therefore required – hence, in my view, the order for 17. The remainder of the trips on the Southern and Eastern lines to be operated with standard EMUs and the “saved” units used to upgrade existing services from 3-car to 6-car. Seems eminently logical therefore that 17 is the number decided on.

      1. But do both units need to be driven for the run to Pukekohe? It’s reasonably flat and IIFC the original spec called for the ability for one unit to be able to haul another out of the not flat CRL.

        1. It’s not flat – there are three reasonable banks to climb to Pukekohe. It doesn’t help to have one unit towing the other dead. It’s still this many tonnes at this many km/h up this gradient. Doing the same job with half as many wheels and half as many batteries just wears them out quicker and pushes them to their limit sooner. It also means that if the one battery unit has a problem, they’re both stuck.

  1. What would speed things up also is if we can find a way to deal with the issue in the long run of lower platforms and increase them to the height where the train is flat the whole way along with metro style seating

    1. But the platforms are at the same height? Only Westfield (now defunct) wasn’t standard within the network.

      1. I think RJ Yap means because of the current standard platform height only the middle carriage can be low floor due to the equipment under the train.

        If the standard platform was higher then all carriages could be level access low floor with no stairs.

        1. not possible without moving the tracks further from the platforms to maintain freight wagon clearances.

    2. I don’t think the step up into four of the carriages really slows things down at all. It’s a much smaller step than Wellington and they manage to run much quicker dwell times than us.

      1. Door well steps to high floor carriages not that different between Auckland & Wellington units. FP/Matangi door well 20mm lower than AM , two steps in either case. You might be thinking of the older units (Ganz & EE) that had one large step.

  2. It is good news that the government has announced funding for two new stations in the franklin area. This will surely help making the business case for puke electrocution (and reduce the costs). I’m not sure that electrication to puke is our highest priority as it will take several years for significant resedential growth to occur there. Other projects like airport LR, airport botany link, third rail and NW busway seem like higher priorities at this stage.

    I wonder if NZTA will come to the party and pay for half the costs.

  3. Given the government’s convoluted and as of yet to be explained or analysed suggestion to somehow fund infrastructure for green field development thus expanding Aucklands residential footprint to places like Ramarama, electrifying the line between Pokeno and Papakura, minimum, should be a no brainer.

    Battery powered trains are a very, very expensive short term band aid solution rather than a far longer term cheaper solution of electrification. Any half thinking planners could see that, surely?

    Maybe this mythical $600 million bribe that Joyce is throwing around can fund it!

    1. Let’s look at this claim of very very expensive for a second, not forgetting the money/time trade off. Based on the figures reported above an EMU costs $8.86m. An EMU with batteries costs $12.18m (most expensive of the two battery options). So the battery system costs $3.34m per train. WIth 17 trains that’s $56.78m for battery power. They report a seven year life, so all up the battery option costs $8.11m per year.

      ATAP reports electrifying Pukekohe to Papakura has a cost of $160m. Using a 6% discount rate to annualised this means the electrification option costs $9.6m per year in the cost of capital. In other words, every year you don’t borrow $160m to build electrification you save $9.6m in borrowing costs.

      Ok this is pretty ropey economics, but it does suggest that the two options cost, broadly, about the same per year. So I don’t think it’s fair to say the battery system is very very expensive, its about the same cost as electrification.

      The interesting thing with them costing about the same per year is, there’s nothing sunk or lost by doing the batteries first. So its perfectly sound to buy the trains, fit the batteries, run them for the seven year life, and in the meantime get on with the electrification programme to keep running the trains after the batteries have run out and have been removed.

      1. Ok the cost might be similar, but after 6 years you get to keep the wires and poles while you have to throw away the batteries to big cost to you and the environment (even if recycled of course)

        1. I didn’t have time to cover it fully but there’s some detailed information about maintenance costs, such as a $3m pa cost to maintain/run the wires. I’ll look at those in more detail on the future

        2. There will be a saving on oil imports.
          NZ generated power is much cheaper and helps us reduce our carbon emissions.

        3. Ian that is accounted for in the estimate, there is no extra cost of throwing away the batteries as they are fully depreciated across the seven years. If you go with the wires, you are still paying for the cost of installing the wires after seven years too, not to mention maintenance.

        4. The maintenance is the scheme of things is bugger all. They are only considering battery power because of the mindless ideological intransigence of central government.

          Honestly they would better buying a fleet of new quality diesel electric locos, attaching them to the SA’s, that they own, and once the line is electrified sell the loco’s to Kiwirail!

        5. I was going to offer a sweepstake on how long before the luddite cranks dismiss this idea and call for the return of diesel hauled SA trains, but…

        6. Yep just a Luddite who sees spending the better part of $100,000,000 on batteries with a life of 8 years idiotic. You struggle with that dead money option, obviously.

          Ask yourself, where would that money be better spent in PT?

          We should electrify the section to Pukekohe bare minimum and now but there isn’t the money from Auckland and central government are too thick to see the benefit.

          So rather than a horribly expensive band aid, that will exist until this government is voted out why not use the resources we have and make it as cost neutral as possible until the line is ultimately wired up.

          Beats your superior intellect suggestions however.

        7. You counting needs work, Waspman. The batteries are nowhere near $100m. Maintenance is not bugger all. All the operational extras required to run a seperate diesel fleet cost substantial money. It’s all there in the report. Nick R ably demonstrates the economic comparison with the proposal just a few posts above. Dead money is dropping over 15% of the lifespan of the motorway bridge, becoming liable for a part of the costs that incurs and generating unnecessary chaos just to bring forward a project that doesn’t actually save any money or provide additional benefit.

      2. But the whole Auckland metro area was electrified for $80 million – I think that’s the point where the economics get ropey. I can’t see how electrifying the 80km gap between Papakura and Hamilton would cost more, given similar km and lower train density.

  4. For (2) and (3) we have two solutions as such to answer those questions.

    It seems sending the wires to Pukekohe will be done when the Southern Motorway is upgraded between Papakura and Drury at that time. Given the bridges need to be lifted at Drury you might as well knock two birds out with one digger – so to say. The Manukau-Papakura upgrade is not complete until apparently next year meaning the next upgrade wont be until at least 2020 starting.

    As for the B-EMU’s – there is another post by another Author meant to be on the way but this is where the Manukau South Link comes back into the mix. To get best bang for buck of the B-EMUs: “suggested that the B-EMUs be used to run from Pukekohe to Manukau outside the peaks while a direct Pukekohe to Britomart express service (with a stop at Papakura and Otahuhu most likely) would occur in the morning and the reverse in the evening. Given the constraints at Britomart pre-CRL and the fact the South commutes within itself (for the most part) running Pukekohe to Manukau every 15 minutes outside of the peak is not an idea that should be dismissed.
    The 17 new B-EMUs gives an extra capacity of 6,375 passenger to the existing 21,375 capacity already available with the 57 EMUs we currently have. If we were to run a standard 3-car B-EMU between Pukekohe and Manukau every 15 minutes (four trains an hour) we have capacity at a rate of 1,500 passengers an hour heading in and out of Manukau to go south (double if running 6-car sets). The question becomes is that enough capacity, not enough or too much capacity? Given Takanini to Pukekohe is the fastest growing residential/population growth area outside of the City Centre coupled with Transform Manukau by Panuku Development Auckland, 1,500 passenger per hour capacity in and out of Manukau City Centre going south would be enough for at least until the City Rail Link goes live. Remember Papakura to Manukau in 14 minutes via the South Link is the fastest way to get to South Auckland’s core reliably (the Southern Motorway is not reliable) so the level of service and the speed compared to alternatives makes the service attractive from go.”

    Source: https://voakl.net/2017/07/24/battery-powered-electric-trains-to-pukekohe-marks-the-return-of-the-manukau-south-link/

    That was written before the Southern Auckland infrastructure announcement yesterday in which the Drury West and Paerata Stations were now in the mix to be built soon. Given that and the new B-EMUs the South Link solves (2) and (3) until more EMU’s are brought through for the CRL.

    1. How do you plan to run a peak express with only two tracks for most of the line? They would just end up crawling behind the train ahead.

      Regarding capacity between Pukekohe and Manukau – short answer is too much. There is no way an off-peak service that doesn’t involve Britomart is going to run into any sort of capacity constraints.

      1. I will let Stu answer that question as he was the one that first touted it in regards to express running.

        As for capacity – you assume all go to the Isthmus which is the wrong assumption with Southern Auckland and the Southern Line in the off peak. Most trips are within the sub region itself so linking up the residential population with its core (it isnt the CBD) would be most prudent.

        1. All the hop data I’ve seen suggests that the proportion of people that go between stations on the Southern line is pretty small.

          Any that is not the point, 3-car off-peak services that run the full length of the southern line are generally nowhere near full, there is no way a 15-min service from Pukekohe to Manukau with five stops in between is going to run into capacity constraints. It is more likely to be waste of expensive EMUs and valuable operating expenditure.

          This would be much better served by getting better off-peak frequency on the existing services so transfers at Puhinui are easier.

        2. Ben’s looking at current commute patterns in general, other modes included, so the numbers could be surprising given it’s too awkward now. Looking forward to the new HOP data too from last year, be interesting what changes there are.

        3. Thanks for the link. That post has no facts that back up the assertion that 4tph to Manukau from the south off-peak will be viable, fig 7.12 seems to suggest the bulk of commutes to Manukau come from nowhere near the railway line.

          It appears to just draw a long bow between the facts that people who live south often work south and the fact that a direct rail connection to Manukau would be quicker than the current options for people who live near the rail line.

          The dismissal of Puhinui transfers based on there being others transferring there as well is odd. I’d think you would have to go well beyond 2030 before this becomes a good use of PT dollars.

        4. This also assumes that Kiwirail and Ports of Auckland are willing to give up their Wiri inland port, which would be rendered virtually useless if the Manukau branch southern link was to cut across the front part of the Wiri sidings. It would be impossible to use the freight sidings while passenger trains are operating all day.

      2. I’m interested in the model that supports your assertion for the Pukekohe to Manukau statement.

        Even if there isn’t a capacity constraint today, if you improved the service, could there be one in the future?

        The conditions that lead to the constraints are as much about our biases as they are about how the network is currently designed to work.

        1. I don’t have a model I’m just using common sense. I’d be equally interested in what modelling suggests this service would be viable, let alone run into capacity constraints.

          The idea that 1500 passengers per hour off peak could come in and out of Manukau to the south is absurd. I doubt more than that pass through Britomart off-peak, and that is with four lines from all stations on the network arriving there. There is only one significant thing to go to in Manukau – MIT.

        2. My expectation is that to start with only a 3 Car unit would be assigned, but if there was capacity to assign a 6 car unit great.

          I’m interested in how the effect of investing in congestion free alternative networks will change travel decisions and what that does for farebox recovery.

        3. The 1500 capacity is with a 3-car unit. There will be people who don’t catch this ‘service’ now as it requires 1 or 2 transfers that would if it was a single service but there is no way this would be enough to justify the running costs, let alone capital expenditure of the new junction. It would just suck up money from other PT. Get the existing line frequencies up meaning a transfer at Puhinui is painless.

          I think farebox recovery will get better and better meaning we can invest in better frequencies, especially off-peak meaning better connections for everyone, not just those on the line between Pukekohe and Manukau.

        4. Turn-up-and-go frequency (10-12 min or better) all day every day on the main routes more important (& upgrade Puhihui Station). If that’s overloaded then consider direct service, otherwise just single seat journey fetishing at expense of frequency.

        5. There apparently isn’t even demand to warrant a direct bus service between Pukekohe and Manukau. That doesn’t bode well for the idea of a direct train service.

  5. This battery-powered option presumes that the existing overhead system can cope with the extra demand while the train is heading towards the CBD, at which time the current will increase substantially. What time will this recharging current be required in addition to the normal running load, and will the recharging time be sufficient to fully recharge the depleted batteries? All of this seems technically feasible, but all to overcome the reluctance of the government to provide the funds for early electrification of the Puke line. As Waspman comments, a VERY expensive solution, using short-life batteries!

    1. Well the trains would be under battery between Pukekohe and Papakura, and under wires charging from Papakura to Britomart and back to Papakura. So about two hours to charge under the wires, should be plenty. Trains spend very little of their time at anything close to full load anyway.

      I’m not sure why you think this is so expensive, electrification costs $160m on top of the fleet costs, it’s by no means cheap!

    2. With regard to charging the batteries, you should never be in a situation where the batteries need to be fully recharged. This is both due to limits in battery chemistry (assuming that they’d be using lithium tech, it’s a “bad idea” to deplete the battery beyond ~30% due to stresses it places on the cells) and operational constraints – If the batteries were to be significantly depleted during normal operation, the capacity of the batteries is insufficient.

      If using lithium tech, assume charging rate of 2-4x capacity (2-4C) – IE: 2Ahr battery recharged in 15m-1hr. Having said that, it’s generally preferred to charge at 1C where possible. I’ve no idea how much power the current EMUs draw, however it _appears_ that the power put into actual traction is about 2.7MW (https://at.govt.nz/media/imported/4678/AT-electric-trains-technical-summary.pdf) – Using those stats, to provide 1hr use (fully depleted) is only 108Ah. Not bad.

      As for current increase, I honestly doubt that you’d notice. It’d be like a train taking off hard to make up time vs taking off slowly, as far as additional load is concerned.

      1. The train supplies current back to the network through regenerative braking at some points anyway, we could probably us only that power and still get enough to charge the batteries.

  6. This is really good news and delivery in 2019 too. Since Pukekohe station is being rebuilt then perhaps this will include a Km or two of OLE with grid feeder to provide an end of line recharging capability for the bemus. Or more importantly to allow for overhead power to initially accelerate the bemus up to running speed before dropping off the OLE and continuing on bttery power. Same at Papakura end although the OLE already in situ for some short distance south, maybe this will be extended if needed.
    This could significantly reduce the power cycling of the battery bank by allowing a much shallower depth of discharge, hence maximising bank life.

    1. A train leaving Pukekohe would need little charge left in the battery as the line drops soon after leaving the station. I would suspect regenerative braking would mean that the battery would still have some charge left when the train reached Papakura. Conversely the battery would need to be well charged when the train left Papakura for the climb to Pukekohe.


  7. I just don’t quite understand the logic of buying 17 three car bemu sets when they are going to be almost permanently run as six car trains. Why not buy nine 6 car sets since that would improve capacity, improve security/policing with full length walk through of train and perhaps increase reliability in event of battery failure since there would be two battery cars?
    Is it that the 3 car bemu is an off-the-shelf design? CAF inot willing to do 6 car bemu?

    1. There is no need to buy 6-car sets, and the existing 6-car trains are years away from reaching capacity. There will by more EMU purchases in the future post-CRL, this is when we should start thinking about 6-car sets.

      1. Ok, again the logic escapes me, so instead of ordering a six car set with just 2 driver cabs, thats one at each end, you say it makes more sense to order two 3 car sets and have them joined to make a six car set and end up with two completely redundant and completely functionless centrally located driver cabs instead of some extra passenger seating and through corridor simply because we dont need the extra passenger capacity now? What about the very significant waste of $ on unused driver cabs or increased opex for more security staff in each 3 cars?
        Please explain

        1. They won’t fit in the depot to start with, fixing that would require a new depot at probably $100m+ as also means things such as train lifts.

          Also there always needs to be one set spare, with 6-car emus you need 6 cars sitting idle/being maintained whereas with 3-cars you can take one set out easier.

          These new trains are the same as the ones we have now with the exception of the batteries. They probably want to keep them the same as much as possible to get these in service faster.

        2. Is there any evidence that single six car sets are cheaper than two three car sets? If they were significantly cheaper then sure, but the amount of trains I see joined together around the world suggests this might not be the case.

          If the only benefit is capacity then as I said this is not today’s problem.

        3. The trains are in operation for say up to 18 hours a day, peaks are say 6 hrs of that. maybe the number of trains running in off peak times goes from 1/4 to 1/2 of the total trains available.
          faults and maintenance can be done easier on a 3 car set without putting another 3 car s in the workshop at the same time. A lot of small faults would only affect 1 carriage, so you are only off lining another 2 cars, pulling out another 5 cars for a single carriage fault is ridiculous.
          As for the extra capacity of 6 cars off peak, well its better to have the higher frequency every time of 2 separate 3 car trains

  8. …and that standard excuse that 6 car set won’t fit in maintenance depot doesn’t work, get the depot sorted out instead, put a canvas tent on one end if needed

    1. A canvas tent over the 25000 Volt OHL on the end of the maintenance shed to protect a $9-million train. Genius. No way that could possibly go wrong. Notify AT immediately and present your invoice.

      1. The point was that some extension to a line at the depot to accomodate a 6 car set should not mean complete demolishing of the existing depot and a new 100million replacement depot constructed as was previously suggested. If a 6 car set sticks out one end of the depot then arrange some simple rain proof cover extension.

        1. The problem isn’t having one end of a train sticking out of the shed in the rain, they all live outside most of the time anyway. The problem is that the shed itself is divided into 75m bays with a “pit” under suspended rails for technicians to access the underside of the trains and also sets of jacks to raise the trains above the rails in adjacent roads. A permanent 6-car train at 144m long, or indeed a 95m long 4-car as rejected in the business case, is too long to fit in the service bays to be repaired or have their 6600km check up, or even to be assembled and commissioned in the first place.

          Building a longer shed is all very well for the next (post CRL) tranche of trains when a new yard and new maintenance depot and new substations and all kinds of other junk is required anyway. Then they may well go for the permanent 6-car solution at that time. But at Wiri, the restricted space between the motorway and Wiri Station Road means there simply isn’t room to extend the shed and still be able to get the units in and out from both ends. Now that the fleet is in constant circulation, any attempt to take the depot out of service to rebuild the shed internally would cause worse problems than this whole saga is intended to fix.

        2. I understand the limitstions of the existing wiri emu depot. The issue of fitting a complete train in a service bay is one of convenience only. We are already talking of future proofing station platforms and the probable need for 9 car trains.
          Do you seriously propose AT construct new depots each time operational conditions demand a longer train set? Or all trains need to be made up of multiple short independant emus despite the staffing and security issues?

          Do we always have to have a full length service bay or is the concept of moving a train to bring the appropriate cars over the required bay area just too difficult to manage? Many maintenance depots have managed this is the past and still do where space resources are limited. Or is New Zealand a special case? Complete train service bays or nothing?

        3. From memory the current depot has the capacity to maintain a fleet of about 80 EMUs, after that a new depot is needed. Post CRL we’re going to need a lot more trains and therefore we’re going to need a new depot somewhere.

        4. David – you appear to have completely missed two points in Dookies comment.

          1. We will have to build a new depot at some stage anyway.
          2. They hoist the trains up when they are over the pit, it’s a bit hard to hoist half a train.

        5. ok, understand. WIth such a short order to delivery time its likely the timeframe would not permit a newly designed 6 car EMU. By ordering 3 car AM class emus with the only serious mod being the AMT fitted with Battery bank, AMB?, and battery management system (BMS) and CAF are probably very keen to get a BEMU into production then 2019 to 2020 delivery looks feasible.
          There will probably be some serious modifications needed to the existing depot so that battery banks/cells can be easily replaced and serviced

        6. Yes, now we’re on the same page. The ultimate goal appears to be to have 120 3-car equivalents rolling around Auckland, and the rest of them may or may not be CAF products, so one way or another a second maintenance depot is on the cards in the long term. There is speculation about where that could be, but I doubt there are anything more than the faintest plans at this stage.

          The current depot has unpowered roads in the maintenance shed that would be perfectly suitable for accessing the batteries in the roof, if that is indeed done from above, so depot modifications to accommodate the new units seem unlikely from this vantage point. The trains already carry two substantial batteries to operate all the onboard equipment, lighting, doors, ventilation, electronics, communications etc, so we’re looking at more of the same.

  9. “Interestingly the 15 normal EMUs have a higher BCR at 3.2, but it doesn’t address the issue of services to Pukekohe and therefore the cost of diesel shuttles.”

    You seem to be implying the batteries EMUs, have further benefits, and thus the BCR would be > 3.2. But surely when they came to a BCR for the battery EMU options, they had already taken into account that cost saving of not running the diesels?

    1. The IPEMUs have additional benefits compared to continuing to run diesel trains, but the also have additional costs. If the additional benefits are less than 3.2 times the additional costs than the BCR for IPEMUS will be lower than continuing to use diesel. AT may believe that there are strategic or operational advantages not covered in the BCR that make that worthwhile, this is likely to be improved service for Pukekohe resulting in improved patronage which may not have been captured in the model.

  10. this is exciting, If it goes well then Wellington can look at the korean version to run EMUs to Otaki and Maymorn/Timberlea.

        1. Yep, that would make sense, although I imagine it would be cheaper to just put some wires up as I don’t think there are any expensive impediments like the Southern Motorway bridge adding to the cost.

          I imagine they would need to boost track capacity through the Hutt first as there is no point extending the network if the trains are already full.

      1. Making Maymorn an electric services terminus isn’t about its current size. Besides being earmarked for future residential growth (as has already been mentioned), it is the next most logical location after Upper Hutt to establish a terminus before Featherston. Though there has been talk for years of eventually building two new stations between Upper Hutt and Maymorn, Maymorn has the advantage of being reasonably flat and having more room for any additional infrastructure that may be required befitting a terminus (e.g. a dock platform or potentially an expanded car parking facility).

    1. You got to be kidding about having a fleet of electric/battery Matangi’s for Otaki and Maymon extensions when the GWRC is fixated of getting rid of the trolley buses which have electric/battery capability and replace them with diesel buses. GWRC might dieselize the Matangi’s 😉

      I agree with you that have a small fleet of electric/battery Matangi’s for Otaki and Maymon extensions is a good idea but I don’t see it happening with the current GWRC management. They have tunnel vision.

    2. More import stuff needs to be sorted with the existing service. For example:
      – Integrated ticketing etc (stop waffling & get on with it, 20+ year farce: GW + govt = useless…)
      – Improved frequency – at least 10 min peak, 15 min interpeak & weekend daytime (7-7), not less than 30 min later evening.
      – Extended hours of service (to midnight) – undoing the damage GW did in 2004.
      – Connecting buses meet every train on main feeder routes.
      – Improved stations, bus stops.
      – Charge for parking at stations (rather than wasting scarce money on seemingly endless expansions).

      If all that’s done then maybe extension, but no real centre or anchor worthy of high frequency beyond Upper Hutt.

  11. I like page 37 of the detailed business case which says: “In the future, with battery development continuing, more housing growth areas further out, such as Pokeno and Huapai could be serviced without the high cost of overhead wiring. The former could be served without further addition to the proposed number of units once the suitable batteries are available. Alternatively, the Pokeno area could be served using the IPEMUs from Pukekohe once conventional electrification reaches there – assumed for 2025 – noting that growth will be progressive as the population of Tuakau and Pokeno grows from approximately 6,000 (current) to 20,000 people over the next 30 years.”

    Yes Pokeno & Tuakau seems the no brainer next step rather than Huapai, due to longer winding route and other issues, good bus lanes should be used out there

    1. Yes, it could well be a game changer in getting trains out to the far reaches of the lines without the time and expense of electrification (at least, initially) sooner rather than later.

      I am interested to see how Pokeno and Tuakau work given they are part of the Waikato, but perhaps also for extensions way out west?

      1. Out west is difficult because of the tunnel between Swanson and Waitakere. The tunnel it too narrow to allow for evacuation of the EMU in case of fire. Also going from anywhere past Waitakere (like Kumeu) to the CBD and beyond always is going to be faster along NW motorway, so I think we’ll never see those IPEMUs out past Swanson.

        1. A long time ago with skinner cars & we have more OOSH as I understand it from previous blog discussions. Let’s move on

        2. They were set up for emergency evacuation through the ends (to some extent), AM class aren’t. Wellington’s Matangi units have emergency ramps that can be deployed from the end doors for evacuation in tight single track tunnels (J’ville, Paekak’ escarpment). Any new trains through Waitakere Tunnel would likely require similar setup to Matangi.

        3. I think we’d need to see the Waitakere tunnel re-bored or daylighted or deviated, which almost certainly needs it to be worth something to both Kiwirail and AT. A revival of freight fortunes to the north in the next two years might have to be the other half of the equation if it’s ever to happen.

      2. When you say ‘way out west’ are you meaning the North West of Auckland where the NAL heads north? or do you mean out west from Pukekohe where there is a rail branch heading out to Waiuku?
        I’m surprised with all the talk of Papakura and Pukekohe main line services using BEMUs that there is practically no mention of the line to Waiuku. Anytime I drive there the roads seem busy with traffic to Pukekohe/. Is the Waiuku area not part of AC and are there no PT plans or housing developments planned there?

        1. It’s definitely part of AC. Not sure why has been off the radar so far, maybe it is surrounded by high quality soils so they don’t want to develop it? Seems like a logical addition to the rail network otherwise.

        2. Maybe not right off the radar, just noticed that TPH diagram above and there down at the bottom between Papakura and Pukekohe a double junction fine line just where the Waiukuline branch connects to the NIMT.
          Recognition at least…

  12. Something doesnt add up here. Papakura to Pukekohe by train is 18.2km. [ From Wiki NIMT]
    The journey time from AT website is 18 minutes. There are no stops and its dual track all the way.
    That works out as an average speed of 37 mph . Why so slow ?
    The elevation at Papakura is 19m while that at Pukekohe is 62m, so quite a gentle climb.

    The other baffling thing is the 3 tph service which is every 20 min requires 2 trains while the story says the IPEMUs being bought are 17! Clearly to recharge over the existing network they require more, ie 15 trains are adding capacity in existing network.
    My thinking is buy 4 only and recharge only at Papakura station between runs. That might take 3 trains with 15min charge The electric traction should mean faster, under 15 mins and the charging should be faster as they are stationary.

    Buying 17 trains is an expensive way to cover a service using 2 trains at 20 min frequency.

    1. We need the extra trains anyway for capacity across the rest of the network, the only additional cost is the batteries and the benefit is a direct connection for Pukekohe rather than the transfer.

      1. What is so terrible about changing trains. A friend the other day changed trains at Newmarket to get to South Auckland as there are no direct trains from West Auckland to South Auckland.
        Every where else with cities with substantial train network people change trains.

        It might make sense if the train arrived at Papakura with 50 people on board and 40 changed to the other train but the numbers arent like that.
        That doesnt mean the future will be the same.

        1. It’s an added transfer within an existing line as opposed to transferring to a different line. This means it is in addition to any other transfers further the journey, it’s a bit like breaking the Mt Eden bus halfway down Mt Eden Rd so people have to transfer to continue along the same road. About as logical as stopping electrification at Henderson and having a shuttle to Swanson.

          Also I imagine at least 4 out of 5 people would be transferring from diesel to electric at Papakura, making the majority transfer doesn’t make a lot of sense. Are you seriously suggesting that most people just travel Pukekohe to Papakura?

          I can’t think of many cities that have a train shuttle to the last station on a given line but I’m sure there must be one or two out there.

        2. Newmarket is clearly a different case to this. At New market almost everyone on the Southern or Western line at Newmarket continues on their line, and transfers between lines are the minority. At Papakura the majority of people on the Pukekohe line transfer to the Southern line. Transfers for the sake of transfers are a poor way to operate a transport network and we can eliminate in for $40-70m.

        3. Read the report – it’s more like 50 people arrive at Papakura and 45 of them change trains. And yes, 50 people on a DMU shuttle is a common thing. At peak it’s often over 100.

    2. One thing if the new trains were only pure battery trains they wouldn’t have to drag around the AC transformers or the dc rectifiers which would be a large saving in weight. They would be altogether a much simpler sort of beast.They could also be built with cab end evacuation capability so they can go through that dam Swanson tunnel which is stuffing up extending the network to Kumeu.

        1. Well there would have to be chargers installed at Pukekohe and Papakura which would be another saving in weight as the duel mode emu have to carry the charging equipment onboard. I am not advocating this but its worth considering because I don’t think the transfer at Papakura is that bigger deal. After all with the new network out here in the south plenty of passengers are transferring to bus every day. Its working far better than I ever thought it would.

  13. Huapai is worth doing for those going from the outer north west only as far as Henderson. It will always make more sense to use the current NW motorway to get from Huapai to the CBD or the airport by car until a proper busway with frequent, reliable & prioritised row is in place. Could be light rail up the NW M’way.

  14. I agree that AC needs to purchase the B-EMU’s, as it is a cheaper option to electrify services to Pukekohe and possibility Pokeno and re-introduce passenger services from Swanson to Helensville.

    With regards to electrifying Papakura to Pukekohe and spending $160 millions to do it, it would make more sense to electrify from Papakura to Hamilton as part of the third rail proposal, for fright and passenger services, especially since there 2 inland freight ports being built in Hamilton.

      1. I am not sure what you mean?, Are you referring to to the proposed east/way motorway of the the much cheaper 3 rail line?

        To me the 3 rail line and electrifying from Papakura to Hamilton to cater for the 2 inland freight ports in Hamilton has a much higher priority to get trucks of the road, operate passenger rail services between NZ’s largest and 4th largest cities.

      2. Personally I think the underground tunnel from Britomart to Mt Eden is a total waste of money when they already have track via Newmarket that only adds on a few more minutes, but they still wasted money.

        1. The track via Newmarket is completely full and cannot run any more trains. The tunnel doubles the number of tracks into the city centre from two to four, doubling the number of trains that can run. It also puts two new stations in very useful places at midtown and uptown, and make the route from the western and southern lines to the city faster and more direct.

  15. Arum1: “I’m not sure that electrication to puke is our highest priority as it will take several years for significant resedential growth to occur there.”

    James B: “I’m not sure there’s any business case that would justify Puke electrocution”

    If we wait until after the residential growth, that residential growth will have developed for for car-dependent residents, There will be large parking lots around big box retail and malls, and the kids will have to be driven to get anywhere. If we provide fantastic public transport to the greenfields developments before the residential growth happens, that growth can be walkable, liveable, people-centred with smaller sections. The very fabric of how that growth develops depends on getting the PT there first.

    Rail down south, and busway, LR or HR out west – whatever works – is needed before the growth. If these battery-powered trains help us to provide that now, great. Talking to a man from Korea recently, this is absolutely what they are doing there. It is only our car dependency that blinds us to this approach.

    1. This is a really really important point. In many ways, how we design transport and connectivity drives (excuse the pun) the urban outcome.

      If very early on we make PT, walking and cycling the focus, then we get a much more satisfactory outcome. That does not mean we ignore the car. But we don’t make it the dominant form from the outset.

    2. 98% sure James B was taking the piss and rolling with the typo, i.e. it would be a strange old world if it made economic sense to electrocute Pukekohe.

      1. 🙂 I’m responsible for too many silly threads, but I had to leave it in there in case someone else had fun with it… made me wonder though if electrocution could keep the development off the fertile soil?

    1. And all sorts of other things to improve the city with the other 90%. You’d be hard-pressed to think of a worse use for that money.

        1. I just had a squiz at AT’s promotional material on that. Trying to stay positive about our city but FFS! It’s one step forward and 3000 back. On top of the EWL it Just makes me want to cry…

  16. I don’t think spending $41-74M on batteries that will last 7 years and having a part of the network requiring special trains is a particularly great outcome.
    There are a lot of unused EMU’s parked at Taumaranui. I’d guess some of these could be used to keep Pukekohe services (or future services to Tuakau or Pokeno) going another couple of years.
    Hopefully thats enough of a compromise for bringing forward $160M on a motorway bridge and overhead lines that would last 50 years. Also allowing for the Huntly bypass and its 20 bridges to be completed, and the Hamilton bypass with its several, thus freeing up construction capacity / getting a better price.

    1. Those are not EMUs parked up at Taumaranui but just unpowered SA and SD cars. The batteries used in the BEMUs will likely exceed a 7 year working life, the 7 Years is probably the warranty period.
      The $41m battery cost includes the complete battery support system, rectifiers, transformers if needed, control electronics, charging and charge balancing electronics, etc.. subsequent battery replacement may only refer to the actual battery cells which should be a fraction of the original $41m
      It appears this AC/AT contract with CAF may be a first for CAF to provide BEMUs

      1. Does this battery powered train decision mean there are surplus diesel trains which Greater Christchurch could use to trial a commuter train system with?

        1. A lack of political will and a poor understanding of how cities work in my opinion too. It is so damn frustrating.

          For us down here in the south it is frustrating because after the earthquakes there was the perfect opportunity to get train loop track right into Christchurch’s CBD but instead the government did the opposite….. obstinately refusing even to trial a commuter rail service on the existing underused tracks….. Stupid is what stupid does….

        2. A few anti-rail idealogues stymied the rail opportunity in Christchurch. One politician in particular – we all know who.

          This lost opportunity is a travesty. And they still show no enthusiasm for making use of the existing lines.

        3. +1, Greater Tauranga, Hamilton, Christchurch and Palmerston North-Wanganui are perfect for rail.

        4. A Palmerston North–Wanganui passenger service: is there really enough commuter traffic on that route to make a rail service viable? I’d have thought now that the Manawatu Gorge road is stuffed, you might have better luck running a Palmerston North–Woodville shuttle service for all those north Wairarapa commuters working in Palmerston North and environs and sick of getting their cars banged up on the sub-par alternative road routes.

  17. It would make sense to use the battery trains for a Huapai service also, it would not need to be every 10 minutes but half-hourly would be a good starting point.

    1. It couldn’t be every ten minutes. It’s single tracked. Extending service to Huapai will almost certainly need the Waitak tunnel lowered and widened.

  18. This is certainly removing pressure on NZTA to rebuild the Drury bridge. Look for this “interim” solution to be around for many many years.

    1. Also think on this:

      So you think it is all right to spend all this money on an interim solution when a large amount of pressure should have been put on NZTA to get that bridge replaced in the very near future so the money could be spent on tranche 2 of the current EMU specification.
      The addition of batteries increases the complexity (using the current 57 as the base) and the maintenance cost.

  19. Don’t know if everyone else saw Ben Ross post in the NZ Herald/Spinoff today?

    Basically the 3rd main case has now been un-redacted and it says that the 3rd main is actually the 2nd best option…. why? Because the 4th main is THE BEST OPTION…. I’ll let that sink in…..
    I hope GA reposts this article since the more it is spread the more pressure can be applied to the government to get this built asap.

    1. Don’t you mean pressure can be applied to the new Labour/Green government? A much easier option than persuading the present dinosaurs about public transport.

      1. I don’t imagine a Labour/Greens/NZF government would need much persuading to approve this (well the 3rd main at least) since all 3 are pro-rail (Greens/NZF in particular are).

  20. I thought thus government had committed to electrification much sooner than later, during the election campaign?

    Secondly I wouldn’t bet a hop card on battery technology improving noticeably. The Nissan Leaf has been around 8 years and the advancements have been very minor and even then largely confined to battery temperature control.

        1. They will run off new overhead lines like the rest of the network; much better than diesel anything.

        2. I remember the 2017 promises from both Labour and Nats that they would electrify Papakura to Puke. Nothing to date. I thought one main reason for not ordering IPEMUs was because the wires would be up within a few years of ordering EMUs? Ie before 2020 2021.

        3. We are not getting battery trains. Electrification to Puke, like 3rd main, seems to be going through incredibly tourtuos NZTA biz case process…

  21. I’ve heard through the grapevine (which I can’t mention) that work will start this year on getting the Drury
    bridges suitable for Emu use.

Leave a Reply

Your email address will not be published. Required fields are marked *