Double SPAD near Crofton West Junction

[Taunton]

The section is question is descending at gradients that vary between 1 in 200 and 1 in 100. Noticeable but not hideous in normal freight operations.

On a descending gradient the brake has to get rid of not only the kinetic energy of the train's movement, but also the potential energy of the height it descends during the braking. The latter appears at a greater rate if the train is descending the same gradient at higher speed, so a brake that can hold a train on a descent at a low speed may not do so if the train is going faster. And of course if it's failing to stop as expected the stopping distance, and thus the energy it needs to dissipate, is that much more. This is one reason the old unfitted freight trains would tend to lose a lot of speed before they started descending, even if the gradient wasn't steep enough to have to stop and pin down brakes.

One would have thought that any such descent would have been mentioned in the report, although you wouldn't in the old days pin down for it - that was for gradients a lot steeper. Anyway, David L Smith's books from Scotland dispelled the old approach of coming over the summit dead slow and going ever faster to keep the couplings tight - you aimed to come close to stopping at the bottom, to pick up the coupler slack from rest, something not an issue nowadays.

I'm still not impressed with loco brakes that can't stop a train of empties in those conditions - it appears to have taken five normal signal sections to pull up from 50 mph - it ran through a YY-Y-R sequence, then a further Y-R at the next junction, and overshot well beyond that. Unfitted freights used to have to pull up just from the distant, like anything else, and ran a LOT faster than walking pace.

 

[ac6000cw]

I believe freight trains in the US (which can be very long!) have these - the End-of-train device connects into the brake pipe as well as being a tail lamp, and reports the pressure back by radio to an indicator in the cab. I think they also have a radio-operated brake valve as well to allow the brakes to be applied from both ends of the train simultaneously.

They use both types - 'passive' ones just report the brake pipe pressure, 'active' ones can also vent the brake pipe (but AFAIK only fully vent it - there is no graduated control, so basically it's only intended for use in emergency situations). The 'active' ones are generally mandatory on trains operating over long, steep gradients i.e. in mountainous areas.

EOTDs/FREDs replaced the caboose, basically.

 

[GB]

I'm still not impressed with loco brakes that can't stop a train of empties in those conditions - it appears to have taken five normal signal sections to pull up from 50 mph - it ran through a YY-Y-R sequence, then a further Y-R at the next junction, and overshot well beyond that. Unfitted freights used to have to pull up just from the distant, like anything else, and ran a LOT faster than walking pace.

You don't need to be impressed. A single loco is not designed to safely stop a 500+ ton train.

What happened back in BR and before doesn't mean its relevant by today's standards...

 

[The Chimaera]

One would have thought that any such descent would have been mentioned in the report, although you wouldn't in the old days pin down for it - that was for gradients a lot steeper. Anyway, David L Smith's books from Scotland dispelled the old approach of coming over the summit dead slow and going ever faster to keep the couplings tight - you aimed to come close to stopping at the bottom, to pick up the coupler slack from rest, something not an issue nowadays.

I'm still not impressed with loco brakes that can't stop a train of empties in those conditions - it appears to have taken five normal signal sections to pull up from 50 mph - it ran through a YY-Y-R sequence, then a further Y-R at the next junction, and overshot well beyond that. Unfitted freights used to have to pull up just from the distant, like anything else, and ran a LOT faster than walking pace.

I think you rather over estimate the braking ability of locomotives they are not designed to stop trains on there own, the laws of physics dictate otherwise. Energy rises with speed and a single loco would take forever to bring a train to a stand on its own, from 50mph on a falling gradient. For the record Loose couple trains were limited to 25mph max and were driven at a speed that the driver new he could pull up within the protection distance, in case of emergency. The drivers new the route they were on and would be travelling well below 25mph on the approach of steep falling gradients and had the assistance of the guards brake and wagon handbrakes if required. Even so runaways were common, going too fast risked the crews lives and was frowned on by all.
Incidentally steam loco's are said to have had better brakes than diesels due the larger wheel diameter gave a larger contact patch with the rail, this would support more braking effort on the wheel which was also helped by the extra area of the wheel aiding cooling of the tyre.

 

[edwin_m]

Incidentally steam loco's are said to have had better brakes than diesels due the larger wheel diameter gave a larger contact patch with the rail, this would support more braking effort on the wheel which was also helped by the extra area of the wheel aiding cooling of the tyre.

I agree with the cooling issue but the brake block surface is curved to match the wheel so the contact patch doesn't depend on wheel diameter. Plus the laws of physics state that friction doesn't depend on the size of the contact patch, just on the force being applied through it.

 

[The Chimaera]

I agree with the cooling issue but the brake block surface is curved to match the wheel so the contact patch doesn't depend on wheel diameter. Plus the laws of physics state that friction doesn't depend on the size of the contact patch, just on the force being applied through it.

The contact patch referred to was wheel to rail, you can only brake a particular wheel so much before you lose adhesion with the rail and the wheel will then slide give negligible braking effect. Adhesion increases will weight upon the axle and also with the size of the contact patch. Or am I reading the second part wrong?
Steam loco’s tend to have 4 blocks per wheel at least for the driving wheels, more Contact area for the friction material should give more braking effort I would have thought.

 

[Llanigraham]

I'm still not impressed with loco brakes that can't stop a train of empties in those conditions - it appears to have taken five normal signal sections to pull up from 50 mph - it ran through a YY-Y-R sequence, then a further Y-R at the next junction, and overshot well beyond that. Unfitted freights used to have to pull up just from the distant, like anything else, and ran a LOT faster than walking pace.

Unfitted freights ran a lot slower than that and would often pin down brakes in advance of downgrades.
And plenty overshot signals!

 

[4F89]

The contact patch referred to was wheel to rail, you can only brake a particular wheel so much before you lose adhesion with the rail and the wheel will then slide give negligible braking effect. Adhesion increases will weight upon the axle and also with the size of the contact patch. Or am I reading the second part wrong?
Steam loco’s tend to have 4 blocks per wheel at least for the driving wheels, more Contact area for the friction material should give more braking effort I would have thought.

More wheel to rail, per wheel contact patch, but less wheels using braking force on a kettle. Six of one, half dozen of the other i think, when weight on the wheels is taken into account

 

[big all]

2x more brake blocks are more likely to double replacement intervals as you wont get more stopping power as the limit is wheel to rail friction
this why regenerative braking is great when the wheel rotation slows the load reduces until the friction is in balance so you never get damaged wheels but maximum available retardation untill perhaps 20mph when friction brakes block or pads takes over??

 

[DelW]

In general terms, the maximum frictional resistance between two surfaces, before they start to slide, is the force normal to the surface, multiplied by the coefficient of friction. The size of the contact patch isn't a factor in the calculation.
However the mechanics of the wheel / rail interface are complex and are still not fully understood, and don't always obey classic mechanics rules. EMD's wheel creep system, which gets better traction force by allowing the wheels to rotate slightly faster than the loco is moving, is a case in point. But that's taking us away from spads.

 

[edwin_m]

The contact patch referred to was wheel to rail, you can only brake a particular wheel so much before you lose adhesion with the rail and the wheel will then slide give negligible braking effect. Adhesion increases will weight upon the axle and also with the size of the contact patch. Or am I reading the second part wrong?
Steam loco’s tend to have 4 blocks per wheel at least for the driving wheels, more Contact area for the friction material should give more braking effort I would have thought.

Noted, apologies I should have picked up from your previous post that you were referring to wheel-rail contact. However the same argument applies - the retarding force depends on the weight on the braked wheels and the coefficient of friction between them and the rail. Try to add too much braking to a wheel and it will just lock and slide, reducing the deceleration and damaging the wheel.

 

[Merle Haggard]

No-one, so far, has been able to explain the introduction of the relaxation provided by the 10 hour rule.
The purpose of the brake test (static, before departure) was not only to show basic faults in train preparation but also less visible but equally crucial problems.
In vacuum brake days, and cold winters, condensation in the system freezing and making blockages was not uncommon, and was shown up by the test. Possibly air systems would be more likely to have condensation present, and I wonder how, in the risk assessment that was presumably done before this change was made, the possibility of temperatures dropping below freezing in the 10 hours between test and departure was covered.

Edited for typo

 

[Taunton]

No-one, so far, has been able to explain the introduction of the relaxation provided by the 10 hour rule.

I suspect it was done wholly for economy in crew time. The report describes, in the little hints that RAIB seem to employ, that the driver arrived on shift and was fully engaged in other activities before departure, hence missing the air pressure buildup, and how the trainee group had to rush off when he turned up as well (thus both starting the chain of events). Someone has devised this to save staff costs.

In any event it was stipulated that it covered where the loco had not been left unattended, obviously not the case here, and probably was intended to cover not needing to go through another test again if, for example, there was a gross delay on departure, or the train had just been standing such as happens with works trains.

I wonder how, in the risk assessment that was presumably done before this change was made

Do not put too much faith in Risk Assessments nowadays, they have become something of a box-ticking exercise, and commonly are developed starting with the end of the process, justifying it, and then working backwards, just covering those risks which will be OK for the desired outcome.

 

[HSTEd]

Possibly air systems would be more likely to have condensation present, and I wonder how, in the risk assessment that was presumably done before this change was made, the possibility of temperatures dropping below freezing in the 10 hours between test and departure was covered.

This gives me shades of the loss of USS Thresher.

 

[Nick_C]

More wheel to rail, per wheel contact patch, but less wheels using braking force on a kettle. Six of one, half dozen of the other i think, when weight on the wheels is taken into account

Most steam locos used for heavy unfitted freights would have had more braked wheels, as usually the tender was braked as well, so a 9F would have 16 braked wheels as opposed to 12 on a 66.

 

[Merle Haggard]

For an empirical assessment of steam vs diesel braking power look at the report into the accident at East Langton on 20/8/65 - it's on Railways Archive.
In summary, a (steam) 9F 2-10-0 had brought a train consisting of 45 wagons of wet slack to Leicester with no problems; 2 wagons were detached and D5383 replaced the 2-10-0. The diesel was overpowered by the train at restrictive signals, and collided with a preceding ballast train. It's fair to say that preparation errors may have been made by the guard in possibly not making it clear to the driver that, although running as 7*, it had no fitted head.
The report also gives an anecdotal example of traincrew escape actions; the second-man took shelter in the engine room and held on to the train heating boiler and, although the loco overturned, survived unharmed. The driver, who took to the rear cab, did not.

 

[edwin_m]

Most steam locos used for heavy unfitted freights would have had more braked wheels, as usually the tender was braked as well, so a 9F would have 16 braked wheels as opposed to 12 on a 66.

The number of wheels isn't relevant. The critical factor is the amount of weight on braked wheels relative to the total weight of the train. Creep control on braking, which I think is provided, but perhaps only for rheostatic braking, would also help the 66 by effectively increasing the coefficient friction.

For an empirical assessment of steam vs diesel braking power look at the report into the accident at East Langton on 20/8/65 - it's on Railways Archive.
In summary, a (steam) 9F 2-10-0 had brought a train consisting of 45 wagons of wet slack to Leicester with no problems; 2 wagons were detached and D5383 replaced the 2-10-0. The diesel was overpowered by the train at restrictive signals, and collided with a preceding ballast train. It's fair to say that preparation errors may have been made by the guard in possibly not making it clear to the driver that, although running as 7*, it had no fitted head.
The report also gives an anecdotal example of traincrew escape actions; the second-man took shelter in the engine room and held on to the train heating boiler and, although the loco overturned, survived unharmed. The driver, who took to the rear cab, did not.

Several other factors that could have come into play here on a quick skim:

• The train had a single Type 2 locomotive onwards from Leicester, which would have been much lighter than a Class 9 steam locomotive and therefore other things equal had much less braking force.
• Presumably the driver of the previous working had been made aware it was unfitted and took this into account on its journey, whereas the diesel driver hadn't been so advised.
• The steam-worked journey from Derby to Leicester was pretty flat and may not have encountered signal checks.
• The guard on the diesel journey seems not to have been aware of very much and didn't contribute to trying to stop it.
• Speculation, but the driver may have been less familiar with diesels. Other accident reports have noted a tendency for steam men to lose their judgment of speed in the very different environment of a diesel, and not necessarily refer to the speedometer.

For all these reasons I don't think we can draw a conclusion about the relative effectiveness of brakes on steam locos versus diesels.

The report also mentions that speeds of unfitted trains were assumed to be reduced on descending gradients steeper than 1 in 200.