Since the majority of the Auckland network went electric last year, people travelling to and from Pukekohe have had to catch one of the old diesel trains as a shuttle from the end of the wires at Papakura. As I understand it the main reason the wires weren’t extended further than Papakura was the cost. It might be one extra station but it would represent about 36km extra of track to wire up.

Coming to the western line soon
Could battery powered trains be coming to Pukekohe

The diesel shuttle is ok as a short term solution but long term we’re going to need to do something about extending electric trains south of Papakura, especially as there is a ton of growth planned for the area with tens of thousands of homes to be built in close proximity to the rail corridor. Electrifying the line was included as part of the Transport for Future Urban Growth consultation recently.

TFUG - Draft Preferred Plan - South

We’ve seen in the past that electrifying this section of track isn’t cheap and combined with new stations to serve those developments and AT have estimated it at over $100 million – I’ve seen some estimates as high as $140 million. Even so the business case we saw in 2012 suggested the economic return was ok with a BCR of 2.1.

At some point, last year I think, there was a suggestion that Auckland Transport were looking at an alternative solution to the traditional stringing up of wires, getting trains with batteries attached. Now it seems AT are talking much more publicly about that with a report from Radio NZ suggesting that this idea is looking more and more promising.

Auckland rail commuters could be riding in battery-powered trains within a few years if the city’s transport agency can find the right technology – and the money.

Auckland Transport said a new fleet of electric trains with large battery packs would be able to serve towns beyond the end of the electrified network.

The agency has been working for more than a year on the project, along with the Spanish train-builder CAF,which supplied Auckland’s 57 new electric trains.

Adding four-tonne battery packs to a new fleet of electric trains is being studied as a cheaper option to extending electrification to the southern town of Pukekohe.

Commuters there have to shuttle in ageing diesel trains to reach the electric trains at Papakura.

Project manager Lloyd Major said battery-powered engines were cutting edge technology, but a new generation of batteries developed for electric cars made it more viable.

“Consequently we looked at the feasibility of doing it in Auckland. The Spanish manufacturer CAF is very confident, to the extent that six months ago we thought we’d need to build a prototype, but now we think it’s more about just finding the right battery.”

There’s perhaps a little irony if it was due to the development of electric cars that battery powered trains became a viable solution, and it’s an interesting solution at that. I don’t know anything about the financials behind option but I imagine it could save tens of millions if not more from not having to run wires and that would obviously be a good thing – although it should be noted that electrification to Pukekohe should really be being paid for by the government like they did with the rest the Auckland network.

The report says that AT would need about 18 trains to serve Pukekohe (and the southern line) and that would also allow the trains currently used on the line to be freed up to bolster the capacity of other services around the network so pursuing such an option could see train capacity improved faster.

While I’m sure there is still plenty of work to be done to see if it is feasible, the idea seems like it could be a good one. Who knows perhaps it might also one day allow trains to travel further afield such as to Pokeno which sits outside Auckland’s boundaries.

What do you think of the idea of battery powered trains?

Share this


  1. Before anyone says we should have got battery powered trains instead of stringing up wires – battery powered trains cannot exist without overhead wire sections because it’s when they traverse the wired sections that the batteries recharge.

  2. Bombardier have already trailed a prototype BEMU in the UK and CAF are offering it as a retro-fit on Birmingham’s new trams, so starting to mature and can be fitted within NZ loading gauge. If AT are smart they might be able to negotiate a good price as guinea pigs. Perhaps explains too why they have seemed so relaxed about ordering new stock. Of course the other thing CAF might be happy to do is a version with a diesel generator that could go all the way to Hamilton!

    1. I thought was being discussed with AT and CAF, as CAF wants to go into the global train market with battery powered trains

    2. Extending Auckland’s network to Hamilton as is would imply a service comparable to an S-Bahn between two cities 100 km apart, without having a real inter-city service. A world first perhaps, but not sure it’s something I’d be gloating about. Ideally we’d have a fully electrified North Island network with HSR bewteen WHA-AKL-HAM-PMR-WGN and whatever other stops you want in between (TGA line?), as well as Auckland’s fully electrified local network. But I’ll keep dreaming.

        1. High speed trains are not possible on the current network and even if they were what do the do once they get to Auckland and are stuck behind the all stops metro trains?

        2. High*er* speed trains could be possible. Queensland runs passenger trains at 160km/h on our track gauge, and test runs of those have been over 200km/h. Japan has numerous diesel and electric MUs on intercity service that run 140km/h and use tilting technology to maintain these speeds through corners. If you can average 140km/h you’d be in Hamilton in less than a hour. Britain runs trains well over 200km/h on legacy lines with our same loading gauge (i.e. physically small trains), admittedly with standard gauge track.

          Within Auckland you could use the third main to bypass metro trains, eventually lengthening this and adding sections of fourth main for passing. Once your on the eastern line it probably doesn’t matter if you timetable it right.

        3. Nick R: the only trains in Britain that run over 200km/h are on the purpose-built high-speed HS1. Trains have been designed to run at 225km/h on the legacy network, but that has never happened in service for signalling and braking reasons.

        4. HSR Dedicated Lines tend to only allow trains to go at least faster than 200km/h with proper cab signalling. So it shouldn’t be a massive burden on the current network.

  3. Perhaps the Government should take the money they’re throwing at electric cars and instead put it into battery electric trains?

    Might find they save more carbon that way :).

    1. I have seen no evidence that the current governmental encumbants wish or plan to save carbon. They signed us up but are MIA when it comes to a strategy.Must ask my local MP what the plan is…

      1. Well the obvious, and more than somewhat elegant use for a small part of the power company sell-off treasure would be rail electrification work. At least it would be an investment in a future market for those companies, along with all the other clear benefits.

        Instead your local MP lobbied for and got hundreds of millions for motorway widening.

  4. The section should just be electrified. It’s ridiculousness in the extreme that NZTA won’t fund this but the same paltry 100 million would be provided without question if it was an intersection upgrade. This is peanuts. The real tragedy is that time is being wasted on half-baked battery powered trains whilst billion dollar additional motorways such as east-west link are rushed through with urgency.

    1. Yup. $100m for 36km is a bargain how much it costs to upgrade one motorway or expressway designation.

      It future proofs the rail line for further station development. It makes total rational sense.

  5. It does seem crazy to me. Extending the electrification bit by bit seems the obvious solution.

    1. If this option is cheaper then why not do it? If the functional outcome is the same why does the specifics of the technical solution matter?

      1. Because while the battery-fitted trains may work just fine on the regular tracks – and need to access those tracks to recharge – we effectively can’t use the bulk of our train fleet on the track to Pukekohe because they don’t have batteries. That’s fine if there are only a limited number of trips from Britomart to Pukekohe a day and can use the battery trains for these journeys. But with the development in the area I would expect the goal to be to increase frequency which places constraints on the number of trains we can use.

        Also batteries are not light. Nor cheap. Nor last forever. I would want to see the BCR on this vs electrifying the whole track.

        1. The flip side to that is if we get 12 trains with batteries for the same price as running wires, then we have 12 new trains for ‘free’

        2. Sure all those issues you raise need to be considered in the assessment. The number of trains bought would need to be consistent with the proposed service levels obviously. But it doesnt give any reason in and of itself that a battery issue would be inferior. As for weight that would need to be considered, but 4T would presumably not be a lot of weight in percentage terms. I dont know the weight of a standard laden passenger train but I would imagine that 4T would represented 2-3% at most.

          Edit: I see Nick R has answered the weight question downthread.

  6. Technology aside, the implications for passengers to and from Pukekohe are significant; no transfer at Papakura and the same service frequency as Papakura.

    All trains on the Southern line will have to be battery equipped so the figure of 18 sets sounds about right. Might have to have some pantograph down system on leaving Papakura for Pukekohe but that seems straight-forward.

      1. Yeah they are unlikely to run all services the full length knowing AT. Also the travel time once these are finally implemented will likely be a bit longer than current DMU’s + transfer due to the additional stations along the way.

        Interested to see if similar can be done out west, not sure if the Waitakere tunnel is a width or height restriction, with the batteries there wont be a need for overhead wires in the tunnels and the photograph can be lowed so don’t see height being a problem, width I am not sure.

        Distance Swanson-Huapai (14.5KM) compared to Papakura-Pukekohe (18KM), means range shouldn’t be an issue.

        Still a bit skeptical about the battery idea, but as it looks like they have changed there mind and now deciding on ordering new units for this purpose rather than retrofitting current ones I am a bit more convinced.

        1. iirc about 18km from Swanson station to Waimauku station, which would increase the catchment for any northwest service.

        2. Waitakere tunnel’s width is reputedly not sufficient to allow for emergency egress from the EMU side-doors. End doors like the Matangi would be ok.

        3. The ADL or ADK DMUs had side doors only. Why was this not a restriction for them?

      2. It will have to be the same frequency. Nothing is surer. Consider the timetabling and dispatching issues at Papakura if it isn’t.

        1. Quite the opposite, no issues with timetabling and is a fairly common practice in most rail systems. Hell they aren’t even planning on increasing frequencies out west past Henderson after the CRL, they want to leave it at the 10 minutes it is now

        2. 18 units is, according to the back of my envelope, about the number needed to run Pukekohe to Britomart at three trains per hour using six car sets. So it seems they are planning exactly that, half the trains on the Southern run from Pukekohe.

          Interestingly, doing that would allow trains to be shuffled around such that they could run all six-car trains at all times on the other lines, except for the Onehunga branch.

          So not only would this ‘electrify’ Pukekohe and remove the transfer, it would also maximise the capacity of Britomart.

        3. +1 Nick. Maybe this has been ATs plan all along but for some reason they have kept it quiet. On this blog we have suggesting for some time that AT needs to get on with ordering more EMUs to make more 6 car sets so if this kills two birds with one stone then great!
          Even if AT do extend the electrification to Pukekohe in 10 years time it will still be worth it as either the battery units can be removed and they just become normal EMUs or they are redeployed out West as Peter N suggested.
          Of course in reality the government should be electrifying the entire NIMT between AKL and WLG and out to Tauranga.

        4. Well the simple fact is that even the best batteries have a limited life and would be effectively be exhausted and fully depreciated in five years or something. It’s basically opex.

          However, I’d assume these battery trains would be literally nothing more than a regular AM class EMU with a big battery bank under the floor and a relatively simple inverter rig. When the battery is pukeroo’d you can put a fresh one in, or just pull it out and be left with a regular train. Nothing lost but the ‘consumables’ cost of the battery.

          I’d say this is something of a stop gap measure to full electrification. Buy the extra fleet we need but spend half a million bucks more per train, or whatever it may be, to avoid the need to wire to Pukekohe yet.

        5. “The report says AT would need about 18 trains to serve Pukekohe”

          Depends on what they mean by the word “trains”. If taken literally (as I did) it indicates a 10 minute frequency. If taken as a 3 car set and a train is actually 2 sets then it would only be sufficent for a 20 minute frequency. Since Lloyd Major (ex Kiwi Rail engineer/manager) is being quoted I tend towards the literal interpretation…but then the article has been reporter-degraded.

          As for recharging at Pukekohe…it scarcely seems worth it for the following reasons:

          A stationary train is a train that is not earning (financially or in terms of other benefits).

          The Papakura-Pukekohe return journey is going to have a time allowance of 40 minutes – 15 minutes running each way with a 10 minute allowance for turnaround and timetable slippage. That’s 10 minutes to charge the battery compared to the best part of 2 hours under the existing catenary. Rapid charging is detrimental to battery cycle life and the cost of battery charge regulation circuitry (some sort of switch-mode power supply) is roughly proportional to the maximum current it is designed to handle.

          A short section of 25 kV catenary is going to need a transformer, switchgear, breakers etc and a feed from the nearest 33 kV line. It’s going to need masts, tensioners (or possibly a solid busbar) plus shielding for the overbridge. It will also raise the issue of track circuit/signalling immunisation. Seems like a lot of expense for 10 minutes of charging (less if the incoming train is running late).

          The article states that the battery range will be 36 km so I am guessing that any in-service charging at Pukekohe has been ruled out.

        6. The EMU’s already have the inverters. My understanding is the AC voltage is transformed and rectified to 1500 volts DC. Presumably the batteries will be 1500 voltages. And will be feed through the normal inverters to the traction motors.

        7. @MFD – The 25kV overhead needs to be connected to at least 110kV due to the load size. Pukekohe has a 110kV linefrom Transpower Bombay, and it terminates at a substation 200m from Pukekohe station platforms. However, it would probably be better to use Drury and its 220kV substation.

        8. That question of signal shielding raises a good point. The EMUs use AC traction motors and the signalling within the electrified area is shielded to cope with the interference they emit. The signals south of Papakura are not up to the same standard and may need to be upgraded to handle AC motors whizzing past, even if there’s no actual 25kV catenary installed. Or, are they planning on fitting the BEMUs with DC traction motors like the KR fleet?

        9. “The 25kV overhead needs to be connected to at least 110kV due to the load size.”

          The context is a battery charging facility for a single train rather than traction supply. My estimates suggest a load of around 0.5 MW and based on my recent experience with installing a 15MW load a 33 KV feed would seem to be more than adequate but if 110 kV is available nearby probably preferrable.

    1. A fundamental basis of AT’s new network is the necessity for transfers for many journeys, so why should this be any different for Pukekohe passengers?

    2. The Puke passengers always got shafted and having no change at Papakura will have them getting shafted again. Whenever there is a delay on the network the Papakura to Puke gets cancelled and put on buses, having a dedicated service only has this happen when there is an issue directly effecting that line or its trains.

  7. Whats the range of the battery? Wonder if same tech could be fitted to a matangi and get them from waikanae to otaki.

    1. Should be 50km before need to recharge and recharge takes 20-30 minutes either while the EMU is stationary or moving under OLE. A retrofitted 3-6 car set EMU can get to Pukekohe and back before need of recharge but the safest thing would be to install at Pukekohe, a set of poles, wires and a 25kv power supply to charge the EMU prior to next service departure.

      Once Papakura-Pukekohe gets its full OLE infra, the battery-fitted EMUs can be redeployed between Swanson and Huapai or even Swanson-Waimauku and then when there are more people living just south of the Bombays, between Pukekohe and Pokeno.

      1. The EMUs have a “shore supply” connection that can be used to power them without the pantograph being activated. This could presumably be used (whether modified or not) for a quick top-up during the turnaround period at Pukekohe.

        1. Simply a big fat cable that plugs into the underframe equipment so that a unit in the depot for servicing can be powered up for testing, repair or whatever without having to be fully on-line. Not all the depot roads are electrified and not all repair/testing/maintenance procedures would be possible with the pantograph up. Very similar looking to the kind of cable that an electric car uses to recharge. I don’t know about current limits of the standard EMU setup – unlikely to be of the order required for fast charging of a traction battery, but the possibility is there for a similar method. i expect there would be some pretty serious safety procedures for manually hooking up and activating a charging cable carrying somewhere between 1000 and 25000 volts. It may be that a short stretch of overhead is still easier.

  8. More money is being spent on a single intersection at Kirkbride Rd. Just electrify already.

    AT could do both of course. But it will take time to develop a prototype, and there will be technology risk. What is the extra cost involved over the cost of the standard CAF trains? Questions, questions.

  9. Complete lack of vision by all participants.
    AT’s thinking is restricted by the Auckland Council boundary.
    Kiwirail and Government thinking is restricted by the $ cost.
    If we are to develop a transit system that we can all use, we must provide for Auckland commuters south of Auckland’s boundary.
    That includes Tuakau, Pokeno, Te Kauwhata, Mercer, Huntly, Hamilton. All the above have new subdivisions and commuters to Auckland by road.
    All are on the existing double track rail line.
    Wake up AT and Kiwirail and electrify the track to Hamilton. Otherwise you just lack long term vision.

      1. “Why do we want to subsidize exurban development? I certainly dont.”

        It’s happening anyway, with motorways and multi-lane roads planned beyond the current edges of Auckland. The question isn’t about whether or not it will or should happen. It’s about how do we want people to travel to and from these places? Should they be totally road-centric, or should they offer choice?

        Rail is the best mode for the likes of Huapai, Kumeu, Tuakau and Pokeno, as every passenger gained on these longer trips can be as valuable in lowering emissions and alleviating congestion as half a dozen passengers on shorter trips in the inner suburbs. It always amazes me how some folk say rail to such places can’t be justified, yet they don’t blink an eye when somebody proposes far more expensive motorway-based, car-centric solutions.

    1. There really aren’t many commuters from Waikato to Auckland – in the 2013 census, “4,827 people from Waikato district and 1,530 people from Hamilton city [gave] a workplace address within the Auckland region”, out of 596,313 people total working in Auckland (, page 11). The number of people from the Waikato district who commute quite far into Auckland would be smaller again.
      In the very long term, it’s possible that those numbers could change, but for now Auckland needs to focus on getting its public transport network working well for the 1.5 million people living in the city itself.

      1. Things have changed a lot since 2013. For starters there is that whole suburb that has been built at Pokeno which presumably most commute to AKL. AKL prices have gone through the roof and a lot more people are moving to the Waikato as a result. I wouldn’t be surprised at all if those numbers have doubled or tripled.

        1. Extending the network to at least Pokeno makes a lot of sense, pity it sits in a different juristiction. Really needs to govt, NZTA and Kiwirail to step in and say it’s worthwhile anyway. There must be some potential around the Waiuku branch as well.

        2. Pokeno Village Estate just isn’t that big though – that development will max out at 2,000 homes, and only 450 have been built so far. 2,000 homes is a reasonably sized suburb, comparably sized to what Todds are doing at Long Bay for example, but it’s not huge. Auckland currently has more than 500,000 homes, and needs to build another 400,000 in the next 30 years. That’s 200 new Pokenos.

          Many of the new Pokeno residents will be ex-Aucklanders and still commuting, but there will also be plenty who are retired or planning to retire in the coming years. And Pokeno sits within the Waikato District, and people commuting from there to Auckland will be going along NZTA-controlled motorways for the most part, which makes it not Auckland Council’s problem. It has enough to worry about already!

        3. However if there was a rail service to Pokeno then a lot of other Waikato residents might drive there then take the train rather than driving to Auckland themselves.

        4. I talked to ARTA years ago about extending the diesel service to Pokeno after they had electrified the rest and would have spare units. They just weren’t interested, their preference was pretend Pokeno wasn’t happening. So Pokeno will have 4000 people, land for a station but no rail.

        5. Pokeno is under Waikato regional rates and as the ‘A’ in AT stands for Auckland it is very unlikely ATs trains will go into the Waikato without funding from the Waikato ratepayers.

        6. Pokeno is within Auckland City Council Boundaries according to Google map. Waikato begins One Kilometre south of Mercer beyond the Waikato. So it ought to have some services.
          It is listed under the Franklin District Plan here: Proposed plan change 24 – Pokeno Structure Plan and new zoning provisions

          Plan change 24 is the first of the projects implementing Franklin’s District Growth Strategy (DGS) and gives effect to the council-adopted Pokeno Structure Plan.

        7. Wayne Moore google maps have it wrong, Auckland stops and Waikato begins just south of the current Pukekohe station. The people of Tuakau and Pokeno will back this up as they get their rates bill from Waikato not Auckland.

        8. and of course you can see that there are no improvements to the bus services either, the improvements coming in October down there are soley for Pukekohe and Waiuku as they are in-region. The 476 (Tuakau) and 50 (Port Waikato) buses do not see any improvement as they are in the Waikato Region, although I am not exactly sure how that works as these buses are operated by Waka Pacific aka NZBus and use AT HOP and the Zone C monthly pass includes these services (and the new “Franklin” simplified zone also includes Tuakau/Port Waikato), 476 is Auckland numbering (I asume setup before the super-city) and 50 is Waikato numbering. So not sure exactly who’s exactly paying for what here.

          I think ultimately something needs to be setup so that public transport can be funded properly for the area, I don’t think Waikato Council is ever really going to be too concerned about getting people to Auckland. I’ve always found it kind of strange how the region stretches as far north as Te Hana but cuts off at Bombay/Buckland in the south…

        9. Yes, it does appear Pokeno is in the Waikato Region it is mentioned in the Waikato Regional Council documents as such, I do wonder why it was included in the Franklin District Plan since it is outside the A.C.C. area.
          Viewing Waikato R.C documents it also appears there is a long-term strategy to have a commuter rail service (or an improved bus service) from not only Pokeno but from Hamilton to Auckland. Considerable pressure is put by North residents of the W.R.C area to have this.

        10. Yes I was confused by the Pokeno Auckland or Waikato thing too a while back. I’m sure I saw on the Paul Homes show when new Auckland was getting sorted that Pokeno was going to be part of it, I wonder if it was changed after submission or something. Think there is a “Welcome to Auckland” sign pretty much at the Bombay Hills somewhere on the motorway if you can go by that. I think from memory as a kid that it used to be part of the old Franklin County Council….but not sure on that.

        11. The southern part of the Franklin District was part of the Auckland Regional Council (and ARTA) and was originally going to be part of the Super City. However, the locals succeeded in removing themsleves from those plans and being put in the Waikato District and Regional Councils. Ironically there were a number from Pokeno in the paper last year complaining about how high their rates were as they had gone from being relatively low value land compared with the rest of Auckland, to having some of the highest values in the Waikato District, which includes towns such as Huntly and Ngaruawahia!

        12. This is false. When the Super-City was proposed to us, we tried, first, to keep all of Franklin out, but when they made it clear that we would not be allowed to opt out, we tried to make the southern boundary to be the river. They also disallowed that. Both we and Tuakau tried hard to get them to change but they said they original ARTA boundary was the watershed to the River and could not be changed. Thus we have friends who live between Pukekohe and Tuakau who ended up with part of their property in Auckland, part in Waikato.

          People here were pretty upset at it. Many surveys were conducted and percentages in favour either of keeping Franklin out entirely or else extending to the boundary to the River. We were told we had no choice.

          In the event I think I am pretty happy that we did end up in Auckland – but I think it would have been a lot better if they could have included farther south.


        13. John, I think we might be both slightly off the mark with this. It looks like the southern boundary was originally proposed to be the Waikato River and that this was fought by the Waikato Regional Council, amongst others to have it aligned along the catchment lines.

          Do you know whether Tuakau and Pokeno were previously part of the old ARC?

        14. You are probably right. I just remember when it was all happening, the local news (Franklin County News) had stuff about how it had been asked that the boundary be the River but someone – could well have been Waikato – objected.

          And, no, I don’t know if Tuakau or Pokeno were part of ARC. Sorry for sounding off about what I am actually vague about! The NZ Bus that serves us used to go to Tuakau and I suppose I thought this meant Tuakau was part of whatever authority it was that serviced Pukekohe – but maybe not.


        15. Looks like Tuakau was part of the old ARC, not sure about Pokeno though.

          Looks like I was (mostly) right about the boundary changes, but wrong on the history of how they actually got there. Didn’t realise they had planned to take the Super City boundary all the way to the Waikato though. There were a few issues with the Regional Council boundaries as they generally follow watersheds. The most famous being Whangamomona being put in Horizons Region as the river drains to the Whanganui but it’s really a Taranaki township, resulted in them declaring independance and electing a goat at the president.

        16. Asked my mum re Pokeno (we used to live there) part of old ARC, she thinks so, was definitely part of former Franklin County Council anyway, so would think it would of been.

        17. Grant to my knowledge Tuakau (so you would also assume Pokeno being that it is slightly further south) has always had Waikato regional council rates and not ARC rates even though they were part of the Franklin district council area. With the creation of the supercity the southern boundary of the old ARC became the cities southern boundary but it swallowed up the entire Rodney district in the north due to the residents opposing being split between Auckland and Northland.
          Either way Auckland and by that AT stop a few kms south of the current Pukekohe railway station so without input from Waikato AT will not send its trains past that point.

  10. Would love to see this strategy to extend electric trains from Swanson to the huge development area at Huapai.

      1. Given the acceleration and speeds are artificially limited anyway it probably won’t even be noticed. From memory the trains are about 150 tonnes, adding 4 tonnes isn’t all that much

        1. With the diesels we currently have two speed-limited sections, one in Paerata, one near the Glasshouse. Driver told me the track quality was some sort of issue there. For us Pukekohe-ites, it isn’t the speed that is the issue. We want to be able to travel straight through, not have to change trains, worry about making the connexion, leave on an earlier train to make sure of making the connexion, etc. I get on in Pukekohe and off in Britomart.


      2. A 3-car is 132 tonnes empty, 160+ full, so 4 tonnes is indeed no big deal. About the same as 50 of the 373 passengers it can carry. The control system detects the weight onboard and adjusts its suspension, power and braking to suit, so that the ride is fairly well standardised regardless of how full the train is. I’d agree with the suggestion that the effect of the battery weight would be negligible.

    1. The trains weight 142 tons empty, and carry another 20 to 30 tonnes of passengers. So a four ton battery is nothing, it would add perhaps 2% to the mass of the vehicle.

  11. So Auckland Transport is fart arsing around again with half baked solutions.

    The obvious solution is simple: electrify to Pukekohe NOW with the big Special Housing Areas already under way between Drury South and Pukekohe.
    The cost of $140m can be easily found by scaling back the $471m Mill Road expressway to something more humane rather than a motorway. Also given the development contributions coming in from the southern urban developments AND given electrification is a CAPEX project spread usually over 30 years (in terms of finance cost) the developments should be able to pay for the project in full of not a few times over.

    So hurry up Auckland Transport your excuses and shortsightedness is getting boring.

  12. Looks smart to me. May well be some first mover advantage with the supplier. Frees up capital to build those intermediate stations sooner to give certainty to the housing developments and to help them form in a proper village-like way around the new stations.

    This next phase of AKL’s spread must be Transit led. We can’t make our usual mistake and leave out Transit only to discover we need it later so expensively, like with AMETI.

    So long as the tech works, of course. Presumably a charging base at Puke, and then on pantograph from Papakura north.

    1. “Presumably a charging base at Puke, and then on pantograph from Papakura north”

      It is much simpler than that. Batteries are charged while the EMUs are under the wires (which constitutes the majority of the km travelled). The battery has sufficient capacity to do the Papakura-Pukekohe return. There may be a trickle charging facility at the stabling yard for maintaining peak charge overnight but the bulk of the charging will be done on the run. The quantity of sets confirms this.

      Based on published data I estimate that this is $20M vs $130M of expenditure.

      1. I comment on this below, but a section of overhead to recharge trains dwelling and laying over at Pukekohe could literally halve the required battery capacity, as they’d need only cover the one way distance. Probably stacks up economically.

  13. A local (Pukekohe) friend told me – based on what, I don’t know – that part of the cost of electrifying to Pukekohe was that the rails themselves needed to be re-lain in parts of the track. Don’t know if there is anything to this or not. Do the electric trains require superior railing??


    1. There have been rumours of Kiwirail requiring that the electrification project be simultaneous with the third-main between Papakura and Pukekohe – presumably because there is also a need for the motorway overbridges near Drury to be rebuilt to allow room for both electrification (height) and 3rd main (width). The track bed can’t be lowered as the water level of the nearby creek makes it a flood risk. The money to electrify is one thing, but the cost of these related works is really someone else’s problem, so it’s all going to be on hold until all three organisations are ready to proceed.

  14. There are at some point going to be stations between Pukekohe and Papakura, yes? Like for the SHA at Paerata North. So it’s not 36km of wires for one station, it’s 36km of wires to serve Pukekohe now and be ready for future stations. Just do it once and do it properly. Cancel a low-value roading project (I have some ideas here) to free up the money – win/win.

    Edit: do the batteries have enough oomph to cope with, say, stopping and reaccelerating five times (two stations between Papakura and Pukekohe, turnaround at Pukekohe) before getting a recharge? If not, battery trains will either hamper the development of the new stations, or we’ll be laying wires anyway.

    1. FWIW, 36Km is round-trip. It’s 18Km from Papakura Station to Pukekohe Station. My understanding is that they hope eventually to have stations at Paerata and Drury – and I don’t know about the planned 5000-house development for where Wesley College now is, supposed to start in about five years. Would they have a station there??


      1. Wesley College is right next to Paerata, so that subdivision will be the main catchment for the Paerata station.

    2. Would’ve been 36km of wires for one station if done when rest of network was done. Business case from a few years ago showed it was only worthwhile doing if also doing those other stations (and development)

      1. Fair enough, the new stations would have seemed a distant prospect when the project was in the planning stages. But any calculations now should include the probability that Drury and Paerata stations will need to be built in 5-10 years.

        Did the business case from a few years ago take into account the probable patronage increases from electrification?

    3. I would imagine that they would still string up some form over overhead at Pukekohe station to give them a boost for the time that they are there so they probably would be able to cope with this

      1. Yes this, a short section of overhead at Pukekohe station would allow flash charging of the batteries while the trains were sat there, it need only cover the station. With 25kv and a bazillion amps at your disposal you could do a lot of charging in five or ten minutes.

        1. Sure 20-30 mins for a full charge, however it it is anything like battery tech for buses (which I assume it is) you’d be able to get to 80% charge in a quarter of that time.

        2. Correct Nick. Its not a requirement to have a battery fully charged before departure if a BEMU can get back to the place with full OLE in time.

        3. “With 25kv and a bazillion amps at your disposal you could do a lot of charging in five or ten minutes.”…and severely diminish the cycle life of the battery and incur additional cost for the charge regulation circuitry and raise major heat dissipation issues. With a few exceptions batteries last a lot longer and exhibit better round-trip efficiency when charged and discharged at modest rates.

  15. Hmm, do we want to beta test this?

    I like that it would offer the possibility of electrics running to destinations some distance beyond overhead lines (how far? I wonder), but on the other hand we should be aiming to electrify to Te Rapa and develop inter-city passenger rail as other commenters have said.

  16. At some point we are going to electrify south to Hamilton, so we might as well get the wires up now rather than messing around with battery trains. They might only be 4 tonnes above the 150 tonnes of current train, but this will be added to their power usage as the accelerate from every station on the existing electrified section, which will add up over time. Might as well get on with electrification, it won’t go to waste.

    1. I have heard rumours here and there ( but seen no official confirmation) that Kiwirail is looking to get rid of there electric locco. If that happens I doubt if they will electrify the NI trunk from AKL to Hamilton.

      1. One of the features of our balkanised railway system is that each of the three existing electrifications is used by a different operator, for two distinct purposes, so there is no interest at all from any of the operators in linking the three, irrespective of what KR does with the NIMT electrics. Any extension southwards would also have the further complications of crossing regional boundaries and not being eligible for NLTF funding, neither of these being issues for NZTA’s roading projects.

        1. Doesn’t matter Mike, new loco’s can run on all 3 systems so could run from Auckland to Wellington provided the Auckland system was extended to Hamilton and the NIMT lines were extended down to where the Wellington network starts. In reality though the NIMT electric system would be upgraded to the Auckland standard just leaving the Wellington DC network as the odd one out.

        2. Of course joining up the three electrifications, with through running, is tehcnically possible – but that’s irrelevant until a capable body or bodies want to do it. With the current setup, that ain’t gonna happen because it’s not in any of their interests.

  17. I think the time keeping across the network will worsen if trains have to run all the way into Britomart from Pukekohe instead of from Papakura. The shuttle gives trains at Papakura more chance to catch up if something goes wrong. More turnaround time.

    1. I doubt the Pukekohe to Papakura section has a big impact on timekeeping as the trains are generally not full at that point so load and unload times would be reasonably consistent. Delays are much more likely to be caused by congested junctions such as Newmarket and Britomart and also unpredictable load/unload times at inner stations where the trains are more likely to be full. Either way not a good reason to hold off a development that has significant PT benefits for a growing area.

      1. However it wasn’t till the shuttle started and the Waitakere trains were abandoned that timekeeping came right. I know this coincided with all electric running but I am sure it has an effect.

        1. Unlikely. It seems, to southern line regulars, that the so called time improvements were simply achieved by switching from all trains scheduled for 55 Papakura to Britomart to having some 54 minutes and others 58 minutes. Be sceptical when you read those numbers, because some of those diesels pushed 50 minutes and the X:36 trains regularly go over an hour. Used to be that the X:49 worked at a pinch. Now my friends and I do not want to catch anything later than an X:36.

          In any case, I am sure that timing would actually be a bit simpler, because the trains that wait at Wiri seem to be Pukekohe connected ones. If they left a bit earlier, maybe this wouldn’t be an issue. More likely, I feel, the train would leave at the same time and if things are, for whatever reason, a bit slow getting to Papakura, they would unleash a bit between Papakura and Takanini and/or Manurewa and Homai and then we’d still wait at Wiri. Apparently, it’s got so that quite a few drivers don’t even bother trying to make better times because they know that any saving they make is just cancelled out by the Wiri wait.

          The point is that there are several reasons to expect this to be time neutral (or, at least, timetable neutral).

  18. What about fitting a diesel engine connected to a generator to make electricity and power the electric motors on the wheels from that… oh wait, that’s been done already.

    1. Nothing significant – the battery weight amounts to only about a 3% increase and would probably do more to reduce wheel-slip than anything negative.

  19. If central government pays for the electrification to Pukekohe then yes, I think put up the wires and do it properly.

    If AT has to pay, then this is a fairly good compromise from AT if the money is coming from their budget.

  20. As an interim idea it has merit if it allows a quick delivery of a single seat electric service to Pukekohe and gets the old DMUs off the run.

    The full electrification of Pukekohe (and beyond) is needed so the Battery EMU option and full electrification to allow EMUs to Pukekohe is needed.
    That is it is not A or B but A AND B.

    If it also frees up some regular EMUs to improve service pre-CRL great, but again this as a tactical solution.

    We will need more EMUs post-CRL opening – so lets not do the usual “once over very lightly [with bronze plate]” for the PT while we triple gold plate the road options at the same [for 10 or more times the money].

    Question, could this be used to deliver a heavy rail option to/from the airport via Otahuhu route [over a LRT service delivered by Onehunga], also as a tactical solution. Possibly as a re-use of the Battery EMUs once the Pukekohe line is electrified?

    Being mindful, that unless we use the [rail] corridors we have reserved they tend to get used for other uses over time, so the longer we delay the airport route protection and use the harder/more expensive it becomes to build.

    And even if that doesn’t come to much, the obvious option for second hand Battery EMUs is Christchurch to jumpstart their system into life.

  21. Rather than fitting a battery carriage, why not just attach a diesel engine at the end of the train? A hybrid option for now until the track is electrified? Crazy not to have a direct service from a populated township all the way to CBD.

    1. What happens if the bemu craps out away from the wires, would you need a diesel powered locomotive to drag it back under the wires?

      1. Same thing that happens if an EMU craps out, you get the next EMU to tow it. The batteries do add some redundancy if the overhead goes out.

        1. I guess the question would be would the battery powered EMU hold enough power to haul a dead EMU if it broke down near Pukekohe and had to be hauled all the way to Papakura?

        2. With Diesels being banned in Britomart, do they have enough power to push/pull a train through the Britomart to Mt Eden tunnel in the event of a power failure?

    2. In surveys the people of the Franklin districts expressed, amongst other things, a strong distaste for diesel fumes being a part of their rail commute experience. This could well be one of the reasons they are taking the BEMU route rather than a diesel-electric option. The other would be that BEMUs can be serviced at the existing Wiri EMU depot, whilst diesels could not.

  22. Great idea .Would it work for bringing trains to the shore and out west.It might just make the business case stronger for rail on the shore and through the tunnels

      1. National have had 8 years to get this right and currently control the purse strings, when they are no longer in government I will treat any equally stupid decisions from the new government with equal disdain.

    1. $1b to replace one loco type with another that makes no actual difference to the effectiveness of the freight network? Let’s not go down that road of financial recklessness. If a future government can come up with that kind of cash, spend it on things that will actually make a difference, not feel-good PR exercises.

        1. Around 75% of the GHG benefits and almost all of the safety benefits are gained by mode shifting from IC-hauled road to IC-hauled rail. Electrified rail without gaining market share is pissing money up the wall. Look up Pareto analysis.

        2. “If you consider not ruining the planet for our children to be PR then sure.”

          Wasting money that could be otherwise spent on measures that promote modal shift from road to rail, does nothing for the planet or children.

          And no, swapping locomotives does not promote modal shift. You may have noticed that the transport industry does not have rail electrification on its list of ways to promote increased rail usage. On the contrary, it would do the opposite, as the financial ramifications would be crippling.

    1. The stats on Electric Battery buses are here Zero emissions, 21.4 MPGe, $0.19 per mile.
      Proterra was the best results I saw up to 24hrs on a single charge, up to 65 mph, on a distant of 650 miles batteries housed in the back, no overheating as one gets from diesel motors making one to take breaks to cool the engine down. I am sure a rail conversion could be done with increase frequency. Frequency from Pukekohe to Papakura could be feed from electric battery buses from towns with an Auckland impact such as Thames, Waiuku, Glenbrook,Port Waikato, Matamata, Paeroa, Kopu, Miranda, Bombay – most of the main roads/highways from these centres are getting congested. Prices for Housing in these centres are increasing showing people are moving there without the travel time being a factor. If we could improve the train commute times and get the price right. I am sure this will convince these people to use it.
      See this tearsheet: this particular model is charged every 62 miles for 10-12 minutes uses an Advanced Lithium Titanate fast-charge energy storage system with a 220 kW peak permanent magnet drive motor and uses 1.7 kWh/mile (w/ full passenger load and HVAC) up to 40 pax. Gross weight is 39,050 lbs.
      I’m guessing a commute train conversion would be similar and with an increase battery storage area could do the same.

  23. I’m amazed how many people here support this. Be under no illusion, once this is done, putting up the wires will go to the bottom of the priority queue (and that is a big queue: airport rail, AWHC etc, etc). It also starts to dilute the benefits of having a standardised fleet and as others have pointed out means you are paying to accelerated 4T of weight up to speed between every station (85% of which have wires over them already!)

    Also AT will only specify the system to just get the trains to Pukekohe. There will be zero chance of them having range to go any further.

    It is of benefit to NZ Inc to wire up between Papakura and Hamilton and also to provide commuting options into Auckland from outside Auckland council areas. Getting the wires to Pukekohe is the first step of this. If we had a governement who understood transport needs, this would have been done by now.

    1. I agree. We in Pukekohe would much rather see them put in the wires. But if they aren’t going to do it, then – sigh! – I think we view this as better than nothing – which is what we have had so far. But I definitely agree. I think it is short-term thinking.

    2. Agree the best option would be to wire it up however if the funding isn’t there from central government then AT can get better bang for it’s buck in other ways (also note that ordering these battery EMU’s means that more of the existing fleet can be upgraded to 6-car EMUs providing extra overall network capacity – so the cost is minimal). They can have quick charge at Pukekohe (and if wanting to go further then putting in some short overhead wires at each station wouldn’t cost much and can top those batteries up).
      Also if you haven’t read already the batteries would be light in comparison to the trains and the EMUs are artificially restricted in acceleration already. This means that there wouldn’t be any reduction in acceleration under battery power.

  24. Electric battery can extend more quickly than hard wire and less likely to have unsightly wires and unlikely to see potential fatalities, the only additional is the overhead charging unit and the new station platforms they insist on having. And, what with Auckland Housing needs this could work in nicely at solving part of this problem by developing towns further out. There no need to rebuild bridges too low for the power cables or protective barriers.

  25. I would have thought batteries could be towed in a separate tender, rather than permanently mounting under the carriages. There is already a lot of equipment under floor, and not a lot of extra space where batteries could be located.
    I’d solve the tunnel height problem for services to Haupai assuming the length of carriages isn’t a problem?
    The 175km electrified for the current Auckland network cost $500M as far as I remember. So 18km from Papakura to Pukekohe shouldn’t cost more than $50M, especially considering there aren’t a whole lot of low bridges or other obstructions.

  26. Have they also looked at maintenance cost?

    How does the cost of maintaining electric cables compare to battery depreciation and replacement cost?

  27. There’s an elephant in the room that nobody has noticed.

    In 2014 at a resource consent hearing related to another matter, KiwiRail explained that it would not permit the current congestion issue north of Papakura to be expanded south, and that any move toward increasing service frequencies in conjunction with electrification would require the electrification project to include construction of a third main.

    I do not know the cost of that, but when you consider the earthworks and signals required, I would suggest the $140m price tag has at least doubled. For the number of passengers on offer, I doubt a business case would stand up.

    If they can come up with a world first and run battery powered AM class over such a long distance, that would be great, but otherwise the obvious solution is to have a diesel engine in the next batch of EMU’s, that can switch in for the run south. That’s existing, and proven technology.

  28. Surely any significant increase in service frequencies would require the third main to be built, regardless whether it’s battery, overhead, diesel or even steam-powered!

    1. Indeed, but what I mean is that a business case for electrification will aim to increase service frequency as part of the justification for the project. If the goal is to do away with the passenger transfer at Papakura, without service frequency being part of the equation, then the other options are more viable.

  29. Like the boring woman I am because & because I am undecided its time for a Pros & Cons List

    Pros: Cheaper Short Run Capex for AT, buys more Trains which is completely necessary anyway pre CRL, CAF sounds like they want a pilot project guinea pigging is cheap, allows after full elect or in just general future to trial them past Swanson, BEMU is more kiwi after all an EMU is an Australian bird & BEMU sounds like Be Moo which is fitting with our glorious dairy heritage.

    Con: Not as environmentally friendly as overhead traction or as efficient no regenerative breaking, tested but not completely proven tech we don’t want like those hoverboard things for our trains to blow up, might not allow for extra stops???, doesn’t help us on trying to get elec all the way to the Hamiltron.

    Queries: Can the batteries handle the two extra stops proposed in the full elect project, do they have the power for 6 car sets (assume yes but still questions), is this the absolute battery limit or could they go further like a Waikuku/Glenbrook station (Would put it on Mission Bush line to keep the historical train people happy) or Pokeno.

    1. I would expect that regenerative braking would be there, to help keep the batteries charged.

  30. Well there is regenerative braking the braking power is stored in the battery instead of being fed back into the wire.. I suppose to go further they just require bigger batteries or quick recharging at the terminal station which has being discussed above.

  31. 35km is Waikanae to Levin one way.

    This could be useful technology for Wellington Kapiti line extension / shuttles too, Levin – Paekakariki / Porirua / Wellington

  32. I am currently in India reviewing the Electric Vehicle manufacturing process.

    A 4 tonne battery pack to move a 155 tonne fully loaded 3 car EMU 36km gives a battery weight ratio of 38.75(4/155). That would be cutting edge indeed!!!!

    Current EVs with a range of 120km have battery weight ratios in the region of 2-8. It would be great to get details on the battery chemistry, controllers and the battery management units being used for the prototype to obtain such a high ratio?

    1. You are not making an apple to apple comparison Mr. Greening. Trains take a whole lot less power to keep them moving than cars due to smooth steel-on-steel contact of minimal area. In addition gradients and changes of direction are much less severe and, in the case of Auckland’s EMUs, the frontal area per unit mass is much less than a typical electric car.

      Diesel-hauled rail freight exhibits around 4 times the fuel efficiency of diesel-hauled road freight in spite of a higher tare to gross mass ratio than road haulage. On that basis and taking into account the lower drag per unit mass for EMU vs road EV a 4 tonne battery seems like state-of-the-art but not wildly ambitious.

      1. It may be fine as long as acceleration is in the 1-2 minute region to get over 50km/hr. If there are stops, gradients, corners or sections that require deceleration/acceleration then passengers will be in for a very loooooooooooooooooooooooooooooong ride.

        1. “It may be fine as long as acceleration is in the 1-2 minute region to get over 50km/hr.”

          Why? Acceleration rate is proportional to power. For a given battery technology mass is proportional to energy not power. To accelerate a mass to a given velocity takes a certain amount of energy irrespective of the rate of acceleration (resistive losses notwithstanding). Let’s see your calculations please.

      2. Also fuel efficiency between Diesel-hauled rail freight and Diesel-hauled road freight is a function of stops, gradients, corners and sections that require deceleration/acceleration energy. You put 1km of road parallel to 1km of rail and the energy required for Diesel-hauled rail and Diesel-hauled road over that 1km at the same/proportional mass, acceleration and velocity will be similar.

        1. You seem to forget the vastly different rolling resistances of rubber on road and steel on rail. Get a road vehicle moving on level ground and it will slow to a stop fairly quickly, while a rail vehicle will just keep on rolling (as many shunters know to their cost).

        2. Right, so your statement “You put 1km of road parallel to 1km of rail and the energy required for Diesel-hauled rail and Diesel-hauled road over that 1km at the same/proportional mass, acceleration and velocity will be similar” is at best incomplete, at worst inaccurate.

        3. In the real world it wouldn’t matter if they could get trucks as efficient as rail, trucks have to negotiate a system completely different to rail. with rail you can allocate a schedule that needs no stops between destinations, not so with roads where they share the space with us all.

        4. TG: I’m not sure how relevant it is that in 2009 state-of-the-trucks even with all the fuel-saving bells and whistles were significantly worse than standard trains in terms of fuel efficiency. It certainly does nothing to support your contention that today’s diesel trucks and diesel trains have near-identical performance over just one kilometre, even if that scenario was relevant to this particular discussion.

        5. There’s no way that the relationship between the open-ended 350+ and the approximate 400-450 can be calculated to such a spurious level of accuracy as 87.5%, not to mention the bias in using the bottom-end rail figure and the best-achieved road figure in your calculation.

          Leaving that aside, if you’ve no better evidence than a 7-year-old very approximate comparison between a hypothetical state-of-the art truck and a normal train, I suggest that you stop digging this particular hole.

        6. 350/400 x 100 = 87.5%, I am merely pointing out that it is possible for similar energy performance between truck and train freight, the study backs this.

        7. I certainly can’t fault your arithmetic, but your conclusion is valid if and only if your implicit assumptions that the following are OK are also valid:

          a) comparing the best truck figure with the worst train figure;
          b) relying on just one 7-year-old report as the only evidence;
          c) comparing a hypothetical state-of-the art truck with a real train; and
          d) expressing the difference between a very approximate figure and a range of figures as a precise percentage.

          It would be good if you were able to respond to these points: just repeating the same thing doesn’t help explain your position – nor make it any more valid!

        8. Yes a,b,c,d backs my original statement about truck vs train freight fuel efficiency over 1km.

          “Also fuel efficiency between Diesel-hauled rail freight and Diesel-hauled road freight is a function of stops, gradients, corners and sections that require deceleration/acceleration energy. You put 1km of road parallel to 1km of rail and the energy required for Diesel-hauled rail and Diesel-hauled road over that 1km at the same/proportional mass, acceleration and velocity will be similar.”

          You implied that rolling resistance had a much bigger part to play in the fuel efficiency between the two. Where as stops, gradients, corners and sections that require deceleration/acceleration is the main difference between truck and train fuel efficiency.

        9. It’s hardly surprising that your assumptions back up your original statements, is it?

          And if all you can do is endlessly repeated your origiinal post, based on distinctly questionable and unsupported assumptions, there’s no point in continuing this discussion.

          And please don’t misrepresent me – I never implied that rolling resistance had a “bigger” (than what?) part to play, rather that you initially ignored it and now (on the basis of what evidence?) discount it, instead parroting your original unsupported assertion – which gets us precisely nowhere.

          And I’ve never yet seen a “corner” on a railway!

        10. Christ Toa, can’t you just make spurious, unsupported arguments about narrow cars?

        11. “You put 1km of road parallel to 1km of rail”

          Except the rail line isn’t parallel to a road. It is on a high speed, low gradient section of track that will have at most three stops, full regenerative braking, and no congestion delay.

    1. I think there is a gap like that somewhere between Westfield and Penrose – at least the air-conditioning stops during that time and the train slows down. I have always assumed that was because of a bridge there that doesn’t have the electric lines continuous. If that is so, couldn’t they just do that for the Drury bridge? Just let the train coast or go on minimal battery power for a short distance?


      1. Coasting through a neutral zone is very different from going under a bridge that is too low for the OLE (over heads), the train can’t lower it paragraph and continue under the bridge.

        1. Trains could lower the pantograph (s) – it happens probably thousands of times a day in New York and London when trains swap between overhead and third-rail electrification. But doing it for large-scale legacy reasons is not the same as doing it for one-off obstructions: the British looked seriously at discontinuous electrification not so long ago but are not using it in any of their current schemes, so they don’t appear to think that it’s a goer.

    2. Why not just dig down a little and lower the whole train line by a metre 100m each side of the bridge? Sure cheaper than replacing a whole bridge/s and less disruptive?! Not that much of a gradient issue either trains will be fine.

        1. So put drainage in and a barrier to stop water from nearby waterways entering it… build a levy with the spoil from the dig!

  33. There are two good possibilities

    1. Nick R = June 9, 2016 at 2:06 pm, Use super capacitors instead then.

    This is sense. Batteries can store a substantial amount of energy, but are very poor when it comes to fast charging and discharge. Supercaps are not so good at storing large amounts but are very good at fast charging and discharge. So any “battery” trains should have both. The supercap releases its energy to accelerate the train, the battery provides the minimal energy needed to keep it running, then the supercap absorbs the energy when the train brakes. The control system balances the two to ensure that the supercap always has sufficient energy for acceleration and sufficient capacity for absorption of braking energy, while the battery has sufficient energy for station to station running.

    This cuts the battery mass down somewhat, at the expense of an increase in mass of supercap, and, as the battery never has to charge or discharge fast, extends battery life from years to decades (hopefully). Meanwhile the supercap has, so far as currently known, a life measured in centuries.

    2. Look at the timetable. A train departs Pukekohe at xx14, xx35 and xx 54 in the morning rush hour, approx every 20 minutes. A train departs Papakura at every 8, 10 or 12 minutes in the am rush hour. Thus the 0714 ex PUK arrives in PAP at 0730, and the passengers then sit around till the electric departs at 0736, which arrives at Britomart at 0832 – a tad over an hour and a quarter. A through electric train, assuming the 16 minutes PUK to PAP could be cut to 14 minutes, could depart PUK at 0722 – a valuable 8 minute saving to passengers from Pukekohe. At present the 0558 ex Britomart arrives at PAP at 0650 and passengers (if any) are decanted to wait for the 0707 to PUK, arriving 0723. It is obvious that the two trains on the PUK to PAP section cross somewhere. It appears that whether overhead or battery+supercap propulsion is used, the signalling will have to be redone to come with the train’s propulsion. So perhaps it might be reasonable to electrify only one track, with a loop at the timetabled crossing point. With both tracks resignalled for services in each direction (‘banalisation’) the effective cost of operating a through service every 20 minutes – half that of the trains operating from Papakura in – would be near halved. Perhaps $40M (assuming that no new stations are built at the moment) as compared to “AT have estimated it at over $100 million”. This $60M saving should not be sneezed at. And if new stations are needed in the future, then electrifying both tracks would be feasible.

    Incidentally, if electrification is done, there is no real necessity for a substation at Pukekohe – a bank of batteries plus supercaps would be able to provide for the power needed at the outer end, fed as needed by the overhead between trains. This arrangement is being widely considered as a means of reducing peak demand on the power supply – often charged at a very high rate compared to the average energy demand rate.

    1. Thanks for number 2 above. The time delays are even more than that sounds, particularly in the evening, when many of us catch an earlier train to Papakura, even though it means a 17-minute wait for the Pukekohe train. This is because we are afraid of the ‘proper’ train being delayed and missing the Pukekohe train – so another twenty-minute wait. This doesn’t happen often – but does happen. It would be so great for us if we didn’t have to worry about that connexion.


      1. The proper train as you but it or ‘connecting service’ as it is often refereed to, with the exception of a substantial delay the shuttle will wait for its connecting service before departing.

        1. Maybe they are supposed to wait but they don’t. After the second time that I took the connecting service, only for it to arrive about one minute before the Pukekohe train’s departure – run toward the Pukekohe train only to see the Pukekohe train leave down the track – after that, I now take the earlier train and put up with the delay. My boss is understanding about my leaving work ten minutes early. The connecting train (both morning and evening) appears only to connect if the incoming is on time. One of the train managers told me they get chewed out if they leave even a minute or two late.


        2. That attitude to timetabled connections doesn’t augur well for the new interconnected bus/train network…

  34. I heard they are going to use 260,000 AAA batteries wired together. Let’s hope they dont get any in backwards.

  35. OK I will be serious about this now. Wouldn’t it be better to use a diesel electric unit for the Pukekohe services and just use a battery to get them around the the central rail link? I mean that is a much shorter distance than Papakura to Pukekohe.

    1. “Wouldn’t it be better to use a diesel electric unit for the Pukekohe services”

      No Mr. mfwic. The proposed battery electrics are a standard EMU with the addition of a battery and some power electronics. On the assumption that they have the batteries fitted to the outer (power) cars they will have performance identical to the other EMUs on the system. With the exception of the above items they will utilise the same spares, the same maintenance procedures and facilities, they will have zero emissions, they will be quieter and cleaner than a diesel-electric, will not require refuelling facilities.

      In contrast a diesel-electric unit would require a total prime-mover output of around 1.8 MW to 2.0 MW to give an acceleration equivalent to the current EMUs (important if the are to co-exist). As a single unit this would typically be a large turbocharged V12 unit with alternator attached that would take up the height of the train and a large portion of the length of one car by the time the cooling system, exhaust silencer, dynamic brakes etc are accommodated . More likely would be something like 4 underfloor diesels of around 500 kW. These, however, are typically fitted with hydraulic transmissions rather than electric; a diesel-electric version with underfloor engines would be non-standard (and hence expensive). Diesels are more maintenance-hungry than electrics.

      From an engineering standpoint adding batteries to thestandard EMU design is elegant; it utilises a lot of existing hardware and retains most of the standardisation benefits of the existing fleet.

        1. It is precisely changing in Papakura (or anywhere) that we don’t want to do. It is inefficient (for the customer) and productive of anxiety.

          I was happier about my travel when we went all the way through from Pukekohe to Britomart and return on the same diesel train. I don’t care what drives the train; I don’t want to change trains.


        2. You change again to go to Manukau, Panmure or anywhere out west, people get over it. How would you like going from Onehunga to Sylvia park with two changes?

        3. An important factor in AT’s new network is the increased requirement for passengers to change between services: why should Pukekohe passengers be different?

        4. This transfer involves neither a change of mode nor a change of route. Is there anywhere else on AT’s network where this occurs? The passengers concerned want to continue along the same rail line in the same direction on the same train. Seems a perfectly reasonable desire..

        5. Changing from one vehicle to another isn’t a good in its own right, it’s only good if it gives you new opportunities directly, or if it makes the operations a lot more efficient which allows resources to be spend on those new opportunities.

          There is a difference between a connection (that opens up new destinations) and a forced transfer (which doesn’t). The Pukekohe shuttle creates a forced transfer, you simply have to do it to go anywhere at all (well except Papakura itself). It’s also not very efficient, certainly keeping what amounts to a whole separate diesel train line running between two stations suburban stations cannot be cheap.

          From Pukekohe many folks will have already make one connection to get from their neighbourhood to the train already, and they might need to make another to get to somewhere not on the southern line also. So adding in what could amount to a third change of vehicles should be avoided.

          Onehunga to Sylvia Park is a problem with two transfers, which is why AT have a frequent service bus planned to make the direct connection between the two.

        6. Ok Nick R what about Onehunga to Manukau, Orakei, Glen Innes etc? Having all line terminate/start at Britomart means if you are not going to Britomart there is a chance you will need to change trains to get where you are going. Having a dedicated service for Pukekohe means they don’t get shafted and put on buses due to delays on other parts of the network and the need to get the train back to Britomart.

        7. So connect a diesel engine onto an EMU when it gets to Papakura. It would increase op costs but still be a hell of a lot cheaper than electrifying all the way to Papakura. Hell you could even pay every existing rail passenger $1/2million to stop travelling and it would be cheaper.

        8. mfwic: dragging of EMUs by diesels used to happen between Paekakariki and Paraparaumu, and Wellington-Featherston for major events at Martinborough. Nothing new under the sun!

          But if they’re being dragged out into the sticks (no offence intended) they’re not available for what they’ve been bought for, thus reducing capacity elsewhere, and they make very expensive and over-equipped loco-hauled carriages. Doesn’t really make a lot of sense (assuming that you were being serious, doubtful given your last sentence – but you never know!).

    2. I’m struggling to see the benefit of what you are proposing here as it would still involve aquiring a new set of units just they would not use a more expensive fuel as they ran across the rest of the network that is electrified.

  36. Allow me to quote directly from my earlier post:

    “a diesel-electric version with underfloor engines would be non-standard”

    As for the Wikipedia reference;

    Hybrid Demu (with underfloor engines) KiHa E200. Quantity built 3 (
    HB-E210 series (with underfloor engines) Quantity built 8 (
    HB-E300 series (with underfloor engines) Quantity built 4 (

    Not sounding like a mass market there…and locomotive hauled stock (since some of the examples in the Wikipedia article are locomotives)? Severely limited in acceleration. The maximum tractive effort that can be exerted is the product of mass of the locomotive and the wheel-rail coefficient of friction (typically 0.25). The mass to be accelerated, however, is the mass of the locomotive plus the mass of the train. Adhesion (or lack thereof) not power is the limiting factor in accelerating a locomotive-hauled train. A 3 car Auckland EMU has 8 axles out of 12 powered. A 3 car diesel-loco-hauled equivalent would have 4 axles out of 16 powered and weigh around 50% more. No possibility of coming anywhere near the acceleration rate of the EMUs.

    From an engineering standpoint adding batteries to thestandard EMU design is elegant.

Leave a Reply

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