Welcome to the very first “Back to the Drawing Board” with Steve Lloyd at Groundforce Shorco technical offices in Leeds. Today, Steve will be explaining how we design our hydraulic props and the kind of actions and effects that act upon it.
Most of the prop make-up is Circular Hollow Section (or CHS) with a hydraulic unit tucked away at one side of the prop that allows the speedy installation, the fine adjustment and the simple pre-stress that is synonymous with hydraulic shoring equipment.There’s a good chance that you already know what a hydraulic prop is, in which case carry on. But if you don’t, you’re in luck – below is a link to Tony’s technical blog and an entry from way back in March 2011 that talks about the basics of a hydraulic unit. So hit the pause button, go have a quick read and come back in 5 minutes.
Read the blog post...
The reason we bother to put these large, temporary steel structures in the ground at all is to hold back the earth and give us a safe space to work in so that the permanent works can be built. And that’s where we will start… Earth load is definitely the biggest of all the forces that acts on a prop. These can be calculated using theories from Rankine, Coulomb, Peck and Terzaghi – some dating back to the 1800’s. Or in more modern times it has become common, and even preferable to use numerical analyses based on detailed computer modelling.
However you come up with your earth load, it will primarily depend on three things: the depth of the dig, the shear strength of the soil and the presence and depth of groundwater. I won’t go any further into Earth pressures now, other than to comment on the Eurocode 7 design approach. We use DA1 in the UK, which means we check the dig three times using different combinations of partial factors and geometry to make sure that everything is safe. As a prop designer, I can then take the worst case generated by these three combinations and apply this to the design of the prop.
The next big action hitting this prop is thermal load. We have pretty big lump of steel here and when it gets warm, it wants to expand. The only problem is, the prop sits between two retaining walls that really don’t want to let it expand. What we get of expansion is an increase in stress.
A good way to picture this is to imaging that the prop is allowed to expand; and then work out (using stiffness equations) how much force or stress it would take to squash the prop back to where it started. And that’s basically how much thermal load you have.
Even though the prop is designed to resist a very significant axial load, not all of the actions upon the prop are axial. The main transverse load for our prop is its own self-weight. In fact, these props can weigh quite a lot. A 15m long MP250 weighs somewhere in the region of 6 Tonnes… but I suppose that’s not a lot in comparison to the 250 tonne axial capacity. But what this transverse force does is induce a moment along the length of the prop… good old WL squared over 8!
This is why tubes are hollow instead of using much smaller diameter solid sections. The moment of inertia of the larger hollow section is so much higher due to its wider geometry. A small solid section would be far too flexible and just wobble about all over the place – it would act more like a cable that a strut.
It’s not just the transverse actions that cause a moment in the props. We talked about the axial forces, but these might not act exactly as we hope. If the axial forces push the prop from a slightly eccentric position, you get a bit of a bending effect induced. And the bigger the axial load or the bigger the offset, the bigger the moment becomes.
To a degree this is mitigated by the fact that the props have swivel pins at each end. But these only cancel out eccentricity in the plane of the swivel. At Groundforce, we always calculate assuming an eccentric moment is generated in the same direction that the self-weight acts. i.e. downwards. This gives the most onerous case. You can get rid of this eccentricity, however by introducing an interface that ensures a perfect fit. Embedding into a grout pack is the most common method. The grout will squeeze and mould to the exact shape and ensure that the load is uniformly transferred.
Those are the common things that you expect to act on a prop. But what about the things you don’t expect? Eurocode expects you to design for any accidental conditions. In the case of a steel prop you have two options.
Option 1 – design each prop in the system such that if its neighbour were to be damaged and cease to do its job, the remaining props would still have enough strength left to take the extra burden. In the case of long wealer runs this is reasonably simple – design each prop to take half the load of the adjacent prop in an emergency. It’s worth noting that partial factors on accidental cases are not applied, so the load case is not so onerous as it might seem. But how do you deal with more complex shaped excavations where the loss of any single prop could be difficult to account for without generating a very uneconomical design?
Well that’s Option 2 – design each prop such that it is robust enough not to completely fail, even if it were struck with some considerable force – e.g. a load dropped on top of it, or an excavator bucket swinging into it. In this case, we apply a theoretical accidental load in the most onerous position and direction and design the prop to prop so that it can withstand it without buckling.
But you don’t just throw all of these forces at the prop willy-nilly. You apply actions or effects in logical combinations and apply various factors to them to account for the varying amounts of uncertainty. I’m not going to go through every single one, but they are basically: CSI; PSI; GAMMA factors.
And the value that they take can depend on: what sort of action you are dealing with; permanent or variable – variable action are more uncertain so the gamma factor (generally 1.5) is higher than that for permanent actions (usually 1.35).
Also, the source of the action; wind, snow, traffic, people. I talked about combinations, and this is where PSI factors come in – you have to consider the source of the action and the probability of them acting at the same time. By applying combination factors it provides a more realistic probable value. You also have to think about illogical combinations and eliminate them – a good example of this is snow load on a prop. I told you earlier that thermal loading is significant on a shoring prop. And snow load can apply a transverse force and act to buckle the prop. But it’s unlikely that you’ll ever get both acting at the same time since high temperatures would completely preclude the chance of having snow.
The country you are in also has an impact; EC allows each nation pick their own values that are specific to them. This can be a bit confusing, but it makes sense really, since the chance of snow on the southern coast of Italy is much smaller than, say, the mountains of Scandinavia. So the factors associated with those counties would reflect that.
In the UK we have a cheeky little pair of equations 6.10a and 6.10b. 6.10a lets us reduce the factor on the leading variable using an appropriate PSI factor and we keep the 1.35 factor on the permanent actions. 6.10b lets us reduce the permanent loads using a CSI factor in exchange for maintaining a higher factor on the leading variable action by omitting the PSI factor. Although except for us, pretty much all other Eurocode countries use equation 6.10 (without a letter), which doesn’t reduce either the permanent actions of the lead variable.
After all of that we’ve only got as far as determining what values of actions and effects to design our prop to – we haven’t even started looking at the prop itself. This is the second half of the process – determining the resistance of a member. Simply put, our props comes in two parts – we said this at the very beginning. There’s the hollow steel part and then there’s the hydraulic part. Let’s start with the hydraulic part.
The primary modes of failure for the hydraulic units are: failure of the connecting pins at the end; bucking of the piston rod; failure of the seals and O-rings and yielding of the cylinder. And it is the cylinder yielding that provides us with our onerous check so this is generally what the hydraulic resistances are based on. (of course, as with any moving parts, mere calculation isn’t enough and the rams themselves are physically tested to pressure far in excess of what would be experiences during it’s working life).
This leaves us with the steel. As with any long member, it’s not the section resistances themselves that give us our limiting strength; it is the buckling resistance. In the case of a CHS, we can completely ignore any torsional buckling modes. But the longer you make something, the more inclined it is to flexural buckling. And the length at which the buckling effect starts to get critical will depend on the cross sectional properties of the prop.
Thicker walls and larger diameters will start to buckle at much longer lengths. You just have to accept a certain amount of strength reduction, but when that goes too far, you just pick a bigger section.
I think that’s about as much detail as I can go into in the time I’ve got. It’s difficult to cover everything in a ten minute video and there are a few subjects that came up that could easily do a whole video of their own.
I’d love to talk more about calculating earth pressures; go into more details on the different load combinations in Eurocodes; the design and testing of the hydraulics would be an interesting one; and the Euler buckling equation and how it is derived would be a great mathematical topic to cover if you like a bit of algebra. But for now: thank you for watching and I hope you’ll keep coming Back to the Drawing Board.