Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations KootK on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Roof Joist camber causes some problems

Status
Not open for further replies.

dcceecy

Structural
Oct 15, 2008
112
We designed a one-story building with open-web steel roof joists and the standard joist camber has caused some problems. It seems that it's a standard practice to have joist cambered per SJI's specification? But we did not specify that the steel W beam parallel to the roof joists should match the camber of joists. Both the steel beam and roof joists (4'-0" OC) are used to support the metal roof deck.

The first problem we have is that there will be a gap between the metal roof deck and steel beam. Because the adjacent joists are cambered and due to the large dead load the camber in roof joists is big. (the metal deck can not just bent for attachment to the steel beam)

The second problem we have is that at the end wall (parallel to the joists) there will also be a gap between the metal deck and the end wall (supporting the edge of roof deck). I was told that one solution to this problem is that stop the metal deck at the last joist and use one span metal deck between the last joist and the end wal. The last span of metal deck will have a slope due to the camber of joists.

When we designed the building, we didn't pay attention to such problems. So the contractor may have a good excuse to ask for more money to solve such problems.

Any suggestions and comments? What is the typical approach to specify joist camber? Just add a note in structural steel section " the steel beam camber shall match the adajcent joists?" or specify "Do not camber the joists"?

Thanks
 
Replies continue below

Recommended for you

How big is the gap between the joist top chord and the steel beam and the top of the wall? The camber in the joist that comes should flatten/level off when the dead load is applied. I have not encountered this problem before and I designed more roofs than I care to admit to. Standard camber should be small as I stated above. Keep in mind many steel beams come with natural camber as well!

If you are certain that the camber will remain or too large for STD SJI, I would contact the joist supplier for suggestion. Alternatively, you can weld a continuous angle or plate (with intermittent welds) to the top flange of the steel beam. I do not recommend changing direction of the deck to single span between the last joist and wall. You can add another joist within inches from the wall or you can anchor a bent plate (or angle) to the wall so that the top of the horizontal leg is at same level as the top chord of the joist. Make sure the new angle and the anchors to the wall are capable to transmit the vertical roof loads and the horizontal shear if the wall is acting as a shear wall.

IMHO, the contractor should not ask for more money since you did not specify for the joists to come with more than the STD camber and the contractor should have accounted for the camber as part of his experience and ways and means.

I usually do not specify joist camber unless it is required. I call for camber on long span joists where concrete slabs are being placed on top of them. This will reduce the concrete thickness as the joist deflects from the weight of the slab. On large projects, this can be many yards of concrete to bring the entire slab to level.


Regards,
Lutfi
 
Thank you for your reply.

The big dead load is from the cable tray (50 psf) hung from the joists.
It seems that the joist fabricator camber the joists according to the total dead load. So there is a big gap between metal dead and steel beam when they attach the metal deck to the steel beam (the cable tray will be hung later )

I can not remember the size of the gap but it should be bigger than 1". I think the gap should also be curved as the cambered joist and not uniform along the span of joists. So I am not sure if adding a steel angle will work (because the top of horizontal leg should follow the joists). Maybe adding shims between the metal deck and end walls or steel beams will work.

I am not sure if adding a joist within inches from the end wall will work in my case. Because the end wall is a shear wall, we want to use solid connections between walls and metal deck. If there is a joist within inches from the wall, the connecion between the joist and metal deck may be damaged if the cable trays are hung from the joists (the joist deflects but the wall did not).
 
Odd structural framing to have a wide flange sandwiched between joists and expecting the same deflection/camber reactions. Difficult to achieve and maintain during loading. As for the perimeter angle, this can be floated then fully attached after dead load is applied. The angle can be cambered easily to match the joist, but you must let it float until load is applied, otherwise it will be higher than the joists.
 
Thank you for your relpy.

The steel beams are needed for lateral load resistant system (moment frame). I think if the maximum gap between the joist and steel beam is 3/8" and the adjacent joist is 4'-0" away. They may just push the metal deck down during the welding. But the gap is too big to be closed in that way.

The end wall is a shear wall. So is that going to cause problem if we leave the cambered angle floating? The cable tray may be installed after all the building construction is done or later. we need solid connections between shear wall and roof deck.

I should admit we design odd structures.
 
Ron, you are on target on the comment and the option for the angle.

The continuous angles can be rolled to a radius that matches the joist camber. The can provide vertical slotted holes and finger tighten the bolts. Tighten and may be weld in place after all DL has been applied and the top of angle and joist are level. Similar detail can be developed for the steel beam (See attached hand sketches). When I said that I used steel beams, I used WF beams where heavy loads are anticipated and where deck changes direction. It is obvious the stiffness of steel joist and WF beam are not the same.


Regards,
Lutfi
 
 http://files.engineering.com/getfile.aspx?folder=fd07b739-82d5-4372-9865-2d29e64feb1c&file=scan0021.pdf
You wouldn't need to try rolling the angles, but rather just deflect them as they are connected.
 
Don't forget that you'll need to isolate the roof flashings at the side walls. If attached before all of the dead load is on, they will pull away from the attachment.

 
Contrary to some of the posts above, this is a very common problem with longer span joists. When we design a structure with longer span joists, we will sometimes detail the structure as to avoid this problem.

Standard camber is located in the SJI catalog. If you run into this problem again you can attack it in a number of ways. Specify 1/2 the camber in the member that is located next to the flat member. Or detail the flat member to have some way of accounting for the camber.

We usually space our joists a little farther apart (5'-6'). This will sometimes allow for the erector to get some "large" guys to stand on the deck/joist and force it down in order to weld.
 
I always avoid having long span members adjacent to perimeter walls due to not only the preload camber problem you note but also the potential reverse situation when fully loaded midspan deflections will be large compared to the short deck span from the wall.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor