high moments at column/base 1

[oneintheeye]

hi, when modelling column loads on a foundation slab and the column loads include a moment the model will show high moments local to this applied load. How do you cater for this? I have had a few suggestions, some seem resonable, some i've used in the past but would like to hear the experts view (you guys).

thanks.

 

[csd72]

This is why most frame bases are modelled as pinned it is often difficult to manage the fully fixed loads..

 

[oneintheeye]

i agree that the frame should be pinned ideally. However this is not our design (the frame) and there are high loads on the kind of frames which I guess (and would hope) that the steel designer is designing economically. I have also on occasions seen details which are certainly fixed when looking at upgrade work etc.

 

[Ussuri]

The Steel Construction Institute used to have a method whereby you model dummy members at the column base as a method of taking account of a partial fixity.

 

[jike]

I am not sure what you are asking. How to analyze; How to detail base plate, anchor bolts or foundation; steel of concrete frame?

 

[oneintheeye]

sorry. I am looking at the foundation itself. Usually piled. In a model I would get very high moment local to load. i.e. the moment applied at the column position is twisting the plate in the model significantly.

 

[Orthotropic]

Are you looking at frame columns sitting on a mat foundation? A large pier/pile cap? If you are modeling a mat/pile cap in SAFE, for example, how much do the moment magnitudes depend upon the size and aspect ratio of your elements?

(slightly off topic) When you have a "pinned" baseplate connection in a moment frame, how do you avoid a "soft story" at your first story of a multi story building?

 

[rmurph17]

Ok I take it your talking about the baseplate:

so assuming that the load from the column is uniformly ditributed and completely distributed through the baseplate to the foundation. It would be helpful if you were to state what the column section size was, the moment at the base, the axial load coming down the column and the orientation.

I could help you as far as designing the baseplate, holding down bolts and pad foundation if you could suply me with the information.

You will probably need to create a moment resisting baseplate using side plates or a concrete encasing.

Also what is the bearing capacity of the soil, the cohesivness of the soil or if it is on top of a pile any details on the existing pile and pile cap

R. Murphy
MICE MIStructE MASCE

 

[csd72]

herewegothen,

Are you talking about the concrete footing/pile cap under the concentrated moment?

To take care of this, bring your anchor bolts down to below the horizontal reinforcement and put a plate at the bottom.
You then have a truss with anchor bolt verticals, reinforcing bar horizontals and a diagonal concrete compression strut.

http://www.gostructural.com/userfiles/file/pdhse1007.pdf

 

[oneintheeye]

there seems to be some confusion here. I have received column loads from steelwork designers. these include moments. When using a model for a piled or even ground bearing the moment will cause a massive moment in the top of the element. Caused i would think by the fact the element in model is a thin plate. How do others rationlize this moment, otherwise you'd have massive reinforcement in top. this is not job specific at the moment. Just occured a number of times in past.

 

[csd72]

Dont treat it as a moment, treat it as a push and a pull on your anchor bolts. This will reduce the effect somewhat.

Refer my previous post, sorry this is not to UK standards but the principal still applies.

 

[oneintheeye]

yes I will have a look through that. Like you say, principle of forces would still apply. Thats one approach I've used in past, up down on bolts. Even then I think the forces generated seem high. although maybe its just a good example of why we dont 'fix' columns.

 

[rtmote]

If the moment is applied about your major axis, then you can do the push/pull (through flange action) on the bolts with bending on the baseplate. If it is on the weak axis then it can design the baseplate for bending (prying action).

You have a choice to put the thickest baseplate (through cantilever projection and limiting stresses) and transfer all the moment to the bolts or, go for a thin plate and allow for elastic deformations which tends to the pinned base approach.

The problems start when you 'design' for a pinned base and you actually have a fixed detail. The questions are, when is a pin a real pin? And is a fixed base really fixed?


Robert Mote

http://www.motagg.com

 

[oneintheeye]

im not designing base plate im designing foundations. i think the steel designer is probably covering both bases maybe

 

[rtmote]

For slab foundations design, if pile are pinned and simple, I often model piles as compression-only springs. The details for 'fixed' embedded piles are difficult to achieve in practice, to the satisfaction of the geotechnical experts.

Do you have horizontal loads as well as moments? One reduces the available capacity for the other.

Robert Mote

http://www.motagg.com

 

[oneintheeye]

I'm struggling to understand why people are not understanding. I am designing the foundations. I am not fixing piles they are 'pinned' supports. The column loads onto the pile cap/foundation are given by others. This includes a moment. When applied to a FEA model. High moments are shown in the top of the elements at the 'column/load' position. This I believe is due to the program not fully recognising the thickness and stiffness of the element. I am aware how to model the piles themselves. This is a question of designing the structural element of the foundation slab.

 

[OldPaperMaker]

herewegothen,
If you are talking about designing a spread footing or pile cap that has an axial load,P and a bending moment, M then you assume a footing size, calculate an Area, A & Moment of Inertia,I of the footing, calc the min & max brg pressure, q from P/A plus or minus Mc/I. Then calc shear & bending moment in the footing. Finally calc the required reinforcing.
You will have to do a few cycles to get the soil bearing pressure to the allowable value that the geotech supplies.
Most reinforced concrete text books have examples of this process. I don't think that there is any way that a FEA program will do this for you.
Regards

 

[rtmote]

You nailed it I think. FEA is lmited.

Engineers should only be allowed to draw , and on a beermat, with a pint in hand.

Robert Mote

http://www.motagg.com

 

[oneintheeye]

right i want to make this clear here. I am well aware how to design a footing, pile cap etc. Occasionally we have a large raft structure which has a number of piles. We use (as many other companies/engineers do) a FEA program to calculate loading on piles where there can be a high number of piles with columns and other loading that is not in any was uniform. To calculate by hand would be a huge undertaking. Other than to design separate pile caps for each column/loading area which will not be as economic. We would produce a model that has pinned or spring supports to model the piles and input the loads. This is where I think the program falls down in that if the columns have moments (not ideal I know but these are sometimes supplied to us). On the comment on FEA being limited. It would not be used to calculate bearing pressures on a ground bearing slab, only moments shears etc. This is described in amny textbooks and the theory is based on a slab on springs to model subgrade reactions. moments generated then will depend on the stiffness of the soil and the stiffness of the soil relative to the slab.

 

[JLNJ]

Your FEA program doesn't like it when you apply moments to the madel? What program are you using?

Maybe you need to define a finer mesh.