us vs. them, joist design 4

[AnimusVox]

Hey everyone,

I'm an EI (I've had two internships and a couple of months of experience at my first job) and I had a question related to steel bar joist design. Mainly, do you designate joists based on the tables provided by the manufacturer or call out applicable loads & load patterns on the drawings and let the manufacturer do the designing?

I've done both but I'm curious what reason(s) warrant doing one over the other. Are there extra costs involved with letting the manufacturer design the joists? Does the manufacturer use any tools more extensive than the loading tables they provide us in manuals?

Thanks in advance for your input everyone.

edit: I clarified I am talking about steel bar joists. Also, we don't designate joist girders (are there engineers that do?) or joists that are subject to point loads. However, the matter of if we designate joists subject to uniform loading has varied between the departments I've been in and who the EOR is / superviser preferences, so I just wanted to get everyone's two cents on their experiences. Thanks again for your responses!

 

[SLTA]

I've only ever called out the joist sizes, with applicable extra loads indicated as required for mech, etc. Sort of a mashup of your methods.

Usually firms have a standardized way of doing it - have you asked what your job's standard is?

And good luck!

Please remember: we're not all guys!

 

[BSVBD]

Assuming you're talking about Open Web Steel Joists:

If it's a simple uniform load, i use the tables and call out the size.

If i have trap and/or point loads, i determine the moment and shear and then specify a KCS.

I have also, often provided load diagrams, if a local joist provider advises me to, since the logic may be more economical.

For Girders, unless it is uniform point loads, i provide loading diagrams.

Contact a local, in state, joist provider and ask their advice. Since they will be bidding the project they will guide you in a realistic and practical direction. Also, if practical, contact two different competitors - everyone has differing, yet, helpful opinions.

If you are not talking about steel joists, none of the above applies. Therefore, please clarify.

 

[KootK]

I always either:

1) Show the loads in some fashion (plan/notes/load diagrams) or;
2) Specify a KCS joints.

I think that manufacturers can optimize a bit more if they know the actual loads. Also, who wants to cycle through all of the joists on a project doing detailed parapet drift load analyses etc. Yuck.


I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[gravityandinertia]

I used to work at an open web steel joist manufacturer doing the design work prior to fabrication. Even though you designate something from the catalog, not everything made with that designation is the same. The designations simply cover a basic rating. Using the additional loading if provided the manufacturer will try to ensure a panel point is at a load location in order to keep a chord from having to be designed for bending. The manufacturers try to optimize joists, while also using the materials they have available.

Typically, raw materials like steel angles are only produced every 6 months, so every 6 months a joist manufacturer purchases materials not knowing what sizes they will need since the work coming their way is unpredictable. As a result, they always try to optimize it to use the smallest size possible because if they use a smaller size and run out, they can always replace it with a bigger size angle, however, if they run out of big angles, they can't manufacture it to be unsafe and are screwed. The end benefit to the Structural Engineer is they they are always going to get a joist as light as the manufacturer can design it to be.

 

[JedClampett]

When I can (simple loadings, flat roof, etc.) I try to pick a joist out of the catalogue (SJI) that's a little bigger than I need. Or when there's a lot of equipment on the roof, a KCS joist, that's a little stouter than required.
I don't like to leave the design to the joist manufacturer. If you give them the dead plus live load, that's it. No margin. If you call them and you gave them 100/250 (100 DL, 250 total), they're going to claim that's it and not one ounce more. Dial tone. If something changes (and it always does) you've got no room, for an air handler for example. If I call out a "K" joist, I know it's got a design capacity and I've built in the margin.
Why are we so concerned about a few dollars of steel, when there's thousands of dollars of furniture or equipment in the building, not to mention human lives?

 

[Gumpmaster]

Jed is spot on. Someone always gets bit down the road when you design to the bare minimum. See also PEMB's. I always specify tighter deflection limits and a little bit higher loads for those as well. The cost is minimal compared to the $4000 chairs and $5000 light fixtures the architect is specifying.

 

[RWW0002]

Where calling out the joist sizes get a little more complicated, and potentially less economic, is when snow drifts etc. are present. For roof joists, I find it easier to give uniform loading on the plans along with snow drift diagrams and net wind uplift diagrams as necessary and let the manufacturer choose the joist size. That leaves the joist and uplift bridging squarely in the joist supplier's hands. It also theoretically gives the contractor the leeway to decide if it is warranted to, say, go with a little heavier joist to avoid diagonal bridging requirements at midspan. If some additional capacity is warranted for future flexibility (and in my opinion it almost always is) reasonable add loads and bend-check loads can (and should) be specified in plan notes. That way the joist design is fairly efficient and that air handler can still be hung.