Full and Half Depth Stiffener Welds

[Kevds122]

Hey everyone.

I currently work for a steel fabrication company and am developing a set of calculations regarding full and half depth beam stiffeners. It is not uncommon to get structural design drawings from the engineer of record that show stiffeners with welding on both sides of the stiffeners to the web and flange(s), which is a bit excessive in my opinion and can be reduced.

The calculations that I am looking to develop would involve determining the ultimate strength of the stiffener plates and then designing welds based on the strength of the stiffeners to request a reduced weld size with a possible lesser length than the weld that I described in the paragraph above (less weld = less labor/less equipment use = saved costs).

The problem I run into is determining the wording in J10-8 of the AISC Manual (14th Edition, pg. 16.1-138)
"Stiffeners required to resist tensile concentrated forces shall be designed in accordance
with the requirements of Section J4.1 and welded to the loaded flange and the
web. The welds to the flange shall be sized for the difference between the required
strength and available strength. The stiffener to web welds shall be sized to transfer
to the web the algebraic difference in tensile force at the ends of the stiffener.

Stiffeners required to resist compressive concentrated forces shall be designed in
accordance with the requirements in Section J4.4 and shall either bear on or be
welded to the loaded flange and welded to the web. The welds to the flange shall be
sized for the difference between the required strength and the applicable limit state
strength. The weld to the web shall be sized to transfer to the web the algebraic difference
in compression force at the ends of the stiffener. For fitted bearing stiffeners,
see Section J7."

My assumption would be to use the follow equations for the strength of the stiffener plates:
Full depth stiffener plates
Rn = 1.8(Fy)(Apb) (Equation J7-1)

Half depth stiffener plates
Rn = (Fy)(Ag) (Equation J4-6)

I could then design the welds based on these values.

Any help would be greatly appreciated!!!!

 

[PersonalProfile]

I am not able to directly assist you with your enquiry, however I note:
1. Despite I am not aware of your situation in detail,I propose that in principle if you change the design [stiffeners' weld details] you must seek consent from the designer. Furthermore you may want to ensure you have appropriate insurance/registration etc.
2. The stiffeners' weld details may be for other than strength - i.e. fully welded for durability etc.

Regards,
Lyle

 

[KootK]

I think this is a great idea and would do the same thing if I were a fabricator. 95% of the stiffeners out there could be clipped short of the opposing flange 4-6 tw without affecting the capacity of the connection at all. The only real exception I can think of is where you've got loads coming in on both sides of the member.

If it were me, I'd do it like this to keep it simple and appealing to EOR's:

1) Create a standard RFI indicating that you'd like to use the EOR's stiffeners and weld sizes. All you want to change is clipping the stiffener back from the unloaded flange's 4-6 tw.

2) create a little brochure that you send along with the RFI that explains why a clipped stiffener is sufficient for nearly all conditions (bearing, crippling, LTB bracing, flange bending...). And supply the relevant AISC clauses to back it up. You'd be surprised how poorly some designers understand the use of stiffeners. 3/8 full height stiffeners are just the stock default where there's no time to really think about it.

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.

 

[KootK]

You could also offer to share 1/2 the savings with the owner/GC.

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.

 

[dhengr]

Kevds122:
I don’ t have the last couple Eds. of the AISC Manuals so I can’t comment on specific formulas and the like. I would like to see the formulas and the text around them or other parts they refer to. It would also be interesting to study the commentary surrounding these topics. What is your engineering education and background, so you can understand and sell what you are trying to do, both at your company and to the EOR. I suspect that on a given job, if you only have a few stiffener plates, you will spend more time on RFI’s and trying to get the EOR to change than you will save in welding time and consumables, if all you are trying to do is reduce the weld size a little. If you have a job with hundreds of the same detail, then it may really pay to try to refine that detail.

Clip stiffener corners at the flg./web fillet, and don’t (or so you can’t) weld into that radiused corner. Avoid fitting to that fillet radius, and with few exceptions no perfect fitting, in height, btwn. the flgs., that’s really time consuming for a fitter. As KootK suggests cut the stiffener short of one flg. or the other, I would say, to avoid the flg./web fillet radius. Also, don’t weld right to the edge or end of a stiff. pl., stop a little short so as not to nick the corners fo the stiff. edge. At starts and stops of welds don’t leave any craters. Always have your welders dwell for a second at the start and weld back into new weld a bit at the ends of welds, so as not to leave cold starts or craters. I really don’t like stitch welds for all their starts and stops, and because it takes more in layout time and over welding so as to pass inspection, that you might just a well do a smaller continuous weld with the same cap’y. quicker and cleaner. Stitch welds allow water into the joint also.

Many engineers don’t know what they are doing when they design stiff. pls., they just follow the cookbook, pick some pl. thickness, and apply some sort of weld without giving it much real thought. How will you determine the loads on your stiff. pls., half the time I’ll bet the EOR can’t tell you what they are. What does the stiff. pl. really do? You mention tension or compression loads on the stiff. pl. In either case, it is really intended to get a large concentrated load down into the beam web or maybe to act as a web buckling stiffener as on a plate girder. If you know the load, then size the welds accordingly. But, some ult. strength of the stiffener will probably still oversize your welds.

 

[TLHS]

Half the time stiffeners probably aren't even require where people have detailed them in for bearing or small moment connections. People seem to severely undderestimate web crippling capacities in most scenarios.

 

[Lomarandil]

TLHS, that is probably partially related to wanting to provide torsional restraint at support points (per AISC F1.(2) and J10.7).

It's always seemed to be a little overly prescriptive to me, but I'm sure in most cases the cost of a stiffener is the lesser of evils to a designer.

 

[KootK]

For me, stiffener laziness is a result of my not being at all convinced that any stiffener savings that I generate will actually make their way back to my clients.

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.