base plate analysis

huangyhg

base plate analysis
i am currently studying to take the struct 1 and struct 2 exams.  i bought a practice exam for each from the ncees and on the struct 2 exam there is a question that leads to designing a baseplate and the anchor bolts for a vertical load and an applied moment.  from a couple of different books i have (and the way i have done it in practice) provide for analyzing the loading on the baseplate as the capacity of the concrete stress block for the allowable bearing of concrete (a uniform load over a small portion of the end of the baseplate).  the ncees solution assumes that the load is triangular in nature with the centroid of the load occuring at the exact location of the flange of the steel column.  i can understand both methods, but they give very different solutions.  the stress block method gives you a thicker baseplate, but a smaller tension force in the anchor bolts.  with the ncees method (for lack of better term) you get a much thinner baseplate and larger tensile forces.  are these both accepted methods?  is one of them preferred? are they both right?  any advice?
akastud
check out our whitepaper library.
i have used two different methods to design base plates.  the one that assumes the center of the load to be under the flange is a relatively good (in my experience) approximation that is easy to do.  a more exact solution assumes the stress distribution is similar to a reinforced concrete beam.  in this case i use the solution by blodgett in the "design of welded structures" which involves the solution of a cubic equation.  with the use of mathematica (and i suppose mathcad) this is no longer difficult to do.
there have been several other threads on this site during the last year with similar topics and may be worth trying to find.
regards,
-mike
mrmikee,
thanks for your response, i have also used blodgett's equation (but i can guarantee you that i will not attempt it on the exam), but all of the design materials i have assume a uniform stress block that is not at all similar to the two methods you describe.  i have searched the other topics, but i have not seen one that talks about the uniform stress block.
akastud
akastud,
yes, i believe both methods are acceptable.  i suppose if you do get a base plate problem on the exam, you should do it the way it is shown in the solutions manual.
daveatkins
akastud,
i didn't notice in your post the part about uniform load.  i have not seen that method myself but i stay in the asd world almost completely.
-mike
mrmikee,
the stress block method is much like a concrete beam where you assume you get a uniform compression load (assumed to ease calculation) over a portion of the compression zone.  in the baseplate analysis it is similar in that a small uniform compression area occurs over some distance at the dge of the plate for the full width.  thank you both for your responses.
akastud
i am currently studying for the exam also and i worked that problem a couple of weeks ago.  frankly i was hacked off about the problem because i couldn't see how you could just assume that the column and base-plate would pivot about the center of the flange.  this would have to assume that the base-plate is flexible enough to allow this.  i just feel this is not a great assumption.  i have always used the concrete stress block approach with successful results but i am not truely satisfied this is all that accurate either because of the flexible nature of the base plate redistributes the load.  anyway if the problem shows up on the exam i guess i will solve it like the solutions manual, but i think that is a bad problem to ask on a multiple choice exam because it requires you to make a pretty important assumption which could be easily disproved.  just my 2 cents.
the best reference that i have come across for base plate design is:
"design in structural steel" by john e. lothers
chapter 6 (p 241): moment-resisting base plates and footings
i think the assumption of a linear stress distribution is more reasonable than a uniformly distributed one, especially if you are designing at service-level loads.  the coefficients for the whtiney stress block used for beam design were based on tests on beams, not on base plates or footings, so how do you know what to use or how big to make your uniform stress block?
have you checked out the design guide for column base plates by the aisc, design guide 1?  it has some good information, but again its all back to assumptions.  the aisc method is not fool-proof either.  your best bet might be to "play the game" and use what the ncees recommends for your test, and your own judgment in practice.
i think for any problem on the seii exam if you state your assumption used weather its conservative or more exact (liberal) you should not be penalized as long as you justify the reason.   note i say should.