@aleggatta: Apologies. I fear that I may have been incorrect in regard to one detail which could have resulted in my previous post being misleading.
In my own defence, it has taken me several days to find a photo showing sufficient detail of the coupler heads to see precisely what is going on with the Delners in use on the Mk5s. In the end I found one showing Mk5s in the shed on the
RevolutioN Trains website, which I have reproduced below. (The full resolution photo can be accessed via the link and scrolling down to the relevant entry.)
From this photo it would appear that there is a star valve only on the top air port but not the bottom one. I presume that this top port is the main res pipe and the lower one is the brake pipe, and that the lack of a star valve on this connection means that any air contained within it would vent should the connection be broken. This being the case, the brake pipes should vent in the event that a train ever parted, which would automatically initiate an emergency brake application on both portions although the pressure in the main res pipe would be preserved. I have still not been able to find a corresponding photo of the loco-mounted coupler head to verify whether or not it is similar, but I suspect that it must be.
To correct my earlier post, this would mean that it would make sense for the procedure for coupling and uncoupling to remain as per current loco practice; in other words, the air supply cocks should be closed when uncoupled and opened when coupled.
I only make this point, because although this might be a 'procedural' issue, it might be deemed to warrant a 'technical' response to mitigate against procedure failing again? I'm not sure any member of the public would be happy to be told 'someone forgot to do it once, so we told him off and he doesn't do that job any more'. If someone forgot, it means someone else can also forget.
As I've said already up-thread, I do not believe that a technical solution is warranted in this situation. Based on the evidence to hand thus far, it is my own opinion that this incident was caused by an isolation cock in the wrong position, for reasons that the RAIB will establish, which caused the main air supply from the loco to the train being isolated. The primary reasons for this are likely to be a) unfamiliarity with the new rolling stock and it's unusual coupling method, b) incorrect or inadequate procedures for coupling and uncoupling this stock, or c) human factors that resulted in a member of staff making a mistake with the positioning of the cock and/or failing to check that it was correctly set. These are procedural issues that can be addressed by improvements to training and assessment of staff competency and tightening up procedures on the depot as well as out on the route.
Could there be a repeat of this event? While we continue to use Human 1.0 there is always the possibility, even where we increase the use of automatic safeguards. But the railway industry is getting better at understanding human factors and designing procedures that reduce the risk of incidents. And the answer is not simply to identify the culprit and remove him/her from duty, but rather to identify what went wrong, learn the lessons and take appropriate remedial steps. Personally, I'd rather have someone who has made the mistake, learned the lesson and is now more vigilant as a consequence than some newbie who is still gaining knowledge and confidence.