Does Network Rail have an over-engineering problem?

[J s]

One thing to bear in mind is precisely which variables you are optimising for. Once upon a time you optimised for material usage as that was where the cost of building lay. Labour was cheap. Nowadays there is much more focus on optimising for speed and minimising labour as they are much more dominant components of the overall cost of construction. If your design is chunkier but can be lifted into place in one go in one night then maybe that’s not such an irrational response?

The railway also has exceptionally low tolerance for risk and a desire to future proof which only promotes more robust conservative designs.

 

[GRALISTAIR]

This doesn't make sense - you surely have to balance the cost of building 1000s of masts strong enough to withstand the impact of a derailed train, with the cost of replacing one slimmer mast, in the very rare event of a train derailing.

Additionally, it is surely much preferable that an electrification mast distorts on impact, than the train body.

That’s what I was thinking and I am glad you said it first.

 

[The exile]

S

do they get the extreme swings we get though?

Over the course of time, yes, though I suspect the tendency to experience all four seasons before lunchtime drops as the climate becomes more continental. In terms of vertical structures, I suspect the toughest feature of our climate is wind - but even there, the difference is probably how often it is very windy rather than the severity of the occasional “extremely windy”

 

[DazT]

I understand that the design on the WCML that tried to address the apparent over engineering was not a success. The signal heads could be lowered to the ground to avoid working at height. I don't think they were ever installed anywhere else but happy to be corrected.

Any at height signals need a walkway that is strong enough to carry the weight of the equipment, the staff up there and what ever heavy tools you need so I think there's much choice!

Raised not lowered, crown post signals, but still require working at height, just that the signal head can be wound up out of position so it's easier to work on.

Crown

Crown UK, a subsidiary company of Avingtrans plc is a world-class supplier of advanced signal housing for rail track installation.…

www.railway-technology.com

On my patch, we have a load on the Tring workstation area, some around Hanslope and north of Polesworth (IIRC all the way to Armitage [former site of])

 

[osmarks]

Perhaps the question could be: are external physical systems necessary when we could put it all in the cab and/or automate it? Removes the maintenance liability and initial infrastructure cost; existing systems elsewhere in the world (thinking very high speed rail) probably already do this. (Note I am not advocating for one or the other.)

The standard for in-cab signalling here is ETCS, which is being installed on the south ECML, but it's proving absurdly expensive for unclear reasons.

 

[Nicholas Lewis]

Additionally, it is surely much preferable that an electrification mast distorts on impact, than the train body.

At Hatfield the OLE mast in the main path of the derailed train gave way and did significant damage to one of the coaches. So when you look at the scale of GWR OLE masts and the signal gantry structure above they will present a very hard structure should a derailed vehicle hit them unless the mast bolts are designed to shear if subjected to high lateral force.

Clearly the engineers also have to have structures that arent vulnerable to failure as that would be equally catastrophic so a balance has to be struck.

 

[RUK]

This doesn't make sense - you surely have to balance the cost of building 1000s of masts strong enough to withstand the impact of a derailed train, with the cost of replacing one slimmer mast, in the very rare event of a train derailing.

Additionally, it is surely much preferable that an electrification mast distorts on impact, than the train body.

I was thinking if the wires could come down and electrocute people.

 

[59CosG95]

In the case of the ECML masts, the slimmest of them (152x152 Universal Columns) have been more prone to Lateral Torsional Buckling than its thicker cousins (e.g. 203UC, 254UC, 308x305UC etc) so aren't available in newer design ranges.

 

[Sir Felix Pole]

I've noticed the same with regard to the electrification of the Welsh valleys: the masts supporting the wires are overbearing and very unattractive. Perhaps unsightly structures are cheaper to erect than elegant ones.

And don't get me started on the ugly footbridges that are appearing across the network...

I went through Abergavenny recently, and was astonished by the absolutely enormous footbridge / lifts recently erected there. It might be appropriate for somewhere like Cardiff Central but is totally over-engineered for what was an attractive country station - it is completely overpowering and dominates the scene. (I know there was an unfortunate incident with the old bridge where someone got electrocuted due to a loose cable slung along it, but even so).

 

[Lucan]

As a structural engineer, I can tell you at a glance and without doing any calculations that the Tulse Hill gantry and the OHL supports here in South Wales are ridiculously over-engineered.

 

[mcmad]

As a structural engineer, I can tell you at a glance and without doing any calculations that the Tulse Hill gantry and the OHL supports here in South Wales are ridiculously over-engineered.

Feel free to offer your services and PI cover to Network Rail for slimmer structures.

 

[Ashley Hill]

I went through Abergavenny recently, and was astonished by the absolutely enormous footbridge / lifts recently erected there. It might be appropriate for somewhere like Cardiff Central but is totally over-engineered for what was an attractive country station - it is completely overpowering and dominates the scene. (I know there was an unfortunate incident with the old bridge where someone got electrocuted due to a loose cable slung along it, but even so).

Future proofed against any future electrification this century?

 

[GRALISTAIR]

As a structural engineer, I can tell you at a glance and without doing any calculations that the Tulse Hill gantry and the OHL supports here in South Wales are ridiculously over-engineered.

Thank you for your input.

 

[Harpo]

Other countries seem to be able to build gantries and masts that blend into the scenery and don't fall over.

What do the health and safety laws look like in those other countries and are they doing maintenance from the structure or by using a ladder from the formation?

 

[DarloRich]

The other thing thats changed today is British Rail used to have the design staff but its all outsourced now

that is not correct. NR have design people.

 

[BRX]

What do the health and safety laws look like in those other countries and are they doing maintenance from the structure or by using a ladder from the formation?

The earlier much slimmer design (as shown in posts 15 and 29) also catered for people on top of it. The need to support a person or two can't explain the amount of structure used on these recent examples.

 

[Harpo]

The earlier much slimmer design (as shown in posts 15 and 29) also catered for people on top of it. The need to support a person or two can't explain the amount of structure used on these recent examples.

Agreed. It was more the point that direct international comparisons rarely work, such as some other nations OHLE costs not including road bridges etc..

Someday someone will ask why we built enormous expensive structures just to suspend a lightbulb where a driver can see it.

 

[sharpener]

The signal gantry is on an "embankment" adjacent to the south circular and the ground conditions are very poor with minimal flexibility in locating the singal for signal spacing reasons. The foundations are two huge steel piles (the largest diameter used for problem OHLE structures).The local ground level is about 4' below the track level and the ground the piles are outside the wing wall.
The is also a singal on the other side of the gantry along with another speed sign (mounted between the gusset and the hidden signal). There is no space here for any ground mounted signals let alone fold down signals and the speed signage.

None of that explains why the main beam has a constant cross-section. This must make it much heavier than necessary, contributing to the need for the massive foundations. I see the same thoughtless over-engineering with motorway signs too, with their vast circular supporting towers - which are not even a good choice of cross-section for an eccentric load.

Here's a slightly earlier (I think) vintage of gantry which appears to be doing much the same job as the one in my OP.

@BRX's example is of a much more elegant tapering structure where the cross-section is appropriate for the loads at every point on the walkway. As we were taught to do in structures design lectures 60 years ago, none of this is new.

 

[DarloRich]

Surely this is about simply having standard items that can be bulk manufactured and installed in the same way- that has a cost bonus.

(That is not to say the bulk design cant be "nicer" but if it is a one size fits all then you have to fit all!)

Here's a slightly earlier (I think) vintage of gantry which appears to be doing much the same job as the one in my OP.

i doubt it would meet modern working at height regulations - we cant see how thick the cross beam is for instance

 

[BRX]

Surely this is about simply having standard items that can be bulk manufactured and installed in the same way- that has a cost bonus.

(That is not to say the bulk design cant be "nicer" but if it is a one size fits all then you have to fit all!)

I can see this makes sense for some things like OHLE stuff but I'm not convinced that this applies in this case - I keep an eye out for other structures (and there are quite a few new signal structures around south london from the Victoria resignalling programme) but don't see a great deal of repetition. There's actually another gantry at the other end of Tulse Hill station for example; it is a totally different design with a kind of trussed beam instead of a solid one.

As a side comment ... it's clear to me that what NR does at stations is often from a kind of standard "kit of parts" and I understand at least the theoretical cost benefits. But I see loads of examples where it simply results in really bad, clunky design, and things that take up much more space than they need to. In at least one place I've seen a platform shelter, open on one side and glazed on the other three. It's then plonked on the platform and I assume someone decides that there needs to be access around the back of it to clean the glass on the back. So the platform needs to be locally widened behind the shelter and this is done on a platform that is already part cantilevered on an embankment. So a whole load of extra structure is needed to create this widening. And all of this lets passengers wait in a shelter with a glazed back that looks across an 80cm wide gap to a high fence (above eye height and standard issue Network Rail prison style). The sensible design here is surely a shelter the back of which is formed by the wall/fence that needs to be there anyway. There's no need for the glazing and no need to provide access to clean it.

i doubt it would meet modern working at height regulations - we cant see how thick the cross beam is for instance

What do you mean by the cross beam?

 

[DarloRich]

I can see this makes sense for some things like OHLE stuff but I'm not convinced that this applies in this case - I keep an eye out for other structures (and there are quite a few new signal structures around south london from the Victoria resignalling programme) but don't see a great deal of repetition. There's actually another gantry at the other end of Tulse Hill station for example; it is a totally different design with a kind of trussed beam instead of a solid one.

it could be as simple as having one on stock to use up or that being the type that was available in the delivery time scale/cost for the project. There could be some hidden engineering reason but never discount the simple option!

But I see loads of examples where it simply results in really bad, clunky design, and things that take up much more space than they need to.

your example sounds like a classic left hand/right hand thing! Someone has chosen the wrong item from the design catalogue.

Not just railways but I look at modern buildings and think - has anyone thought how that part is going to be cleaned or repaired? You see it a lot with glazing and fancy shapes to roof lines etc. Flats are terrible for this!

What do you mean by the cross beam?

The part holding the signal head. I am not an engineer so wont be using the right terms!

 

[sharpener]

As a structural engineer, I can tell you at a glance and without doing any calculations that the Tulse Hill gantry and the OHL supports here in South Wales are ridiculously over-engineered.

It is not my professional branch of engineering but I have always had an interest in bridge design, and all the classic forms look "right" because not only is the material in the right place but there is no material in the wrong place.

Glad to have my impressions of this dreadful looking gantry confirmed by someone who really knows.

 

[Coleridge]

What do the health and safety laws look like in those other countries and are they doing maintenance from the structure or by using a ladder from the formation?

I would imagine that Switzerland has some of the strictest health and safety laws in the world, as well as some of the least obtrusive lineside hardware. But they also have a general ethic regarding the appearance of their country, on the railway and elsewhere, that we don't have.

I'm afraid I have no idea whether the Swiss use a ladder to carry out maintenance.

 

[AlterEgo]

I would imagine that Switzerland has some of the strictest health and safety laws in the world

Switzerland has a very different H+S culture than you imagine.

 

[yoyothehobo]

I mean its two standard steel I sections with some steel mesh footway between them, the signal heads themselves are attached to both beams and are underslung with an area to work on the signals. The main stantion as has been suggested is build off ground quite a bit below rail level, with that ground being very poor in nature.

The structure seems well designed to cater for the twisting effects of wind load and that there is a long cantelever arm.

Its a simple design, it does the job and is made mostly of standard manufactured parts.

 

[Sorcerer]

I would imagine that Switzerland has some of the strictest health and safety laws in the world, as well as some of the least obtrusive lineside hardware. But they also have a general ethic regarding the appearance of their country, on the railway and elsewhere, that we don't have.

Switzerland's railway has very little lineside fencing, and if there is any then it's only small fences that most adults could easily climb over. I rode the network for a week in early September and often times there were footpaths and cycle paths right next to the running line. Over here that would drive the railway health and safety inspectors mad.

 

[sh24]

It is not my professional branch of engineering but I have always had an interest in bridge design, and all the classic forms look "right" because not only is the material in the right place but there is no material in the wrong place.

Glad to have my impressions of this dreadful looking gantry confirmed by someone who really knows.

One of the best engineers I have worked with always said - if it looks right, it probably is right.

 

[quantinghome]

TBH I'd only be concerned about inefficiency if this was a widespread signal gantry design being rolled out over the whole railway. However, it appears to be a one-off for whatever sound or unsound reason, so quite frankly, who cares?

EDIT: If this is indeed a bespoke design, a tapered beam would almost require one-off production and hence would be very expensive, whereas a uniform section beam would be an off-the-shelf product and hence the cheaper option, even if it is structurally inefficient. The examples of tapered gantries shown upthread are almost certainly mass-produced as part of a large signalling scheme.

 

[yoyothehobo]

Switzerland's railway has very little lineside fencing, and if there is any then it's only small fences that most adults could easily climb over. I rode the network for a week in early September and often times there were footpaths and cycle paths right next to the running line. Over here that would drive the railway health and safety inspectors mad.

Maybe in rural areas, however my recollection is of a lot of large concrete barriers, predominantly for noise purposes which look far worse than this whole "over engineered OLE"

 

[GRALISTAIR]

Material in the right place means away from the neutral axis!