What puzzles me is why Network Rail, as infrastructure supplier, seem to have no obligations to specify the state of the infrastructure that they are providing. For instance, a guarantee of the minimum co-efficient of friction of their railhead.
The condition of the tracks is neither a constant factor nor can it be determined comprehensively in real time.
This is my understanding of some of the issues surrounding adhesion and the responsibility for associated delays, although anything involving delay attribution is rarely straightforward or simple:
The responsibility for performance issues caused by low adhesion (freight trains slipping to a stand, or passenger units delayed by defensive driving necessary because leaves are causing poor braking) is usually determined by whether railhead conditions are considered unusually bad due to "railhead contamination" - I think the precise wording for this state of affairs has recently been changed from "exceptional railhead conditions" to "reportable railhead conditions." If a train slips to a stand due to very poor railhead conditions (rather than a fault with the train), the driver would report that to the signaller and it would be an important factor in determining the responsibility for the delay. As an example, if a Railhead Treatment Train was supposed to operate on a route but didn't for some reason, and the resulting poor railhead conditions cause delays because a train slips to a stand, Network Rail would be responsible for those delays.
This publication from the Rail Delivery Group "Managing Low Adhesion" is quite interesting, and includes a table reproduced below:
https://www.raildeliverygroup.com/publications/13003-managing-low-adhesion-edition-7-0/file.html
| Adhesion Level | Typically | Description |
| High | >15% | Clean rails wet or dry |
| Medium | 10-15% | Damp rails with some contamination |
| Low | 5-9% | Typical autumn mornings due to dew / dampness often combined with light overnight rust |
| Exceptionally Low | <5% | Severe rail contamination often due to leaves but sometimes other pollution |
Based on this scale, Network Rail do have a sort of "minimum co-efficient of friction" they are supposed to provide - I think the general idea is that they are responsible for taking preventative measures to avoid "exceptionally low" adhesion levels (which are a safety risk in addition to a cause of delays).
Adhesion issues caused by typical wet & windy Cumbrian weather in winter (outside of leaf fall season) are unlikely to be considered "exceptional" and it's the responsibility of the freight operator to ensure that the locos they use are capable of hauling their trains. A "fair weather only" loco is not much use, especially in Cumbria!
When pairs of class 90s were being used by DB on the heavy Daventry - Grangemouth service for Malcolms (1600 tons trailing) there would occasionally be times when it had to run with a single class 90. In that case traffic would be left behind to keep the train within the published weight limit for a class 90 (970 tons trailing up Shap and Beattock) rather than risk the train slipping to a stand - even though a class 90 could probably manage a higher load on a dry rail. If a single class 66 had to substitute for a pair of 90s, it could manage 1600 tons without risk of slipping to a stand, but would not be able to maintain the booked timings for a pair of 90s over Shap and Beattock.
The class 88s have an authorised max load on this route of 1400 tons trailing, a 44% increase over the class 90 but still relying on 4 axles to transfer the power to the rail.
A Modern Railways article on the class 88s has a telling sentence regarding adhesion issues on this route with these locos:
Just a couple of times the orange Catherine wheel flashes, meaning wheelslip, but it is automatically corrected almost instantly. It isn’t always this easy, witness a large bag of sand in the back cab.
https://www.modernrailways.com/article/class-88-over-shap
The Freight train loads book says the following:
The manual assumes standard train resistance and adhesion parameters. Train resistance can alter significantly due to atmospheric conditions such as headwinds. Adhesion can be significantly reduced by adverse railhead conditions.
...Seasonal variations may also be required in certain sections based on historic issues.
Autumn weather conditions sometimes require freight trains to be "downloaded" (i.e. have their maximum weight reduced). Network Rail monitor leaf fall (to the extent of having statistics on the % of leaves left on the trees in different areas of the country) and predict how bad conditions will be, due to factors like predicted winds bringing down more leaves. The conditions for different areas are colour coded depending on how bad they are expected to be, with "red" days being bad and "black" being the worst.
For example, trains from South Wales to Dee Marsh (Shotton steel works) can be affected by leaves on the steep climb north of Abergavenny (Llanvihangel bank, with a sustained climb of about 4 miles at 1-in-82 easing to 1-in-95). The normal load is usually 25 wagons of steel coil, but this can be reduced depending on predicted conditions, to 22 or 20 or even 16 on the absolute worst days. This mitigates against "historic issues" where trains have been known to slip to a stand in previous years. Stubbornly persisting with a load of 25 wagons during Autumn despite knowing the risk of slipping to a stand would not really be an option, especially considering that delays to other trains can cost upwards of £50 per minute of delay to each train involved, which can then snowball as those late trains delay more trains in turn, including their return workings and so on...
A quick glance at the delays to trains stood behind the failed 4S43 on 17th December showed at least 500 minutes delay even before any subsequent knock-on delays to other services are considered. Operators often have a "cap" on the maximum amount of delay minutes for a particular incident that they can be charged for, which works a bit like an insurance policy to limit the financial blow from any single incident.
The issues with 88s over Shap and Beattock are well known. I think the problem with using single 88s on 4S43 is that the timing load - the "
tonnage which a nominated train is deemed as being able to convey in order to maintain point to point running times" is evidently quite close to, and sometimes exceeds, the absolute maximum that the loco can cope with in poor weather conditions.
1400 tons is not a particularly heavy load, even up Shap, so having regular issues with this weight has to be considered as an issue with the type of loco used. Unlike a heavy steel train to Dee Marsh which normally runs close to its maximum possible weight, 1400 tons over Shap can be considered a moderate weight which a modern freight loco should be able to handle on its own all year round, including autumn and winter.
I think single 88s are more suitable for less steeply graded routes such as Tilbury to Daventry or Daventry to Trafford Park, but that geographical limitation seriously limits their usefulness as a freight loco when the WCML from London to Scotland is the main artery for electric freight. I suppose the underlying issue is that the freight operators as a whole don't currently have the right mix of traction for Anglo-Scottish intermodal, with the 88s and 90s having weight and adhesion limitations when used singly, and 66s being reliable but too slow up the gradients. It would be interesting to know how the class 93s would have performed with the same train in the same conditions - when doing a google search I read an AI summary (which appeared to be drawing on info from a paywalled article) which said that the class 93 wheelslip prevention system is better than the one on the otherwise similar class 88s. The new 99s should be great over Shap and Beattock, but GBRF don't currently have the contract for the trains from Daventry to Scotland - although having more reliable and capable locos will be a good bargaining chip when contracts are renewed.
