Joist Analyis / Modification

[sponcyv]

Hey guys. I am reviewing an existing roof which will be getting new roof top units. I do not know the depth of the joists, so I have to assume the joist size. The only way I know how to do this is to take the joist length and existing dead and live loads and find the smallest joist in the joist in the vulcraft manual that works.

Once I've assumed a joist size, I then analyze the joist in a spreadsheet that we created in-house. The spreadsheet tells me how much the joist is overstressed in moment and shear. It then checks web and chord reinforcement that I add.

We have gotten information that this way of reinforcing the joists is very laborious and expensive. The engineer that designed the job that I'm reviewing put in wide flanged beams across the bottom chord of the joists to share the load with each joist. The contractor thought this was too much and sought us to review the job.

I'm wondering if there is any ideology differences out there or methods of analyzing joists in this situation. Any input would be helpful.

 

[JAE]

Well, first off I'd not assume a joist size because that is just too speculative for my tastes.

I would get from the site a joist tag if possible, which many times indicates the fabricator name and project number. The fabricator could then possibly tell you the actual joist size. Using SJI's historical data book, I could look up the joist capacity and then compare that capacity with the actual in-place dead loads and current applied live/wind loads to see if the joist is within acceptable limits.

If there's no joist tag, another option would be to get out there and measure the chord sizes, web sizes, etc. and produce an approximation of load capacity.

Another option is load testing per IBC Chapter 17.

Finally, adding beams under the joists or adding supplemental joists/beams to support the units can be considered.

If I know the joist sizes with some level of confidence, I will allow up to about a 5% overstress on the presumed joist capacity.

If I don't know the joist sizes and field measure, I won't allow the 5% overstress.

 

[JedClampett]

My advice is to not get too far out on a limb on this. Trying to figure out joist capacity from observation is risky. There's numerous threads regarding this on the board. Add to that the fact that joist are not made for concentrated loads and you're entering uncharted territory. And if something comes down, you're going to be very lonely. I just got an Email this morning for a SJI Publication No. 12 on Modification and Evaluation of Steel Joists (

http://www.steeljoist.org).

That might help.
But due to the difficulty of determining the actual capacity of joists, we usually provide a separate frame just like was proposed on your project.

 

[sponcyv]

Thanks to both of you for your input!

JAE - I agree with everything you say and the 5% increase is also a rule that we use. I am just reviewing this job and have been asked to spend less than a day's worth of time on this.

Jed - Thanks for the reference. I will check it out.

I basically wanted to know if everyone out there was following the same type of protocol as our company.

 

[BAretired]

I had a very similar project a couple of years ago, except that I spent considerably more than a day on it. Some of the proposed mechanical units were quite heavy but snow drift resulting from the units projecting above the existing roof was even more critical than the weight of the units. There were no shop drawings available for the joists but the joists could all be measured and calipered.

I ended up not completing the job because the client (the Contractor installing the units) became impatient with the amount of time I was taking.

The engineer who took over the project visited my office and reviewed all of my assumptions, measurements and calculations. I didn't follow the job any further, but I believe he ended up assuming that the existing joists were adequate only for the original specified load so he reinforced the joists to carry all new mechanical loads but ignored the snow drifts.

He was much braver than I and I was quite happy to lose that job.

BA

 

[TXStructural]

Any supplemental beam/truss needs to have about the same stiffness, or it has to replace the existing.
The truss will most likely be stiffer than a shallow, single beam member. The truss will take all of the load, until it can't take more (as the beam deflects), then all of the load transfers (maybe suddenly) to the beam as the truss fails.

Reinforcing only the bottom chord invites failure of the top chord and web members - and it is unlikely that the original truss design has more capacity than what was required.

I've seen rod welded along angle chord members on each side, top and bottom, to strengthen open web bar joists. If the web can take it, this seem viable. I've designed beams placed parallel to the joists to take new loads, and used platforms to place the loads on the new members. The deck can flex between the new beams and existing joists, preventing excessive loading of the joists. Alternatively, place the new members as needed and use standoffs which penetrate the deck to load the new members.

 

[SteveGregory]

I am adding a ditto to Jed's post, get SJI Publication No. 12.

 

[sponcyv]

Thanks again for everyone's responses. I ordered SJI Technical Digest 12 and it is great!

 

[JedClampett]

I referenced SJI Publication No. 12, but I hadn't got it yet. Thanks for the review. I'll have our librarian order it.