Design of base plate with axial load and bending moment

[Chandan321]

Hi All,
I am trying to design a base plate with axial load and bending moment. I have different cases of eccentricity: L/6 < e < L/2, L/2 < e < L and e > L.

I found that different references have different methods of designing it:

1) Blodgett uses the assumption of plane section remains plane and solves a cubic equation of forces and strains to get: concrete bearing stress, bearing length and anchor bolt tension.

2) AISC Design Guide 1 says that the above assumption may not be correct as the base plate and concrete may not be in contact everywhere. The anchor bolt tension and concrete bearing stress are independent of each other. The Design guide assumes that the bearing stress will reach max allowable value. The guide then finds the bearing length from summation of moments, and then the anchor bolt tension from summation of forces.

Each of the above methods yield different results. I want your suggestion as to what method is generally used in the industry practices for designing base plate with axial load and bending moment.

Thanks for your time.

 

[paddingtongreen]

I used the analogous, working stress, cracked concrete section using a cross section the size of the baseplate and the bolts being the rebar. I used this to check the concrete bearing stress and that stress to load the baseplate for it's design.

Timing has a lot to do with the outcome of a rain dance.

 

[transmissiontowers]

ASCE Manual 48 and the new Manual 113 have methods outlined to determine the base plate thickness. There are several cases to study. The base plate on leveling nuts and base plate on concrete where the bolts only have tension. I have used the Blodgett method for hand calculations for many years. In MOP 48 and 113, you basically determine the bolt loads and then assume the bending plane location and length and sum moments about the bend plane and then solve for the thickness required for the plate to remain below the yield stress (assuming LRFD had overloads on the applied loads).

You can get a free base plate program from Hilti called Profis that will do a FE analysis of a plate and bolts but there is little documentation of what it is doing. GTSTRUDL has developed a Base Plate Wizard that creates a FE mesh and boundary conditions that you run in GTStrudl with tons of documentation.

_____________________________________
I have been called "A storehouse of worthless information" many times.

 

[dik]

The AISC also has a Design Guide for this. I just modified the methodology to match LRFD (limit states, here).

Dik

 

[Chandan321]

Thanks for all your posts.

All of you have essentially used the Blodgett method. But AISC Design guide 1 says:

The plate and the concrete may not be in contact everywhere. So, there is no reason that we can assume plane sections remain plane. Also, the strain diagram may not be linear in the tensile part. So, the strain in the anchor bolt may not be dependent on the concrete strain. So, the anchor bolt force is assumed independent of the concrete stress.

What is you opinion on this.

Thanks for your time.

 

[ishvaaag]

Well it says "may not be dependent". Perhaps more accurate would be "may not bear some practical relationship to". Then "for practical reasons may be assumed", meaning that by the assumption given one delivers a safe design, irrespective of the concrete strain standing. But this is not exactly that is shown to be independent, but that is practical to forfeit evaluation of the relationship.

 

[Chandan321]

ishvaaag, I could not understand your logic. Could you explain clearly what you are trying to express.

Thanks

 

[ishvaaag]

Well, if you have a problem where the math becomes intrincate and so not very convenient to use, the problem is one where simple methods would be good, and you devise some simplified method that making omission of the intricacies of the general behaviour gives a sound design, then one can opt to follow such path. AISC may be saying that. We are doing this every day when we apply Coulomb's theory for earth push: just a convenient coverall proposition. In fact many of our cherished calculations are just such things, only that the beauty and perfection of the math involved makes them look more true to the matter. When the math is uglier, simpler and less -but enough- accurate solutions become more attractive.

 

[ChipB]

Pretension your bolts such that the residual force in the bolts at the extreme distance is greater than the tension the bolts would have under the design moment. Once you do that, you've essentially brought the resultant force back in such that e<L/6, therefore it's always bearing on the concrete. Remember you need to add those forces to your column reaction to make sure base plate bearing < 0.35f'c