Compound deflections for steel beams/girders 2

[Browen]

Hello all,

Novice structural engineer here, been lurking for a few years now. I could use some advice, as I may or may not be being gaslit by one of my colleagues.

Say I have a beam (steel in this scenario, but I don't think that's important) which is supported by girders at both ends, framing into the midspans of the girders. Do the deflection criteria for the beam (L/240 for total load or 1" is what I'm using) depend on the deflections of the supporting girders? For example, if the midspan deflection of my beam is 0.5" but the midspan deflection of the girders is also 0.5", do I not have any wiggle room with the beam since its compound deflection is 1"?

I understand one purpose of the deflection criteria is to prevent damage to the floor due to differential deflections of the underlying beams. For my above example, the differential deflection of my beam is 0.5", which is less than the 1" typical limit. Would I be correct in continuing to design based on differential deflection instead of compound deflection?

Thanks ahead.

 

[CANPRO]

Most often you should be checking the L/criteria based on relative deflections of the member in question - in your example, its the 0.5" value. This is because, as you suggest, its the relative deflections that matter to the finishes and user comfort.

There are cases where you should check the global deflection as well. Again using your example, if you only have 3/4" clearance to some sort of equipment or another structure below the beam in question then you need to check the global deflection which includes girder deflections as well - in your case, the global deflection would be excessive, even though you passed your standard L/240 check.

 

[XR250]

I think there other situations where the global deflection is important - namely occupant comfort.
I usually do not get too excited about this unless it looks like a Tinkertoy model.

 

[CANPRO]

XR250, agreed that there are other situations where global deflection is important. Although I've always put user comfort in the "local deflection" camp, not global. I suppose if you had a very large open space like a suspended gymnasium and the deflection was excessive you'd maybe see/feel yourself moving relative to the exterior walls. I'm curious what kind of situation you're picturing where global deflection would be a user comfort issue.

 

[phamENG]

I agree with CANPRO and XR250.

For checking against the code mandated deflection criteria, use just the member in question.

For absolute spatial constraints or occupant comfort, look at the global deflection.

Some other cases where global or a local/global perspective is valuable:
-Ponding on a roof (compounding deflections can create a larger "pond" and lead to a collapse of your roof)
-Framing discontinuities where the local member deflection is fine, but the resulting deflection in your floor deck either violate the requirements or will just create a sudden change in magnitude or even direction of deflection that will be noticeable.
-Vibrations due to human occupancy. This is less of a deflection thing, but but the dynamic response is closely related to the stiffness which, of course, drives deflection, too.

 

[Agent666]

I'd disagree with the just use the isolated member camp, you need to be mindful of the total cumulative deflections with some caveats.

If you are screeding your slab to a level and unpropped, then it is less of a concern (simply because top of slab deflections will be lower), provided you are allowing for the additional weight of concrete from the ponding effects on individual beams(which in turn will increase deflections further... Rinse and repeat until you get convergence). This can become significant depending on the magnitude of the deflections. Only long term effects reflect at top of concrete level as far as the occupant is concerned with unpropped/screeded to level construction. If soffit is exposed to view you'd also need to be mindful of the steel deflection from a visual standpoint.

If you are screeding to a thickness (unpropped) or dealing with propped construction then my opinion is that you need to be mindful of the total deflection as the deflections are more likely to hit recognised limits. The occupant doesn't know what framing is, they just experience the total sag of the floor system as a whole. Guidelines in this part of the world recommend that the engineer checks the diagonal distance with the total sum of the deflection of the entire floor system(delta max from screenshot below). If you are supporting beams off beams off beams off beams and designing each beam to the max deflection limits then you're going to potentially have issues as it adds up.

 

[r13]

Which deflection should I concern with?