Thermal Axial Loads / Foundation Design

[Redtelis]

So, I do not have experience on the design of structural elements exposed to environment (thermal effects etc).

I am currently looking the design of a steel raised walkway and the idea is that's its going to be restrained at the supports therefore no allowance for horizontal movements (i.e. bearings). As a result of the restraint supports, thermal effects to the steel members will be induced (axial loads).
The primary walkway steel beams are to be supported directly on pad foundation to transfer the loads back to adequate bearing strata. The connection between the steel beam and the pad foundation will be a base plate resin anchored/hd bolts to transfer the forces from the steel to the concrete.
My question is about the forces and what kind of actions are applied to the pad foundation for the design. We will have the axial load to the base plate which is the vertical load of the walkway and then am I right to say that we have the thermal axial loads as horizontal/shear loads? If yes, does the pad foundation requires to be designed for this horizontal/shear load? I am asking because the thermal axial loads are huge (2500kN) and I have the suspicion that I got something wrong in my thinking....

 

[Redtelis]

3d visual

 

[jayrod12]

Depending on the expected temperature swing, I'd expect that the steel columns and the pad foundations will be flexible enough to just accommodate the thermal movement internally. The only thermal movement seem by the columns and therefore foundation is for the portion of steel structure between the columns based on your screenshot. Run the numbers and see what is the expected total movement for the maximum temperature swing. And then make a call at that point whether it warrants further analysis.

 

[dik]

what is the length of the structure? and what sort of temperature range?

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[Redtelis]

The o/a length is 25m and the temperature range is about 52 Celcius

 

[1503-44]

You must allow thermal movement somewhere to eliminate that thermal force. It is typically done by allowing one, or both ends to move on slide bearings. If you let both ends slide, there will be no thermal force anywhere; only friction force to worry about.

 

[jayrod12]

Based on your screenshot, it would appear then that the distance between the columns in around 8m. At a temperature change of 52 degrees Celsius that equates to about 5mm of total shrinkage/expansion between the columns assuming it is installed at either the maximum, or minimum expected temperature (which would be a conservative assumption). In my eyes, there will be enough flexibility in the columns, connections and foundations to accommodate this movement without looking at it further. The columns will just slightly bend towards or away from each other to accommodate the movement.

If installed somewhere near the middle range of the temperature, then you're only looking at 2.5mm of movement total from installation. That would require each column to only move 1.25mm. I can only imagine they're flexible enough to accommodate that.

 

[JoshPlumSE]

If installed somewhere near the middle range of the temperature, then you're only looking at 2.5mm of movement total from installation. That would require each column to only move 1.25mm. I can only imagine they're flexible enough to accommodate that.

I didn't run the numbers, but I was thinking the same sort of thing that Jayrod said here. Another aspect of this is to think about any bolted connections and any slip that might occur in those bolted connections. My tendency is to say that we (as engineers) tend to make too much out of thermal loads for a normal range of ambient temperatures.

 

[1503-44]

Its not the columns that are the problem, as everyone has already said above. Its the ends. There's a good chance that the thermal force will move, bust, or buckle beams, or break whatever is attempting to stop it. Maybe only a few mm, but move, or break, it will. So do allow a small gap for that movement.

I don't know the cross-sectional area of your 25m long beams, so I can't tell you if your 2500kN force is correct. I also do not know the low, high, or the specified installation temperatures. I suspect something may be wrong, because if installation temp is at 50% of range, it would take around an area of 43,225mm2 to reach that 2500 kN force. That seems like a large x-sect area of steel. I estimate that thermal force should be close to beam x-sect area x 58MPa, if you do not allow any thermal movement. I think that force would move the abutments, or buckle your beams, whichever came first. I do not see sense in stopping all movement, but maybe you have a reason I can't yet understand.

When I designed conveyor belt towers for wood chip and paper Mills, we always provided slotted connections to keep thermal loads off the towers. If you do that, you reduce the overturning moment to only the friction load. Just the friction load at the top of a tower created a pretty high overturning moment on the foundation. A full thermal load would not have knocked the tower over, since tower deflection was small, but it would have exceeded the max soil bearing, perhaps off-centering the tower.

 

[canwesteng]

Conveyor belt towers, or bents, are a good example. They are not normally designed with slide bearings, the columns are just allowed to deflect. Personally reviewing the sketch, those columns are quite stout so I would let these slide, or put some thought into what force is require to deflect the columns. If you are just designing the pads below the bents, you should look at how much rotation you need in the foundation (assuming fixed base columns) to allow for the columns to move 5mm.

 

[1503-44]

I guess you are thinking like 1 tower? A chain of conveyor towers cannot be allowed to deflect, because that deflection would be cumulatively passed on to the next tower through the bridge and so on, until possibly creating too much movement and eccentricity at some tower N. Handling each bridges expansion via sliding gaps at the end of each conveyor bridge on all towers allows all to remain almost perfectly vertical.

 

[JoshPlumSE]

There's a good chance that the thermal force will move, bust, or buckle beams, or break whatever is attempting to stop it. Maybe only a few mm, but move, or break, it will. So do allow a small gap for that movement.

Hence my point about whether you have bolted connections or not. Thermal expansion is always a self limiting load. So, often there is enough "slop" in the bolted connections that you will get a little bit of movement there and nothing the forces shown in your analysis model will never develop in the actual structure.

 

[canwesteng]

Well at some point yes, you need to make up the movement and reset. Will depend on how much movement you can tolerate at each bent, but it could 200, 300 feet. I think it's more economical to use bents between the end spans, but it's all a matter of opinion I guess.

 

[1503-44]

It is easier to handle it at each tower. That way no thermal force greater than friction can develop at any tower and all bridges of roughly equal spans can be designed similar, without the need to consider potentially variable thermal loads in bridge or tower members, which could happen if chained together without expansion joints. Plus connections are similar, expn gaps are all equal, with none of excessive width. In a long continuous chain, the one and only expn gap width could be several inches wide,

 

[HTURKAK]

I will suggest you, to provide abutment supports with elastomeric bearing pad and connection to anchors with slotted holes.. With this set up,
You could limit the axial force developing due to thermal changes , and limit the moment transferred to the abutments.

The integral conn. detail with intermediate piers are OK. The flexibility of pier columns will accommodate the thermal movement.

The following snap is from Steel Bridge Bearing Design and Detailing Guidelines AASHTO..

You may reach the full doc. from the link=https://www.aisc.org/globalassets/nsba/aashto-nsba-collab-docs/g-9.1-2004-steel-bridge-bearing-design-and-detailing-guidelines.pdf