Structural Movement

[craigory28]

I have a building that is five years old, that is under going some movement. The building is two stories and the bays are 22’ X 44’. The 44’ purlins span north – south. The first floor is on grade. The second floor is composite. The roof is joist and joist girders. At the perimeter there are precast panels spanning horizontally to the steel columns with ribbon window between them.

The perimeter of the building has partitioned offices. The office partitions extend less than one foot above the ceiling and do not attach to the roof. I was called to check because an office on the second floor on the south side had a ceiling tile fall down.

Upon further investigation the partitions which were not at the perimeter show ½” gaps at the top which go down to nothing at the bottom. This is very apparent on the second floor south offices. On the second floor east and first floor south offices this shows up as well but the ceiling tiles have not been shuffled. It is barely noticeable on the first floor east offices. (The west side abuts an existing building and there are no windows on the north side.)

Our climate has temperatures in the winter in the 20s with the extreme to -10 Fahrenheit. The summers are in the 80s with an extreme at 100.

The precast wall panels are about 8” thick with rigid insulation on the inside face of them.

Has any else experienced anything like this? Does anyone have any ideas on the cause? I have a few theories but I’m interested in seeing what other people have to say.

 

[JAE]

Please clarify - when you say "the partitions which were not at the perimeter show ½” gaps at the top which go down to nothing at the bottom" - are the gaps vertical cracks that are weider at the top and hairline at the bottom (near the second floor level)?

 

[JedClampett]

I don't know if this has anything to do with your problem, but what is the lateral load resisting system for this building? If the precast panels are meant to act as shear walls, they seem discontinuous.

 

[craigory28]

JAE - The partitions, which are perpendicular to the permiter of the building, have a gap at that top of about 1/2" and go to hairline at the bottom.

Jed - The lateral resistance for this building is masonry shear walls. So there should be very little deflection.

 

[JAE]

Well, it could be due to a multitude of causes - possibly deflection or torsional deflection of the perimeter beams (from the weight of the precast). Excessive downward deflection of the second floor framing could also be a culprit. The crack description seems to imply that the interior support of the wall has moved downward OR the exterior beam has moved downward, or both.

Could you survey the floor to see what is level and what is not?

 

[craigory28]

JAE

Thank you for your responses.

The precast panels span to the columns and are not attached to the beams. I checked the columns and they are plumb. I also check the panel and it is plumb as well.

Your deflection idea is one I thought about and was going to run some calculations on. I could surveying the floor would be possible. However it would not explain why this phenomena is occuring at the first floor which is a slab on grade.

Here is the only thing I am coming up with and was hoping someone else has seen it and knew right away. I am wondering if the precast panel bowed inward at one point pushing the walls in. The thermal difference from one side of the panel to other was so great it would have caused this.

Craig

 

[apsix]

If the building is founded on a reactive clay the interior slab/footings could have been pushed up due to heave. After 5 years the moisture content under the slab would likely be substantially higher than around the perimeter, for the right (wrong?) soil and climatic conditions, causing swelling of any reactive clay.

 

[pba]

I think apsix may be on the right track - If you get similar problems on suspended and ground bearing slabs it does suggest that the ground is the problem.

A level survey would be useful in establishing the current shape. If possible get levels on the underside of the floor beams as well - This should give an further information to support foundation movement if they exhibit level differences betwen the ends.

 

[craigory28]

The soil under the building according to the soil report is, "...very stiff to hard comparative consistency silty clay to clayey silt and silt and loose to firm fine to coarse sand and fine sandy silt." The soil capacity was 4000 psf.

We typically do not have reactive clays in this enviroment and none of the adjacent buildings on the site experienced heaving of the slab. Part of the project is a manufacturing floor that shows no cracking and it is subjected to forklift loads. We also have 6" of drainage fill under the slab.

 

[craigory28]

Some information I found out...

PCI Design Handbook has a calculation for thermal bow in wall panel. With the spans I had (22') and assuming a 30 degree temperature differental, I could have a bow of 0.20". Since We have temps from 100 down to -10 and the inside typically 70. The 0.20 could either bow in or out for a total movement of 0.40". Enough if the perpedicular partitions were built tight to the precast panel in the summer, that they would be pushed in about that distance now.

On a different building on the same campus, I found from one of their maintance people that the precast panel was spanning 45' and deflecting over 2". It was tearing up the roofing.

 

[LdsUtd]

it doesn't make sense to me that the tilt walls bowing in would cause cracks in the partitions. on the other hand, the tilt walls bowing out makes sense in terms of causing cracks in the partitions. this is assuming that the partitions are essentially fixed to the tilt walls.

the hole in this theory is that the partitions around the exterior aren't cracked, but that the interior partitions are. this also takes away from JAE's post that the torsional deflection caused by the tilt walls could cause this.

a way of looking more into apsix's theory is to see if the partition cracks at the second floor occur more over the columns, or over the beams/slab. If they occur over the columns, it makes sense that the footings on the first floor could have pushed up.

Just an idea: If by composite you mean steel beams plus a concrete slab over them, where the steel beams have studs so that they are effectively large T-beams, then this might be true: The slab acts as a one way slab over the joists/beams, with the slab deflecting down in between the supports and the slab being "high" over the supports (joists/beams). If the partition runs in line with the slab, it could deflect the same way as the slab. If the partition is fixed at the bottom to the slab, it could theoretically crack at the top at locations over the joists/beams.

Forgive me if I'm wrong, and the deflections aren't enough to cause that, but I'm fairly fresh out of school, and just brainstorming.