Max speed on UK network

[notadriver]

There’s no reason for ETCS to be so restrictive - I’m pretty sure the TVM 430 system on HS1 doesn’t force a more restrictive driving style.

 

[hexagon789]

It will be the cumulative effects of allowances for poor rail conditions, adverse weather, braking systems at the limits of tolerances, etc etc. All the things that a driver usually assimilates mentally through a simple running brake test and allows for as appropriate.

== Doublepost prevention - post automatically merged: 5 Aug 2021 ==



I’m not certain of the exact year, weren’t the original 3h59 schedules slowed down a little bit after a couple of years of experience? Certainly there were about 3 or 4 ‘Scottish Pullmans’ each each direction by the mid 1990s, some of them had more stops than just York and Newcastle.

There were two timed at 3h59 each way in 1992, plus a 3rd at 4h08 that called at Darlington.

Only a year later all the recovery time was re-inserted extending the schedule to 4h15 in May 1993. Throughout the 1990s the schedule varied, sometimes 3h59, sometimes just over 4 hours.

I believe GNER only ran a 3h59 schedule for one timetable period in the 2000s after initially getting rid of the 'fast' schedule in 1999. Post-2004 I believe it was eased to 4h02 then eventually switched to a more standard pattern timetable with a Peterborough stop in most Edinburgh fasts in addition to the usual York and Newcastle.

== Doublepost prevention - post automatically merged: 5 Aug 2021 ==

There’s no reason for ETCS to be so restrictive - I’m pretty sure the TVM 430 system on HS1 doesn’t force a more restrictive driving style.

Max 50% braking as standard I believe, as in you can use no more than 50% of the brake and that will be enough to bring the speed down without triggering an intervention.

 

[notadriver]

Max 50% braking as standard I believe, as in you can use no more than 50% of the brake and that will be enough to bring the speed down without triggering an intervention.

That seems plenty enough in normal driving ?

 

[edwin_m]

Max 50% braking as standard I believe, as in you can use no more than 50% of the brake and that will be enough to bring the speed down without triggering an intervention.

That seems plenty enough in normal driving ?

Ideally you probably aim to stop on the regenerative brake as far as practical, with the friction brake used only in the event of misjudgment or system failure. That reduces power consumption, pad wear and particulate emissions. Something like an 80x that has numerous motored axles is probably getting close to that.

 

[hexagon789]

That seems plenty enough in normal driving ?

That is standard on French LGVs to the best of my knowledge, so yes I would say so.

Ideally you probably aim to stop on the regenerative brake as far as practical, with the friction brake used only in the event of misjudgment or system failure. That reduces power consumption, pad wear and particulate emissions. Something like an 80x that has numerous motored axles is probably getting close to that.

In a previous thread a driver for Eurostar said that on 373s high-speed trimming was done on the rheo only but station calls were done on about 50% braking with mixed use of rheo and air with air only (tread and discs) below 40km/h (25mph) - where the rheo fades out.

 

[edwin_m]

That is standard on French LGVs to the best of my knowledge, so yes I would say so.


In a previous thread a driver for Eurostar said that on 373s high-speed trimming was done on the rheo only but station calls were done on about 50% braking with mixed use of rheo and air with air only (tread and discs) below 40km/h (25mph) - where the rheo fades out.

The 373 probably has less dynamic (rheostatic or regenerative) braking than other trains, having a fairly small number of motored axles that it can be applied to. Many more modern trains have more, especially those designed for rapid acceleration between more frequent stops. More modern traction will probably also regenerate down to a much lower speed.

 

[hexagon789]

The 373 probably has less dynamic (rheostatic or regenerative) braking than other trains, having a fairly small number of motored axles that it can be applied to. Many more modern trains have more, especially those designed for rapid acceleration between more frequent stops. More modern traction will probably also regenerate down to a much lower speed.

390s can rheo/regen down to 7mph but the several low speed slide incidents, culminating in the buffer stop collision at Liverpool Lime Street, when they were new led to this being increased (to nearer 20mph I believe) not long after.

 

[alf]

I believe this is also true of the 80x. Initially they regened down to about 7 mph but it was reset to 15 mph after an incident or near incident.

 

[notadriver]

That is standard on French LGVs to the best of my knowledge, so yes I would say so.


In a previous thread a driver for Eurostar said that on 373s high-speed trimming was done on the rheo only but station calls were done on about 50% braking with mixed use of rheo and air with air only (tread and discs) below 40km/h (25mph) - where the rheo fades out.

If driven properly surely one should never have to use the brake to ‘trim’ at high speed unless slowing for speed restrictions or stations? Use of the brake to stay within the limit means the train has been wasteful of traction power (on HS1)

 

[snowball]

If driven properly surely one should never have to use the brake to ‘trim’ at high speed unless slowing for speed restrictions or stations? Use of the brake to stay within the limit means the train has been wasteful of traction power (on HS1)

I know nothing about these matters but I would have thought if you were, say, going up Shap and then down the other side, you might well need to brake.

 

[waverley47]

I know nothing about these matters but I would have thought if you were, say, going up Shap and then down the other side, you might well need to brake.

Indeed. Trains have very little rolling resistance, and the average train will continue to accelerate to its terminal velocity (gravity - (rolling resistance + air resistance)) on any slope more than about 1 in 90. On a level track, the rolling resistance of a train is minimal, however more than that gradient is enough, without brakes applied, to move a train under gravity.

The average train will be braking for a decent proportion of its journey, and while harder braking on approach to stations may not result in longer journey times, on downhill sections that would result in a more sawtooth speed profile of acceleration and braking, and would inevitably lead to maintenance and journey time issues.

 

[AM9]

Indeed. Trains have very little rolling resistance, and the average train will continue to accelerate to its terminal velocity (gravity - (rolling resistance + air resistance)) on any slope more than about 1 in 90. On a level track, the rolling resistance of a train is minimal, however more than that gradient is enough, without brakes applied, to move a train under gravity.

That's true, but the terminal velocity on a downward slope will be determined by the gradient, but also by the initial velocity, given that no slope is infinitely long. So, cresting at 300km/h and descending a 1:50 will increase the speed by some measure, but the air resistance will already be limiting the maximum speed, whereas at 160/h on the same gradient, the air resistance would be just over a quarter of that at the higher speed, (air resistance is proportional to the square of the speed for a given cross sectional area).
I have no inside knowledge of this, but I would imagine that running under ATO, the HS2 trains could be running under a pre-defined speed profile, i.e. with virtually all dynamics of the train and route known, (current speed, rolling resistance, air resistance curve, actual weight including passengers, maximum acceleration rate, actual line voltage, gradient profile ahead and of course the headway clearance) that the applied power can be finely trimmed to maintain the speed over peaks and through dips to minimise brake applications and reduce overall power consumption.
Sorry for the long sentence.

 

[hexagon789]

If driven properly surely one should never have to use the brake to ‘trim’ at high speed unless slowing for speed restrictions or stations? Use of the brake to stay within the limit means the train has been wasteful of traction power (on HS1)

Im just repeating what the poster put. I assume they meant that if travelling at say 300km/h and the TVM430 starts flashing then braking would be done only on the rheo at this point to reduce speed to that demanded by the TVM430 of for example a preceeding train was being caught up for some reason.

== Doublepost prevention - post automatically merged: 6 Aug 2021 ==

I believe this is also true of the 80x. Initially they regened down to about 7 mph but it was reset to 15 mph after an incident or near incident.

Wasn't there a spate of issues with the friction brake being very slow to blend in at low speeds on the 80x?

 

[Jozhua]

The US very well may have achieved the highest steam top speed, but they didn't put speedos on them so that basically discounts everything lol

 

[mike57]

I know my father is convinced there was 140mph running in daily service on the ECML when the IC225 sets entered traffic, he had a memorable run on the Scottish Pullman (1530 ex-Kings Cross) which pulled into York at 1703, a whopping 12 mins early against a tight non-stop schedule. Admittedly the sets were in their original 2+8 formation but it shows what was achievable with a clear run and a good driver.

That would imply some significant distance running at above 125mph. Thinking about the route, Newark flat crossing, Doncaster, the initial run out of London are going to be well below 125mph I would have thought. I can remember being on a southbound Flying Scotsman in pre Selby Diversion HST days, we left Newcastle some 10mins late and arrived in Kings Cross on time. Pre GPS days so I had no easy way of confirming, but my impression was Darlington-York was taken quicker than usual, and you could feel that the full power and braking was being used. Couldnt happen today.

 

[Wilts Wanderer]

It would be interesting to know how much engineering padding time there was in the 3h59 schedules. BR tended to be quite liberal with allowances in many places - blocks of [4] approaching a key junction weren’t unheard of - although individual trains with tight schedules might omit them. That being said, the ECML was recently modernised so the frequency of TSRs etc would be low compared to, say, the WCML which was infamous for pedestrian schedules through the 1990s.

(Memories of 15 min stops at Rugby, Preston, etc… yawn!)

 

[dk1]

It would be interesting to know how much engineering padding time there was in the 3h59 schedules. BR tended to be quite liberal with allowances in many places - blocks of [4] approaching a key junction weren’t unheard of - although individual trains with tight schedules might omit them. That being said, the ECML was recently modernised so the frequency of TSRs etc would be low compared to, say, the WCML which was infamous for pedestrian schedules through the 1990s.

(Memories of 15 min stops at Rugby, Preston, etc… yawn!)

I don’t recall there being too much recovery in that schedule. I used to follow it most mornings on TRUST at work. Reliability wasn’t great & it has been said before that right time arrivals for both the 06:00up & 15:00 down Scottish Pullmans stood at around 30%.

 

[Annetts key]

ETCS will have a few interesting impacts on train performance, the assessment is that the braking curves are 15% longer than current defensive driving styles, on average. Not particularly conducive to good journey times if you have to pass through any sort of traffic bottleneck.

Why?! That’s really silly. Is it just because the system can’t judge actual rail conditions so errs on the safe side?

Are you talking about the trains performance without encountering restricted signals/block sections. Or are you referring to trains appearing to “lose” time when they encounter restricted signals/block sections?

If the later, surely it’s a similar situation to the high speed trains that currently use the GWML ATP system between Bristol and London Paddington.

Yes, if a train is encountering double yellow, single yellow or red aspects, it takes longer to get to the signal, as the computer enforces a suitable braking curve on the assumption that the signal has not changed aspect.

In some places, that is where this is likely to occur, an infill beacon or infill loop is provided to update the on board computer. But even where this is not the case, unless the train ahead is diverging to a different line/route, it does not make much difference, as you are limited by the time it takes for the train ahead to clear the next section.

Obviously it can make a difference if the train ahead does diverge via a junction. But if the timetable has been worked out correctly, this should not normally occur.

Or am I missing something?

 

[class26]

That would imply some significant distance running at above 125mph. Thinking about the route, Newark flat crossing, Doncaster, the initial run out of London are going to be well below 125mph I would have thought. I can remember being on a southbound Flying Scotsman in pre Selby Diversion HST days, we left Newcastle some 10mins late and arrived in Kings Cross on time. Pre GPS days so I had no easy way of confirming, but my impression was Darlington-York was taken quicker than usual, and you could feel that the full power and braking was being used. Couldnt happen today.

Newark flat crossing is 100 mph as is Grantham station itself, 125 between the two.

 

[hexagon789]

It would be interesting to know how much engineering padding time there was in the 3h59 schedules. BR tended to be quite liberal with allowances in many places - blocks of [4] approaching a key junction weren’t unheard of - although individual trains with tight schedules might omit them. That being said, the ECML was recently modernised so the frequency of TSRs etc would be low compared to, say, the WCML which was infamous for pedestrian schedules through the 1990s.

(Memories of 15 min stops at Rugby, Preston, etc… yawn!)

None is the short answer.

Iirc technically the northbound runs had one minute pathing allowance, but southbound there was nothing.

 

[jfollows]

In the 29/5/94 to 24/9/94 timetable I see
1S31 15:00 King's Cross to Glasgow Central
[1.5]<2> after Retford
16/25 pass Doncaster
<2.5> before Colton Junction
York 16:45.5-16p48.5
[1] before Darlington
[3] before Newcastle
Newcastle 17:39.5-17:42.5
[2]<1> after Drem
Edinburgh arrive 19:05

I make that a total of 13 minutes of extra time for a 4h05 schedule with two stops.
Removing the York stop would save 4 minutes I think (there were no non-stop down trains, but a non-stop up train was 4 minutes quicker than a stopper).
3h59 with two stops would require a reduction to 7 minutes of the extra time in the schedule
3h59 omitting the York stop would require a reduction to 11 minutes which is more realistic

However, was the schedule realistic in the first place without the recovery time?

 

[The Planner]

5½ in diamond time? basically admits the schedule was always going to be ropey.

 

[Wilts Wanderer]

5½ in diamond time? basically admits the schedule was always going to be ropey.

That was my thought as well.

By comparison in today’s timetable the standard benchmark is [1] for every 30-50 miles dependent on route and TSR strategy, and very little mandated use of <> time at all. However schedules are still slow due to application of pathing time ( ) - ie catching other trains up on paper - and increased numbers of stops in schedules. I would be very surprised if there’s anything like those allowances now. I’d expect probably [1] approaching each of Peterborough, Doncaster, York, Darlington, Newcastle, Berwick and Portobello Jcn / Edinburgh.

To bring back to topic - with those sorts of allowances in the mid 1990s schedules, how critical was 125mph running to achieve overall journeytimes?

 

[jfollows]

1E01 05:40 Edinburgh to King's Cross with 1 stop had (pre-Covid timetable)
<1> after Prestonpans
[1] before Berwick
[1] before Heaton
Newcastle 07:02-07:05
<0.5> before Darlington
[1] before York
[1] before Doncaster
[1] after Retford
<1> before Welwyn
[1] before King's Cross
King's Cross arrive 09:40
4h00 including 8.5 minutes of extra time

 

[Bald Rick]

To bring back to topic - with those sorts of allowances in the mid 1990s schedules, how critical was 125mph running to achieve overall journeytimes?

Very.

The absence of box time was because the then Regional Civil Engineer bought some Dynamic Track Stabilisers to eliminate TSRs after track renewal. Didn’t quite work out like that as it happens.

 

[Watershed]

That was my thought as well.

By comparison in today’s timetable the standard benchmark is [1] for every 30-50 miles dependent on route and TSR strategy, and very little mandated use of <> time at all. However schedules are still slow due to application of pathing time ( ) - ie catching other trains up on paper - and increased numbers of stops in schedules. I would be very surprised if there’s anything like those allowances now. I’d expect probably [1] approaching each of Peterborough, Doncaster, York, Darlington, Newcastle, Berwick and Portobello Jcn / Edinburgh.

To bring back to topic - with those sorts of allowances in the mid 1990s schedules, how critical was 125mph running to achieve overall journeytimes?

Yes, there are 7 mins engineering allowances in a Kings Cross-Edinburgh run, but there are an extra 4 mins performance allowance, at Woolmer Green, Colton Jn, Darlington and Drem.

 

[Wilts Wanderer]

Thank you both, equally fascinating responses! It’s interesting to see performance allowance (I.e. <>) still in action within ECML schedules, is that mandated by Rules of the Plan or just the preference of the operator?

 

[zwk500]

Thank you both, equally fascinating responses! It’s interesting to see performance allowance (I.e. <>) still in action within ECML schedules, is that mandated by Rules of the Plan or just the preference of the operator?

Performance allowance is part of the Timetable Planning Rules, as they are now.

 

[edwin_m]

Are you talking about the trains performance without encountering restricted signals/block sections. Or are you referring to trains appearing to “lose” time when they encounter restricted signals/block sections?

If the later, surely it’s a similar situation to the high speed trains that currently use the GWML ATP system between Bristol and London Paddington.

Yes, if a train is encountering double yellow, single yellow or red aspects, it takes longer to get to the signal, as the computer enforces a suitable braking curve on the assumption that the signal has not changed aspect.

In some places, that is where this is likely to occur, an infill beacon or infill loop is provided to update the on board computer. But even where this is not the case, unless the train ahead is diverging to a different line/route, it does not make much difference, as you are limited by the time it takes for the train ahead to clear the next section.

Obviously it can make a difference if the train ahead does diverge via a junction. But if the timetable has been worked out correctly, this should not normally occur.

Or am I missing something?

ETCS level 2 updates movement authorities (which replace signals) to the train via GSM-R. So if the movement authority is extended (equivalent to an adverse signal clearing ahead of the train) the train is able to accelerate almost immediately. It doesn't have to wait to pass over a beacon or loop to update the system. I presume it also avoids the limitations of approach control, because the train is being supervised to pass the junction at the maximum permitted speed for its route, not to obey the aspect of an approach controlled signal that is probably more restrictive.

 

[Wilts Wanderer]

ETCS level 2 updates movement authorities (which replace signals) to the train via GSM-R. So if the movement authority is extended (equivalent to an adverse signal clearing ahead of the train) the train is able to accelerate almost immediately. It doesn't have to wait to pass over a beacon or loop to update the system. I presume it also avoids the limitations of approach control, because the train is being supervised to pass the junction at the maximum permitted speed for its route, not to obey the aspect of an approach controlled signal that is probably more restrictive.

Is GSMR a continuous feed or does it update via fixed eurobalises? I’d be surprised if the braking curve updates in real time.

Edit - as I engage brain - or do the Eurobalises just describe the route characteristics to the train. GSMR being a radio system of course.