Last week at the Council’s Transport and Infrastructure Committee, Kiwirail gave an update about the state of the network and the work they’re doing to get it ready for the opening of the City Rail Link. There were a few aspects that stood out to me, so I’ve pulled them together in this post. You can watch the whole thing here – the presentation covers the first 25 minutes, followed by questions from Councillors.
Britomart Points Failures
The rail network this year has been plagued by a wide variety of faults and other failures. One area that has popped up a lot is points failures in Britomart, including one on the morning of this presentation. This is particularly notable because the points machines in Britomart were all replaced over the Christmas shutdown as part of the work to realign the tracks in preparation for the CRL.
KiwiRail said that following forensic examination by the manufacturer, they’ve finally found the cause of the issue to be a manufacturing defect with the frequency converter inside the points machines. Furthermore, they said there are 16 of these faulty frequency converters already installed around the network – six of which are inside Britomart – with more going into the CRL and some in Wellington. They’ll do what they can to proactively manage them, but ultimately, they all need to be replaced. Hopefully this is something that can be done overnight and will not need a full network shutdown.
KiwiRail also told Councillors that as part of investigating this issue they’ve had an international expert helping audit the issue. They’re going to extend that audit to a “complete network signalling/control system interface review” to make sure that everything is fit for purpose to run trains when the CRL opens.
CRL Commissioning
Kiwirail noted that the Matariki weekend network closure saw both ends of the CRL connected up to the signalling system and they’re now working on commissioning the signals within the tunnels.
They’re also starting to liven up the traction power systems, and say they’re working to a key date in August where the entire CRL will be livened.
This work is all leading up to the first test train running through the tunnels, which is currently planned for November. There will then be a further six to nine months of testing needed – and we know from CRL that then will also be significant driver training required.
Somewhat related: the Matariki Weekend also saw the electrification of Papakura to Pukekohe, livened up for the first-time. This means electrification of that section is now largely complete – although, as we’ve covered before, KiwiRail are keeping the lines closed for the rebuild works, with services expected again early next year.
Fixing the Network
Despite all of the work done so far on fixing the network, KiwiRail were also very keen to highlight just how much more work is still needed to get the network ready for the CRL. This is work that needs to happen over the next year or so.
Fixing the tracks south of Papakura is just one of the tasks of the Rail Network Rebuild, which is fixing/replacing things like the foundations under the tracks, improving drainage etc. As we’d previously learnt, Kiwirail had already used up all the funding they had on the Inner South, East and West sections meaning there was a lot of uncertainty over the outer sections of the Western and Southern lines.
One of the pleasant surprises from the Budget this year was that the government will provide an additional $159 million towards completing the rebuild work, which they say has to be completed by February 2026.
The government also provided $107 million to be spent over the next 12 months to help catch up on other overdue renewals in Auckland and Wellington. KiwiRail say they’re just finalising the exact share of that for Auckland – likely to be around $50 million – along with a work plan to spend it in order to replace these expired or expiring assets.
Finally, the charges that Auckland Transport pay to KiwiRail to access and help maintain the rail network have been increasing substantially over the last few years. Not that long ago, it was around $25 million a year, which is apparently far below what it actually costs to maintain the network. I understand that last year it was around $55 million, but this year it has risen to around $92 million.
All up, that means over the next year or so, hundreds of millions will be spent to fix and maintain the rail network. KiwiRail say that, on top of the typical Christmas and long weekend closures, they’ll also need ‘extended access‘ – which means more long closures of the network are needed.
Kiwirail say they’ll likely share those future closure plans towards the end of July.
Future Funding
Concerningly, while Kiwirail have a large amount of funding confirmed right now for maintenance, they also say they do not have assured funding for FY26 onwards. They’re working with AT and One Rail (who run the trains) on a business case to secure the maintenance and renewals funding needed to sustain the network after the CRL opens.
Rail Screeching
There was a bit of noise in the news earlier this year about the level of screeching on the tracks, especially from residents surrounding Hobson Bay. KiwiRail told Councillors they’ve first had to wait for rail grinding work, as part of fixing the tracks, to be completed. Most of the network is already completed and the final sections are due to finish around September. Some of the noise does settle in the months following grinding, but starting towards the end of the year they’ll be installing lubrication pots in key areas to further help reduce noise.
CRL Day One
Almost all of the work referenced by KiwiRail was in support of getting the network ready for the CRL opening. They reiterated quite a lot that this meant getting the network to be able to run a maximum 16 trains per hour (per direction) through the tunnels.
They showed off this map to highlight just what that would look like, though this also seems like a case of burying reality under a lot of detail.
We’ve seen the basic concept behind this plan before. Just like then, I’m not convinced by it. Here are some of the reasons why, along with a few other observations:
- Poor off-peak service – Most of the lines you see on the map above are peak time, peak direction only services. Once the peak is over, services will only run every 15 minutes on the main lines. While that’s an improvement on the current 3 trains per hour, I don’t consider a 15-minute wait good enough for our ‘Rapid Transit Network’. In my opinion, services need to be at least every 10 minutes off-peak to qualify as rapid transit.
The peak time, peak direction focus also means that effectively people travelling counter-peak (for example, from New Lynn to Henderson in the AM peak), will see a reduction in service from the current 6tph (trains per hour) at peak times.
Back in 2013, Auckland Transport was promising services every 10 minutes. AT needs to do better to support the massive transformation that the CRL promises to be.Planned rail frequencies in the 2013 Regional Public Transport Plan - Confusing services – With so many different service patterns, including terminating services at numerous locations around the network, we’re concerned this will result in reliability issues and a great deal of confusion for users. It screams of the focus being about optimising to a model, and not serving customers. Especially with the southern line doubling back on itself – making for a hot mess for users at Otāhuhu in particular, where trains branded as the same line go to four different destinations.
- Henderson Upgrade Redundant? – It’s notable that both lines to the west go to Swanson. What does this mean for all the planned upgrades to Henderson Station? We’ve long advocated for Swanson to be upgraded instead.
- Onehunga Line short run – The questions around Henderson get even stronger now the suggestion is that the Onehunga Line will terminate at Maungawhau (previously they were to go to Henderson too). To me, the change makes sense by freeing up train slots on the western line to run the cure service at 6tph all day. Especially as freight services on NAL in the absence of the Marsden Link to North Port are unlikely to be at the volume KR seem to be reserving slots for?
The KiwiRail team stressed to council several times the idea that “this is a freight railway with metro services on top”.
This is historically accurate. Prior to re-investment this century – starting with Project Dart, double-tracking the western line, etc – passenger services were reduced to almost nothing, and KiwiRail freight did largely have the network to itself.
However, with Dart, electrification, CRL, the 3rd main, RNIP, RCF track replacement, etc, this is now clearly a mixed freight and passenger system. This mix of course makes for competing demands on the network and its operation. KiwiRail, as a logistics State-Owned Enterprise, has a strong need to prioritise its freight task. Looking at this proposed metro services pattern, that point of view seems fairly evident.
So what might a more legible and customer-focused service look like? Taking the core aim of every Metro system, from the point of view of the people using it (and therefore, this should be AT’s point of view also), the holy grail is a “turn-up-and-go” service, sufficiently reliable and legible, with clean and safe trains and stations. In our view this means a minimum 6tph all day, with a simple pattern that’s as regular as possible and doesn’t require ordinary people to have a deep knowledge of the network. A user-focused design.
In our view, there are a number of ways this could be delivered, working within the constraints of the freight and intercity services. The pattern below is just one example. Mostly we’d like to see AT take the lead on designing a more ideal passenger service – one that delivers on its promise of “putting people and places at the heart of how we design and deliver our transport system” – rather than settling for something designed for a “freight railway with metro services on top.”
Our CRL Day One Model
Through-routing enabled by the CRL enables a basic two-line pattern, with peak overlays as necessary. Here, our assumption is there’ll be more freight traffic to Ports of Auckland than to Northland, limiting the eastern line to 4tph off peak. So, we’ve terminated the red line at Otahuhu in order to remove the confusing double-back – but this also allows for increased capacity over lower frequency, as inbound (citybound) trains will be starting empty at Otahuhu.
The Onehunga Line gets a passing loop midway and a 50% increase in frequency, though at the cost of a forced transfer at Penrose. Work would be needed to align those platforms better, in order to improve this transfer for passengers.
This change also removes a major conflict at that junction which will increase reliability, safety, and capacity on the whole system. It also enables 6tph on the western line core pattern, which we suggest taking right through to Manukau City, taking advantage of the 3rd main to Wiri which separates freight and metro services there – to the delight of KiwiRail and Auckland Transport both!
We haven’t shown any express services from the deep south. These are possible in peak pattern, though they do take additional space/time to work through the pattern. The Red Line in the south will be able to run more frequent trains as that market grows, but for now this looks like plenty of service.
Keen readers will note the old Mt Eden platforms at Maungawhau are unused by Metro services in this pattern. Well, perhaps Te Huia could terminate there? A nice new station, very well connected to the rest of the network.
We acknowledge Auckland is an under-tracked complex little system. This is all the more reason to tackle the problem of its operation with imagination and ambition. Especially as getting it to work to its very best as a Metro passenger service is the most effective way to ensure future investment, which is vital to getting to that nirvana of separated freight and passenger networks.
No doubt our readers will have other thoughts – please weigh-in in the comments below!