Freeze-Thaw Effects on Steel Structures 1

[STpipe]

I don't usually ask a lot of question but here goes.

For some background info, my boss has developed a new joint connection detail for deck panels that will be subjected to cyclic loading. Several other senior individuals in my company have voiced their skepticism that it will work as intended by my boss. Other engineers outside of our company who have looked at the detail have voiced their concerns about possible water infiltration into the joint and the potential effects of freeze-thaw in cold weather climates. So this leads to two questions:

1. I've been searching online for the past few days, but can't seem to find any sort of documentation that deals with the effects of freeze-thaw cycles on purely steel structures and how one would go about analyzing this. On this website, all I was able to find was a mention about how freeze-thaw cycles had managed to permanently deform an HSS column.

2. One of the proposed solutions to prevent water infiltration in the joint would be to use some epoxy to seal the joint. So I was wondering how well do epoxies hold up under cyclic loading or is this something I'll have to take up with the manufacturer?

 

[Ron]

It would help to see a detail of the joint.

In general, steel itself is not affected by common freeze-thaw cycling since steel does not absorb water. It can be affected by the stress induced by water that freezes in confined space such as the HSS section you mentioned, or in joints that might fill with water and expand upon freezing. Concrete in contact with steel and itself affected by freeze-thaw cycling, can adversely affect the adjacent steel.

As for epoxy sealing of joints....not recommended unless there is little or no chance of differential joint movement. Flexible, properly profiled sealant joints are recommended.

 

[hokie66]

Agree with Ron. A quality polyurethane sealant, properly installed and maintained, would be the best way to keep water out of joints.

 

[STpipe]

Hi Ron,

Thanks for the response. I would love to provide a sketch of the detail, however it is a bit proprietary and would easily make me identifiable. This issue is also a bit political within the company with my boss dead set on going forward with this detail, while the company's president is dead set against it unless there's more engineering done on the detail to ensure it works. Essentially I'm stuck in the middle and I'll need to come up with evidence that will either assuage the company's president's fears, or show to my boss that it's not advisable to use the detail despite his "engineering judgment" which says it will be.

The scenario you presented (water freezing and expanding in a confined space which induces stresses on the steel) is what I'm concerned about. I'm looking for literature or help on how to quantify those stresses which I can then use to analyse the joint detail and determine whether or not we should be concerned. Good to know about the epoxy and differential movement at the joint. I do anticipate there will be cases where this occurs so that would be a non-starter.

hokie66,

When looking at the detail with my boss, my first thought was a polyurethane sealant. I'll do some research to see whether this can be applicable to my case.

 

[dhengr]

STpipe:
I’ve rarely designed or built a hollow exterior steel structure which might take on water without trying darn hard to find some way to allow it to drain. That’s drain holes at low spots, intentional openings, coped or clipped corners, or some such. If it might or can take on water, there is a fair chance that it will. I agree with others that epoxy is not a good idea. And, the one caution about caulking is that you detail and explain that on your design drawings, so there can be no suggestion that you are trying to hide a missing or inferior weld. And, I’d still put in a drain or breather hole. I don’t think we are worried about a film of moisture in a faying surface btwn. two pieces of a joint. You should normally be dealing with a sufficient volume of water so that it expands during freezing and stresses and strains the structure that contains it. These forces and movements of the ice tend to be perpendicular to the plane of the plates of the structure, their weak orientation, and likely haven’t been designed for. Also, recall the frozen milk bottle example; if the fluid can expand in a weak direction (unconfined direction), it might with little harm to the container. We realize that this detail is one of the last of the great national secrets, but you still have to provide enough pertinent info. so that there can be a meaningful discussion. Why don’t you wet/fill the detail with water, put it in a freezer, and see what happens?

 

[DamsInc]

Reminds me of the gondola tower failure in whistler due to ice jacking. Might find some useful information in the attached report.

 

[amecENG]

If you are looking for a design force to apply to a FEM model you could use the crushing strength of ice (~3MPa to 5MPa). The expansion force of the ice cannot be greater than is compressive strength or it will just crush itself.

 

[DamsInc]

"If you are looking for a design force to apply to a FEM model you could use the crushing strength of ice (~3MPa to 5MPa). The expansion force of the ice cannot be greater than is compressive strength or it will just crush itself. "

How can this be true in a confined area? The ice can't just crush to relieve the stress. Just as water is incompressible.

 

[STpipe]

dhengr,

I was thinking along the same lines with your example of the milk bottle and ice expanding in the 'weak' or unconfined direction leaving the structure intact. That's the case with my detail, and it's not fully confined by steel. I've attached a sketch showing my general concern - it doesn't show the detail in question but still illustrates the potential issues. So in theory, one would expect the ice to expand freely towards the opening at the top of the joint without inducing any stresses on the walls of those panels? However, that would also be undesirable since the intent is to have a wearing surface above the joint and the pressure from the expanding ice could potentially crack/damage it.

DBond,

Thanks for the report. It's a very interesting read that I'll have to look into more detail once I get home this evening and see what other information I can extract from it.

amecENG,

That's precisely the sort of information I would be looking for. My boss was of the same mind (taking the crushing strength of ice) and applying it as a pressure, however DBond doesn't seem to be in agreement?

 

[rb1957]

i think that as the water freezes it expands. this is an enforced displacement if the water is confined, and not equal to an internal load of 5psi ... IMHO.

if there is a contained space that could fill with water can you

1) add a drain hole ?
or
2) fill the space with some flexible sealant ?

if this is a battle between your boss and his boss, i reckon his boss'll win (CEO's are like that !) why not make a sample and see ?? (whoever loses quits, without bonus !)

Quando Omni Flunkus Moritati

 

[DamsInc]

".....however DBond doesn't seem to be in agreement?"

Well, it depends on the detail. That assumption may hold in unconfined conditions, it's a different story when it is confined.

 

[STpipe]

rb1975,

The intent is to definitely fill the space with a sealant, but as I'm told in this topic, something like epoxy wouldn't be suitable. As for the upcoming battle between my boss and his boss, it's slightly more complicated than that as they are brothers! I'm hoping we will test a sample just to see what happens, which is something I'll be pushing for. No idea if they will accept it.

DBond,

Would you say in my case that it would be closer to the unconfined case?

 

[DamsInc]

Yes, I would. But I was thinking more along the lines of an HSS filling with water, not knowing the details of your situation. I don't know how much of an effect a few mm of ice would have on a detail like that.

 

[madmantrapper]

I've seen water freeze in pipes of all sizes, unconfined and confined, break open. I've seen scaffold legs get water in them that were plugged with dirt freeze and break with the top sides open. Don't know how this relates to what you are doing.

 

[TomDOT]

Don't let water collect in contact with steel, much less inside.

 

[ameyerrenke]

Put in a weep hole. There is no way to make steel stronger than freezing water. Without some means to allow for the water's expansion, it will expand anyway, and it will deform anything that tries to prevent it from doing so.