HS2 loss of ATO - what are the ramifications?

[Crithylum]

Alot has been said about the top speed reduction, however there is very little about the ATO. Assuming the train spends the entire distance between OOC and Birmingham International at top speed (which is entirely unrealistic) the top speed reduction will cost ~180s (depending on exactly how the distance is measured). ~120s is likely more realistic.

For the deceleration time calculations below, I am comparing to if it were to proceed at the 320 km/h top speed and instantaneously stop at the station. I have also assumed deceleration at a constant rate.

Aggressive ATO (1 m/s2) - 44s
Moderate ATO (0.8 m/s2) - 56s
Gentle ATO / aggressive driver (0.6 m/s2) - 74s
Moderate driving (0.45 m/s2) - 99s
Gentle driving (0.3 m/s2) - 148s

Given that they will need to do the timings for the slowest driver, this will possibly add up to 2 minutes per stop. The time penalty from a lower line speed is less (as you spend less time braking), but I wouldn't be surprised if this added 4 minutes on a Curzon Street to Euston journey in total.

This begs the question, how much are they really saving and have they calculated the costs? Does it even cost that much to implement ATO on top of ETCS?

Finally, I thought ATO was required so that the doors would align with the emergency evacuation shafts? Or is this not required now that the brakes will produce less heat due to the lower top speed?

 

[Nym]

No impact at all.
It was never going to be ATO other than in the musings of optimisim of a political industry “expert” trying to sell their products.

Nowhere in the world has ATO High Speed, and there’s nothing preventing a competent driver on clear track having the rules available to accelerate at the limit of adhesion.
Indeed in Protected Manual on 2009TS 1.14m/s/s is available.

 

[Harpo]

Nowhere in the world has ATO High Speed

Sure? I have had a live demonstration of ATO over a long distance in the cab (official visit) of a high speed passenger service from Madrid to Zaragoza. Interestingly the driver chose to switch it off as it was gaining too much time. Also of note was the driver receiving and taking inward (cab) telephone calls on the move.

 

[357]

Also of note was the driver receiving and taking inward (cab) telephone calls on the move.

I can't think of a TOC in this country that completely prohibits it. A couple have bizarre rules about GSMR usage but even they permit calls between the guard and driver.

and there’s nothing preventing a competent driver on clear track having the rules available to accelerate at the limit of adhesion.
Indeed in Protected Manual on 2009TS 1.14m/s/s is available.

Exactly this.

Aggressive ATO (1 m/s2) - 44s
Moderate ATO (0.8 m/s2) - 56s
Gentle ATO / aggressive driver (0.6 m/s2) - 74s
Moderate driving (0.45 m/s2) - 99s
Gentle driving (0.3 m/s2) - 148s

There shouldn't and doesn't need to be a large difference between drivers. If delays are caused leaving stations, it is a training and policy issue.

When stopping at stations, there is more of a natural variation - but again this can be managed by training and increasing confidence. I know a number of drivers who never use more than the initial brake setting, and others who are very reluctant to use full service or emergency even when the situation clearly requires it out of fear of repercussions from management.

 

[Crithylum]

I never said anything about acceleration. I have assumed no difference in acceleration.

My experience on the Elizabeth line west of Paddington is that the T4 stoppers almost always pick up small delays, and that these are worst with unconfident drivers.

 

[zwk500]

This begs the question, how much are they really saving and have they calculated the costs? Does it even cost that much to implement ATO on top of ETCS?

Yes. ETCS on it's own is "only" required to be able to stop the train getting too fast and to bring it to a stand if necessary. That is considerably easier to mitigate failure risks than a system that must also be capable of speeding the train up.

 

[HSTEd]

Nowhere in the world has ATO High Speed, and there’s nothing preventing a competent driver on clear track having the rules available to accelerate at the limit of adhesion.

China has GoA 3 automation on a high speed railway.....

EDIT: The Beijing–Zhangjiakou high-speed railway is fitted with driverless (but person on board) automatic train operation and opened in time for the 2022 Winter Olympics.

== Doublepost prevention - post automatically merged: 20 May 2026 ==

Yes. ETCS on it's own is "only" required to be able to stop the train getting too fast and to bring it to a stand if necessary. That is considerably easier to mitigate failure risks than a system that must also be capable of speeding the train up.

On the other hand, it doesn't have to achieve the same level of reliability in speeding the train up as the ETCS system has to be able to achieve in slowing the train down....

Forgive me if I am skeptical that deleting ATO will save a meaningful sum of money compared to the cost of the scheme, and once it is gone we will likely never get it back.

 

[zwk500]

Forgive me if I am skeptical that deleting ATO will save a meaningful sum of money compared to the cost of the scheme,

In the context of a £100bn+ railway, I agree. However it does remove a significant risk to the project certification and authorisation timescales.

and once it is gone we will likely never get it back.

I think we will not have another chance to do it until the signalling equipment needs renewal.

 

[Bletchleyite]

China has GoA 3 automation on a high speed railway.....

EDIT: The Beijing–Zhangjiakou high-speed railway is fitted with driverless (but person on board) automatic train operation and opened in time for the 2022 Winter Olympics.

It's a pity we can't get to "guard only operation" on HS2, really, even if a driver would need to board at Birmingham Interchange to take the train north onto the classic lines. It would be a significant cost saving day to day, and is easier done on fully segregated high speed lines than even something like the DLR.

(The latter of course means that no current drivers' jobs would be threatened by such a thing, either)

 

[AlexSJW]

At no point has there been a suggestion that people won't get on the HS2 trains. Nor does the classic line need any more people being encouraged on to the train.

So his answer is completely irrelevant.

But it generates a good soundbite (oh scrap it and give 1% of the funding to a nice interior desingner) and it's said by somebody who goes viral on Tiktok so it must be brilliant.

FWIW I find a lot of what he says interesting, but that specific answer is missing the point of HS2 completely.

You never fully got his points which he clearly stated.
He was for the mindset of marketing. Make the experience attractive, enjoyable, and efficient - and straight onto the train with no station delays, which saves time. No good having a super-dupa fast train if you wait at the station, making it slower overall. Also good connections which HS2 lacks.

He never said reduce capacity, or make it ridiculously slow. He was on about making it faster would be part of it, but clearly not fast for fast sake. The route, the Y, was also ridiculous. He rightly stated the sums guys got it wrong. And boy they got it hopelessly wrong.

He is one hundred percent correct.

BTW, Jony Ive designed the Smart Phone.

 

[Krokodil]

Who is Rory Sutherland?

An advertising executive.

So someone completely unqualified where infrastructure planning is concerned.

 

[Stossgebet]

Back onto the subject of ATO then...

Why is it not possible to update the HS2 infrasteucture to ATO a few years after it opens?
same question for the max speed?

My simple thought process is this. When they get HS2 (or whatever they then call it) extended towards Crewe, would it be possible build that extention to the original 360kph and ATO operation, and use it as the test track to certify the additional speed and method of operation, before opening it to passenger use, and upgrade phase 1 at the same time, in a dual upgrade?

 

[Krokodil]

Back onto the subject of ATO then...

Why is it not possible to update the HS2 infrasteucture to ATO a few years after it opens?
same question for the max speed?

My simple thought process is this. When they get HS2 (or whatever they then call it) extended towards Crewe, would it be possible build that extention to the original 360kph and ATO operation, and use it as the test track to certify the additional speed and method of operation, before opening it to passenger use, and upgrade phase 1 at the same time, in a dual upgrade?

I suppose that it'll depend on whether the ATO kit just "bolts on" (not necessarily in a literal sense) to the ETCS kit that's being fitted anyway.

In any case, it's only worth fitting if the Eastern leg ever happens, that's when you really need the reduced headways.

 

[blimmo]

It's a pity we can't get to "guard only operation" on HS2, really, even if a driver would need to board at Birmingham Interchange to take the train north onto the classic lines. It would be a significant cost saving day to day, and is easier done on fully segregated high speed lines than even something like the DLR.

(The latter of course means that no current drivers' jobs would be threatened by such a thing, either)

Spending more upfront to reduce operating costs? Impossible!

 

[HSTEd]

It's a pity we can't get to "guard only operation" on HS2, really, even if a driver would need to board at Birmingham Interchange to take the train north onto the classic lines. It would be a significant cost saving day to day, and is easier done on fully segregated high speed lines than even something like the DLR.

(The latter of course means that no current drivers' jobs would be threatened by such a thing, either)

It's certainly an interesting path to go down.

I assume you'd end up with a safety trained "guard" in a modular buffet.
After all, if they have to be present anyway they can serve tea and biscuits!

 

[bahnause]

Spending more upfront to reduce operating costs? Impossible!

Operating costs are rarely the main business case for ATO. It might not be cheaper at all. It‘s capacity.

 

[zwk500]

Why is it not possible to update the HS2 infrasteucture to ATO a few years after it opens?
same question for the max speed?

It's very possible, but would require access to run the testing and certification runs.

My simple thought process is this. When they get HS2 (or whatever they then call it) extended towards Crewe, would it be possible build that extention to the original 360kph and ATO operation, and use it as the test track to certify the additional speed and method of operation, before opening it to passenger use, and upgrade phase 1 at the same time, in a dual upgrade?

I expect that further delay of delivery to the north to allow testing for upgrades to the south would be undesirable from a commercial and political standpoint.

More likely is a programme somewhat similar to the East Coast Digital Programme - passenger services run "as they are" while the infrastructure is upgraded to support ATO and overnight testing is done (leaving windows for maintenance) to get the certification for passenger service.

I suspect the government is not going to want to incur additional capital costs on a section once it opens, and therefore the next opportunity for ATO is very likely to be when the trains or signalling equipment need replacement.

 

[Zomboid]

very possible, but would require access to run the testing and certification runs.

The issues with introducing ATO on HS2 after services start with manual driving wouldn't be mostly technical/ engineering.

 

[MatthewHutton]

Operating costs are rarely the main business case for ATO. It might not be cheaper at all. It‘s capacity.

You don’t need the capacity since the Y was abandoned. You do need to do the certification as quickly as possible.

== Doublepost prevention - post automatically merged: 22 May 2026 ==

Spending more upfront to reduce operating costs? Impossible!

It’s not always worth it.

== Doublepost prevention - post automatically merged: 22 May 2026 ==

The time penalty from a lower line speed is less (as you spend less time braking), but I wouldn't be surprised if this added 4 minutes on a Curzon Street to Euston journey in total.

The penalty in total from 320km/h to 0 to 320km/h with a 2 minute standard dwell should be like 5.5, maybe 6 minutes at a push. The Sanyo Shinkansen manages 300 to 0 to 300 with 4.5 minutes of penalty and probably with a one minute dwell.

EDIT: Didcot parkway has a 4 minute penalty including a 2 minute dwell from 200-0-200km/h.