Surrey Quays Suspension

The problem is yesterday a driver reported a wrong side failure relating to the AWS for signal number EL262.

A wrong side failure is defined as something that makes the railway more dangerous rather than less.

For example, a track circuit which the signalling system thinks has a train on it, when there is actually is no train there is a 'right side' failure because the signalling system has protected the failure by holding signals at red, preventing points from moving, etc.

Now consider a track circuit which the signalling system thinks is free from trains - yet it actually has a train on it. This is a 'wrong side' failure as the train is not protected by the signalling system - and as Clapham Junction in 1987 proved, death and serious injury are very likely as a result if immediate action is not taken.

In the wrong side failure on the ELL, we are talking about an AWS telling the driver the signal ahead is at green when in actual fact it is at yellow or red, this could in a worst case scenario cause a train to collide with something else (readers may which to search the RAIBs website for a report into an incident at Wooton Basset near Swindon to appreciate the importance of correctly working AWS equipment).

Now while I cannot comment on the specifics of this case (because I don't have all the facts), a guide to 'wrong side' failure testing may bad helpful for forum readers.

Testing in the case of 'wrong side' failures is done in accordance with a strict procedure, designed to pick up ANY flaws in the signalling system. This is laid down in the "Signal Technicians Maintenance Handbook" (SMTH) which gives details of which tests are required and more importantly what level of competence is needed before the piece of equipment against which the allegation has been made can be returned to service.

The actual checks will be carried out by 'level 1' (usually first line S&T response staff) testers with initial checks like wire counting (making sure ALL relevant equipment (which includes the controlling realys, transformers, etc) has the correct number of wires terminated on them), various visual examinations, checks for moisture ingress, doors securely locked, etc. take place with the results accurately recorded - this is known as the 'non destructive testing' phase.

In the 'destructive testing' phase, items of equipment and line side cables are checked for continuity and insulation properties and the presence of spurious voltages that shouldn't be there. In the case of a 48 core lineside cable this may have to take place overnight because of the number of separate bits of signalling equipment that are fed from that one cable. The results from this testing will show whether there is moisture ingress to the equipment, whether there is a short circuit between cable cores or whether a cable core has become connected to earth through damage (ALL signalling circuits MUST be totally independent and NEVER earthed - due to the potential of two earths to cause electricity to by-pass relay contacts and cause a disaster).

The results of both sets of tests are then forwarded up the chain of command to the level 2 tester  (usually S&T depot manager level) who reviews the tests and decides whether there is anything more that still needs doing.

However because of the serious nature of 'wrong side' failures, in most cases the SMTH requires the level 2 then has to go up another level to a person with a level 3 (usually area S&T engineer level) compliance who reviews the results and if they are happy it is the level 3 that gives permission for the equipment to be returned to service.

Thus, while I don't have the exact specifics to hand it sounds like because of the wrong side failure allegation yesterday, the capacity of the signalling system was reduced (e.g. A signaller was, say, having to caution every train past the signal before EL262 (because the AWS on EL262 could give a green AWS indication when the signal is actually at red). As a consequence the train service was reduced and amended to make things easier for the signallers and testers.

Now until ALL the evidence has been reviewed and the appropriate level 2 or 3 tester and they are 100% happy with what they have found, the revised service has had to stay in place. It is quite possible that some of the testing could only had done overnight (say a couple of 48 core lineside cables need fully testing affecting multiple signals and track circuits) and not all of it was completed, or alternatively maybe a problem was discovered and a piece of kit needed to be changed (note the decision to do this comes from the level 3), or maybe its a issue between the offical diagrams not agreeing with what is installed on site or situation of simply too much to test and too few qualified staff.

In such situations it's easy to see that the WSF investigation may take some time to complete - however until that investigation is complete and the signalling can be PROVED to be in perfect working order, the equipment the allegation relates to must remain out of service - regardless of how much delay it causes.

Normal service now seems to have resumed, so it seems the issue has been sorted.

Yes from the S&T side all relevant logging systems will be examined during the non-distructive testing phase, along with things like train describer data, platform CCTV images and any relevant voice communications between the signaller and the driver. (Note the S&T equipment includes EVERYTHING used in the signalling of trains and includes the signalling panel / workstation in the signal box as well as interlockings and the trackside kit)

However it is important to note that logging can only get you so far. In the case of an allegation of a signal displaying the wrong aspect on a SSI area for example, the logger might tell you what aspect the signal was being commanded to show - but it cannot actually confirm what lamp was lit. For that you need a forward facing CCTV image because you MUST NOT discount the possibility of a defect in the cable from the module to the signal head, a wiring issue in the head or a 'great train robbery' style tampering with the aspects has taken place. As a general rule the further up the chain (or the closer to the 'brain') the logging stops the more equipment and cables you have to test - but with relay interlocking there is a limit on just how much you can practically monitor - as it is all retro fitted years after the original signalling was installed. SSI has the logging built in so is easier when it comes to wrong side failure testing but even then it has its limits.

It should be noted however that in at least two cases in the past 5 years, the wrong side failure has been traced to a data defect in the SSI interlocking that had somehow got missed when the data was being tested before commissioning and the defect had been there for quite a while - it was simply luck that the particular circumstances that triggered it to show up had not occurred before.

As for the OTMR data, this will usually be requested from the TOC as part of the investigation, but the download and interpretation of the data has to be done by the TOC with the results communicated to the S&T level 2 or 3 tester.

S&T = Signals and Telecoms.

My question is which is signal EL262?

When I saw the suspension yesterday it was Surrey Quays to New Cross northbound only. The only thing I can think that would cause that would be something that allowed northbound trains to travel ECS only but not in passenger service - otherwise you would very quickly end up with an overabundance of trains at New Cross!

Indeed

The only involvement S&T have with telecoms are lineside direct dial telephones, the cable that connects them to the nearest location case, plus modems directly connected to signalling equipment. Inter location case lineside cabling, telephone concentrators and suchlike are the responsibility of a dedicated Telecoms department.