seismic joint between buildings

huangyhg

seismic joint between buildings
i'm working on the design of a pedestrian bridge that will connect to a large parking garage structure.  the general consensus in the office is that there should be a joint separating the bridge from the garage.  i'm not sure i understand why.  does anyone havbe any ideas?
generally two dissimilar structures such as you have decsribed should be seperated for lateral evaluation purposes. i don't know the situation of your structures, however, in most cases it would be best to isolate the two structures, due inpart to lateral loads such as wind or seismic.
1st: the two building will likly to have two different periods due to dynamic loads (wind, sesmic).
2: you still can model your thing without the joint but:
you will get a hell of concentrated complix stresses around the bridge supports and in the bridge   
i agree with erv and rawand.  also, even if you had the "perfect site" with no seismic and good soils, you would still have to contend with the different responses to thermal changes.  using the joint separation is the right thing to do, when ever you have a change in geometry.
thanks for all the input, it's been very helpful.  one more question related to this:  if this bridge is connected to a small building (housing stairs and elevators), would it be feasible to deign them with a joint?  the bridge is 250' long by 16' wide steel truss suppported on 3 concrete piers and the building is 80' long by 35' wide with moment frames in the short direction and braced frames in the long direction.  both bridge and building are the same height.  does similar reasoning apply that since the lateral system are very different (concrete piers vs. steel moment frames) that the responses would be different and thus warrant a seismic joint separation?
i agree with the concept but to make sure for a particular situation:
say for instance you have a 500' long, tilt up warehouse bldg. it has full length and height load bearing partition walls that separate the structure into 5 equal pieces.
you would put in joints at each interior wall for thermal changes and also enough of a gap to prevent seismic pounding, even though the periods of each segment should be the same?
1- it is easy to disconnect this bridge from ur building using proper steel joints (movment joint)
red text
2- these two buildings will not act the same
3- as stated: joining the two will lead to unpredicted stresses.
green text
i would consider the joints between tilt-up panels as covering the requirements for expansion/contraction joints.  the panels are usually on the order of 8' wide.
   
sorry jheidt, i did not explain thoroughly.
i meant seismic movement joints at the roof level on one side of each of the 4 interior tilt bearing walls. like a typical expansion joint at the roof, located adjacent to each bearing wall.
i think rawand is indicating that they would be required.
the way i read the ibc is that this seismic joint is only required in class 'd' 1620.3.6. but it may be a good idea for the bridge to bldg. situation no matter the class since they are so dissimilar, and also due to wind effects.
have to also consider expansion and contraction. bridges go through quite a temparature differential.