Length of Building 1

[RSMENGR]

Dear All:

Uusually we provide an expansion gap for every 30 to 45 metres spacing as given in various codes. Reasons is given that we have to accmmodate the temperature expansion/contraction as well as better lateral load behaviour.

We have a problem where our architects want to provide a 90 m long building and that too without any expansion/contraction gaps. Now suppose we wish to provide this building on base isolators like friction pendulum etc. what difficulty we might face if we have such a long building on base isolators?

The building is to be located in an area where we have temperature variation 10 degrees C +/- .

Best regards

RSMEngr

 

[14159]

I had a solid answer for you until you started talking about base isolators.

I've designed 3-4 300-ft long multi-story buildings without expansion joints, mostly large hospitals. These were all in the USA, mostly in the southeast. Our office used this as a cutoff for 15-20 years, so your 90 m without EJ wouldn't scare me at all. Expansion joints cause such chaos and difficulty that I would not even think of including them for the small lengths that you listed.

I've never designed a building with base isolators, but I don't see why a 90 m building causes special challenges not faced by smaller plans. That's not a very large plan. I suppose you have a larger mass moment of inertia, so your building might twist more, but that should be included in your analysis.

DBD

 

[rapt]

DBDavis,

If you calculate the moments and shears in the columns due to the effects of temperature change and slab shrinkage you would never do a 300' building without expansion joints.

As long as there are no stiff restraints at the ends, about 70m is the maximum and even then the effects on the columns need to be checked and steps may need to be taken to relieve shrinkage effects.

If there are stiff elements nears the ends then shorter lengths between joints are required and steps must definitely be taken to relieve shrinkage effects.

 

[astructurale]

10 deg C = 50 deg F. That's pretty good temperature differential. I'm most familiar with precast so PCI (Precast/Prestressed Concrete Institue latest ed -6) would not allow a 300ft (90m) building based on that temperature differential (Fig. 3.10.20, which is also the recommendations of the Federal Constrution Council in the US). However, I have seen buidings at 300ft without an expansion joint.

When it comes down to it, its an engineering judgement issue, and you know the structure you are building better than anyone. You should account for the types of connections you are using, the column stiffnesses in simple span structures, the relative stiffness between the beams and columns in framed structures, the location of lateral load resisiting elements, and the weather exposure conditions. If you can justify your building while condsidering the effects on all of those factors then you should be fine. You may want to even doing a quick and dirty check of the build up of forces due to expansion of your structure to see if you are even in a reasonable range for your structure type.

 

[14159]

"If you calculate the moments and shears in the columns due to the effects of temperature change and slab shrinkage you would never do a 300' building without expansion joints...."

Your point is well taken.

I should've tempered my comment by mentioning engineering judgment. We had smaller buildings with EJs depending on the geometry, climate, etc. It all comes down to the judgment of the principal in charge.

We talked about this stuff in the office, but the bottom line was that the firm has done high profile work for 30+ years and done that kind of thing many times without a problem. The overall firm attitude was that this stuff is typically blown way out of proportion.

The calculations for temperature and shrinkage effects are usually extremely idealistic and ignore many things that help, like the foundation moving just a tiny bit, column isolation joints, etc.

I've done this before and it's what convinced me: re-analyze the structure with those thermal and shrinkage effects, only this time, allow the column foundation to move just a little bit and see what happens. I did this on a large parking garage and the forces dropped dramatically.

DBD

 

[JAE]

I've designed a roof system that was almost 700 ft. long. No expansion joints as it was a clear span (cable stayed roof system) but the roof was set on teflon all around so thermal changes wouldn't be transmitted down to the structure below. If you design for and deal with the thermal forces, you can really do almost anything.

But I've always wondered about how much thermal changes really affect a structure - all the columns, beams, skin, etc. have relatively similar thermal coefficients and much of the building skin is in pieces so a cumulative effect doesn't always occur - and everything experiences the thermal change together so the differential doesn't always add up to much structurally.

AISC has a chart that implies a length of 500 ft.(150 m) for steel framed buildings is acceptable for a 55 deg. F temperature swing.

 

[SacreBleu]

JAE,
An example where thermal changes are more critical is a mostly steel-framed building that has masonry (CMU) shear piers or shear walls, and the piers/walls are spaced relatively far apart. The steel beam lines supported by the masonry have a tendendency to go through large changes of length, effectively tearing at the beam seated on the masonry. (The beam line acts a drag strut, connected to the masonry). We had quite a problem during construction when the temperature of the beams fluctuated from day to night.

 

[rapt]

JAE,

I have seen the effects of a 100m long concrete beam tied into columns with 5MPa prestress applied. The shortening was equivalent to a strain of about .0003. Half of the end columns sheared off and there was a lot of cracking in a lot of other columns and also in the beams. The repair cost was enormous.

This is roughly equivalent to the shrinkage strain for a 1st floor slab supported by columns to footings (where the footings cannot move horizontally as they did not in this case). Add to this a similar shortening for temperature and the end columns are very sick.

Agreed for the higher floors in a multi-storey building, everything moves together so the differential floor shortening is small, but at the ground it is enormous.

Yes, everything in the structure has similar thermal co-efficients, but the columns are short and vertically alligned and the slab is long and horizontally aligned and this horizontal shortening of the slab, restrained by the columns connected to something immovable like the ground causes the problems. 20C differential results in about .2mm/m of shortening, so for 100m, the slab shortening would be 20mm due to temperature plus about 30 for shrinkage giving 50mm total. Each end column will experience about half of this so 25mm horizontal movement at the top of the column. Slab cracking and Creep will reduce the effect of the shrinkage component but the moments and shears induced are enormous.

Also, for the person doing the temperature conversion, a 10C differential does not equal 50F differential, it is about 18F, but 10C is a small differential between summer and winter or airconditioned and during construction conditions plus shrinkage must be added.

 

[JAE]

Good points.

 

[pba]

One approach which I've used in the past on steel frame buildings is to consider the beam members as subjected to a compression force due to the restraint of thermal expansion - E = stress/strain - thus stress = E * strain.

Typically here is the UK we use a possible temperature range of 40 (degrees C). This is plus or minus 20C which gives about 50 N/mm2. This reduces the available stress for resisting bending but not usually so much as to lead to silly steel sizes. (UK steel stresses are 275 and 355 N/mm2)

This approach can also be used where expasion joints are provided but not to cater for the full temperature range. I have to say I've never tried that - If you are going to provide the joint - DO IT CORRECTLY.

 

[JKW05]

There is a publication by the National Academy of Sciences titled "Expansion Joints in Buildings, Technical Report No. 65" that suggests empirical and analytical methods to approach this question. It can be found on line:

http://books.nap.edu/catalog/9801.html

or Google "Expansion Joints in Buildings"

Hope this helps.

Respectfully,

JKW