......are there any conditions.......

[whole diary]

[Network SouthEast]

There are normal passenger trains in service that can de-ice third rail. A member of public might have trouble spotting an Electrostar that has a de-icing tank, for example, because the tank is below sole-bar level.

[stuving]

Older locomotives and carriages - note not modern multiple units - had tread brakes as rightly pointed out in an earlier post so the driving wheels were cleaned off. Also, steam engines and older diesels with steam boilers had the ability to spray steam onto the tracks under the driving wheels.

There is - of course - a Railway Group Standard: "Low Adhesion between the Wheel and the Rail ^ Managing the Risk" (GE/RT8040, Issue Two, February 2009).

Its important requirement is contained in a single paragraph:

2.1 Joint requirements for infrastructure managers and railway
undertakings
2.1.1 General requirements
2.1.1.1 Infrastructure managers and railway undertakings shall jointly implement measures to reduce the risks generated by low adhesion between the wheel and the rail that cannot be eliminated by local treatment at specific sites.

It then goes on to the requirement for site specific [sic] action plans, to define that local treatment. It mentions train-mounted equipment (sanding), but not other more general measures such as rail treatment trains (Sandite and jetting).

However, there is a guidance note that gives lists of potential measures:  "Guidance on Low Adhesion between the Wheel and the Rail ^ Managing the Risk" (GE/GN8540, Issue One, February 2009).

The main list is as follows:

GN22 Example of possible measures include:
a) Vegetation clearance
b) Water jetting (including frequency)
c) Sandite application (including frequency)
d) Traction Gel application
e) Hand sanding
f) SPAD^▸ (Signal Passed At Danger) prediction equipment fitted at level crossings
g) Driver briefing
h) Take signalled route out of use
i) Special box instructions restricting certain moves
j) Selected trains to perform running brake tests
k) Train strengthening
l) Use of tread braked stock instead of disc braked stock
m) Fitting scrubbing devices to the wheels of rolling stock to eliminate or reduce
contamination of the wheels
n) Fitting devices to the wheels of rolling stock to enhance the performance of the
braking system
o) Fitting sanders to trains
p) Turning off flange lubricators
q) Monitoring and inspection of low adhesion sites.

and a little later:

GN25 In developing the plans, consideration should be given to any train borne equipment fitted
to improve braking performance under conditions of low adhesion, including:
a) Optimised wheel slide protection (WSP) equipment
b) Scrubber brakes
c) Track brakes
d) Sanding equipment including:
i) Emergency sanding device
ii) Automatic sanding device
iii) Dual function sanders (braking and traction)
iv) Variable rate sanders.
GN26 Plans should also consider the benefits, where provided, of systems to detect wheel slide
and alert the driver or systems that provide remote monitoring to alert drivers of low
adhesion.

... and, if you know of a way to improve adhesion with a kitchen sink ...

The fact that something is in the list does not mean it works, or even that all parties think it does. Tread brakes are thought to help by a lot of drivers, but attempts to measure the effect show it is at best marginal. The RAIB^▸ (Rail Accident Investigation Branch), in report 25/2006 part 3, looked at the evidence, and said this:

194 Disc brakes are provided on modern rolling stock whereas the stock they replaced was
equipped with tread brakes. Some drivers have suggested that tread brakes are better than
disc brakes at stopping trains in low adhesion conditions because they act directly on the
wheel, having the effect of cleaning the wheels and improving adhesion accordingly. This
may, in part, be attributable to the fact that tread-braked units achieve deceleration rates of
0.7m/s2, whereas modern disc braked rolling stock can achieve deceleration rates of 1.0m/
s2 or greater. Thus the risk of exceeding available adhesion will be greater, although WSP
should intervene to ensure that braking rate is matched to available adhesion. It may also
be the case that drivers of modern trains are more aware of wheelslide as the display on the
driver^s desk will indicate if WSP activity is taking place.

195 The AWG low adhesion manual addresses the issue of tread braking and disc braking.
Quoting work carried out by British Rail Research in which the relative performance of the
two brake types was examined, the AWG Manual makes the following assertions (RAIB
comments in brackets):

• Tread braked stock generally has lower operating speeds and lower braking rates
  than disc braked stock. This can help to create the impression that the brake itself is
  contributing to better performance.
• When rail conditions are able to support full braking demand, disc brakes are superior
  to tread brakes (this is borne out by drivers of modern rolling stock who agree that in dry
  conditions disc braking is more effective).
• When rail conditions are poor (i.e. in drizzle) a good WSP with disc brakes will stop the
  train in a comparable distance to a tread braked train (this runs counter to the evidence
  from drivers who have indicated that modern rolling stock was not as effective at
  stopping under drizzly conditions in autumn 2005 as the tread braked stock that it had
  replaced).
• When rail conditions are severe a good WSP system will stop the train in a comparable
  distance to a tread braked train (however, the incident at Lewes on 30 November 2005
  described in the Part 2 report suggests that WSP alone is not sufficient when drizzle falls
  onto a contaminated rail) but will avoid damage to the wheels in so doing, whereas the
  tread-braked train is likely to suffer wheel damage.

196 Although there may be no evidence to prove that tread braking yields benefits in stopping
performance under low adhesion conditions, one advantage of the cleaning effect on
wheel treads is to improve electrical contact between the wheel and the rail. For this
reason, auxiliary tread brakes (sometimes known as scrubber blocks) were fitted to some
disc braked vehicles to aid track circuit operation by cleaning and roughening the wheel
treads. There is insufficient evidence to judge whether vehicles equipped with auxiliary
tread brakes are less likely to be involved in adhesion-related station overruns because the
number of units so equipped is small.

I don't know if WSP (wheelslide prevention) could be applied to tread brakes, but I get the impression it could not. In any case, no-one is going to go back to tread brakes only, so it would have to be additional wheel-tread conditioning equipment. And that would need scientific evidence that it would be effective - which appears to be lacking.

[IndustryInsider]

From the RAIB^▸ (Rail Accident Investigation Branch)

Buffer stop collision at Chester station

...
The buffer stop was of an old design with only minimal capacity to absorb energy. The train destroyed it before overriding its remains to mount the platform where it came to rest. The front bogie was lifted off the track as the front of the leading vehicle rode up onto the platform.
 

I though this was relevant here - it is braking rather than traction effects of low adhesion, but closely related.

Thanks for posting the link.  If anyone is in any doubt as to why modern standards insist all new terminal platforms, such as the rebuilt 4,5 and 6 at Reading, have better quality buffers and roughly a carriage length of safety overlap, there is the proof!

[Chris from Nailsea]

The Rail Accident Investigation Branch (RAIB^▸ (Rail Accident Investigation Branch)) has released its report into a buffer stop collision at Chester station on 20 November 2013.

RAIB has made three recommendations and identified two learning points.

Summary:

At 12:12 hrs on Wednesday 20 November 2013 the 10:10 hrs passenger train from London Euston to Chester collided with the buffer stop at the end of the platform as it arrived at Chester station. The impact destroyed the buffer stop and caused the leading vehicle of the train to start to override the platform, demolishing a glass screen and damaging the platform. It was fortunate that there was no-one in this area at the time.

Several passengers fell over as the train stopped, but did not report injuries. One passenger was taken to hospital as a precaution and was released later the same day. The front of the train was damaged by the impact and the leading bogie was derailed.

Platform 1 at Chester station was closed until 04:15 hrs on 22 November for recovery of the train and repair of the buffer stop.

The accident was caused by the train sliding on wet rails that were also contaminated with leaf residue and traces of lubricating oil. The train was not equipped with automatic sanding equipment, which could have applied sand to the rails to improve adhesion. The train was fitted with emergency sanding equipment but the driver did not activate this until the train was too close to the buffers to be able to stop before striking them.

The train involved was a class 221 unit operated by Virgin Trains. The RAIB analysed records of low rail adhesion incidents on Network Rail infrastructure during Autumn 2013 and found several other cases of class 221 trains, and the similar class 220, encountering low adhesion when attempting to stop. In some cases the driver had to use the emergency sanding equipment. The RAIB analysis found that, per attempted stop, class 220 and 221 trains were the most likely type of train to be involved in a low adhesion incident. Almost all other types of train are fitted with automatic sanding equipment, or equivalent.

The RAIB has identified two key learning points and made three recommendations. The operators of class 220 and 221 trains are recommended to fit equipment to their trains to automatically apply sand when wheel slide is detected during heavy braking (Virgin Trains has recently informed the RAIB of its intention to fit such equipment to its fleet of class 221 trains). A recommendation is made to Virgin Trains to amend its operating instructions to drivers and a recommendation is made to RSSB^▸ (Rail Safety and Standards Board) to update the standard which governs the fitment of train sanding equipment.

The first learning point concerns the analysis of data from train data recorders to provide information on the location of low adhesion conditions on the network. The second learning point is that infrastructure managers need to be aware of changes in traffic patterns which necessitate the reassessment of the adequacy of buffer stops in terminal platforms.

[stuving]

There's quite a lot about how WSP systems work in this report, which is relevant to the discussion earlier about the merits of tread brakes. This is the relevant bit:

64 The second function of the WSP is provided to maximise the retardation of the train in the prevailing friction conditions. Research has shown that a degree of sliding contact between wheel and rail is beneficial in that it ^conditions^ the wheel and improves the friction between wheel and rail. The WSP on the class 221 was designed to attempt to keep the speed of the wheels 17% below the speed of the vehicle during braking in wheel slide conditions.

So when the wheel wants to slide, the WSP (rather like a car's ABS) brakes it in a succession of tiny slides alternating with freely speeding up again. That's meant to "wipe" the wheel and rail on each other, as well as preventing any localised flatting. I guess there is a minor point left: whether the there is a delay for this to have effect that tread brakes or scrubbers would avoid.

There was also this assessment of the available braking:

81 The fact that the WSP indicator lit up as soon as the brake was applied in the lowest setting indicates that the friction available between the wheels and rails was too low to support even the initial brake application (ie indicating a coefficient of friction less than 0.019).

The OTDR^▸ (On Train Data Recorder) showed a deceleration of less than 0.1g - even with all this clever WSP stuff in action. Not much, to stop a train.