Does Network Rail have an over-engineering problem?

[BRX]

Here's a gantry near Denmark Hill. I think this is one of the older style ones - but has been fitted with the new LED type signal.

I wondered if part of the explanation for the structure of the new one at tulse hill was the access platforms, behind the lamps, which project outwards from the main arm. I could see that maybe this would introduce some twisting forces. But it looks like this older design also has such platforms.

 

[LickeyIncliner]

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.

Apologies for coming late to this discussion. I entirely understand that poor ground conditions require chunky foundations, but isn't that also an argument making the structure above them - the signal gantry shown in the OP - as light as possible? To put it politely, it doesn't look like the lightest possible design...

 

[BRX]

I entirely understand that poor ground conditions require chunky foundations, but isn't that also an argument making the structure above them - the signal gantry shown in the OP - as light as possible?

Yes!

 

[domcoop7]

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.

I've always thought the new motorway gantries (the cylindrical ones) are ridiculous. I am not a structural engineer, but I was recently looking into loft conversion and also wondering about the weight of our box room (which is above the front door and goes across the top of the stairs, so as you walk upstairs you go under that room. The joists on that room go paralell with the stairs, so in other words, they don't span between the two brick walls but run in the same direction and end on one (wooden) beam that itself goes across the stairs). It's a 1930s semi, and the wooden beam is only a few inches thick and 6-8 inches deep. And it holds the weight of the joists themselves and circa 500kg of cupboards and wardrobes and clothes plus another 200kg of people if three of us go in there.

When looking into the loft conversion, the default Bob the Builder methd is to put a huge overengineered RSJ craned in from the side and build a floor on top of it. But another technique was to sister the joists on the ceiling by doubling up the ceiling joists and putting steels in to bridge gaps. I was told a 10mm thick and 200mm wide flat piece of steel weighing only can support a load of over 11kN which if I understand correctly is over a tonne. In normal use with the maximum expected static load of a habitable room that spec over a 3 metre span would deflect under weight by up to 1mm. If someone did put a 1 tonne weight on the middle of the span it would deflect by 7.5mm. So to make sure and allow adequate clearance they specified 15mm thick steel. The bar itself weighs under 10kg. I can now safely have elephants in the loft if need be.

Now obviously my steel bar is not self-supporting or cantilevered, but the picture in the OP looks like the sort of thing that would hold a crane up at a docks that could lift trucks and containers in the air! Surely there must be a reason for specifying it?

 

[quantinghome]

I don't think there's any doubt that a tapered section would be a more efficient use of material and could have worked in this case. The question is whether a suitable tapered section is a standard product or bespoke. I suspect it's the latter, in which case the oversized but standard uniform section would probably have been cheaper.

 

[sharpener]

But another technique was to sister the joists on the ceiling by doubling up the ceiling joists and putting steels in to bridge gaps.

A useful dodge for loft conversions is the flitch beam, a sandwich of two timbers with a thin steel plate between them. The steel carries the load but the timbers prevent it twisiting or buckling. Our loft conversion ppl used it in one place where there was an awkward gap to span. There is also a ridge beam but apart from those they avoided steel in the construction.

== Doublepost prevention - post automatically merged: Today at 09:14 ==

I don't think there's any doubt that a tapered section would be a more efficient use of material and could have worked in this case. The question is whether a suitable tapered section is a standard product or bespoke. I suspect it's the latter, in which case the oversized but standard uniform section would probably have been cheaper.

Yes, you can't easily roll a tapered steel beam with flanges! But you quite often see a beam which has been flame cut on the diagonal and welded back together the other way round. And castellated beams made like this which makes them deeper and at the same time moves material away from the neutral axis, which @GRALISTAIR would approve of! It must be cheap enough to do or ppl would not bother.

 

[hwl]

Apologies for coming late to this discussion. I entirely understand that poor ground conditions require chunky foundations, but isn't that also an argument making the structure above them - the signal gantry shown in the OP - as light as possible? To put it politely, it doesn't look like the lightest possible design...

I can't understand that either!
The key differences with the Denmark Hill example is that Tulse Hill example has two speed signs and the signal not visible in the photo in the 1st post on the reverse side is a banner repeater. Given problematic local lighting conditions (sun a certain times of day) they may have wanted some of the signals to appear against a large solid black background which that overly chunky beam section provides.

 

[WAO]

The BR structures are elegant because the cantilever has been designed to reflect the shear stress and bending moment diagrams. It is therefore tapered - not expensive to fabricate. Because a slim beam is quite sufficient, it requires less bracing and a slimmer post, all of which is both economical in materials and construction cost - it is still in a sense "over designed" in that it will have quite a high safety factor over design.

The Tulse Hill example has a large fixed section cantilever which is unnecessary and ADDS to the foundation problem in difficult soil conditions. I would still not criticise the designers too much as if external will likely have all sorts of risk/life stipulations in their contract. You get what you ask for - and pay for!

WAO

 

[quantinghome]

Yes, you can't easily roll a tapered steel beam with flanges! But you quite often see a beam which has been flame cut on the diagonal and welded back together the other way round. And castellated beams made like this which makes them deeper and at the same time moves material away from the neutral axis, which @GRALISTAIR would approve of! It must be cheap enough to do or ppl would not bother.

Of course non-uniform steel sections exist and have established processes for production. But you've still got to go through that process and if it can't be bought off-the-shelf or mass-produced as part of a major scheme to give economies of scale, it's going to be very expensive.