Multilinear springs - how to define them properly

Hi,

I'm working in STAAD.Pro V8i

I need to define a support with the following:

FX  - multilinear spring

FY - fixed

FZ - multilinear spring

MX - fixed

MY - fixed

MZ - releazed

So far I successfuly used "FIXED BUT" with Linear springs for FX and FY where I defigned KFX and KFZ as 16666kN/m but when I tried to define a multi-linear spring I could not assign it and it was not recorded in the file.

'Help' topic on this subject is not very clear. Could I assign Multilinear Springs in such way that they will substitute previously defined KFX and KFZ or 'FIXED BUT" and "Multilinear Spring' do not work in conjuction at all? Than, what degrees of freedom actually got assigned when I use 'Multilinear' ?

Could you, please, help me with this problem?

Thank  you,

Len Bogdanov

 

  • Dear Team,

    Apparently, this topic does not initiate a lot of interest but I'll try again; let me tell you about my 'achievements' sofar...

    I assigned MultiLinear Spring(s) "on top" of "Fixed but" supports (FX and FZ) and they do work (thou the pictograms at supports changed to old-style rectangles).

    To run analyses I had to erase "P-DELTA 33 ANALYSIS" and put "PERFORM ANALYSIS" instead.

    Statements in the 'run' sequence:

    "++Re-analyzing for bi/multi-linear spring support"

    and

    "++bi/multi-linear spring converged. Iters = (number of iteration)"

    convinced me that mulilinear spring are legitimate and considered in the analyses.

    My next question is: Can I put for each lateral spring both positive and negative deflections?

    I need it because the Pile head can deflect both +FX and -FX as well as +FZ and -FZ

    Than, 'k' values would decline symmetrically at both positive and negative side of the curve and the spring would fully and more realistically represent pile behavior.

    Thank you in advance

    Best Regards,

    Len

  • First, I have to clarify myself  whether multilinear springs are a series of tangent lines or secant lines of soil load-displacement curve?  This hasn't be defined very clearly in STAAD manual.  I understood they are tangent  approximation at various displacement rather than secant one starting from origin.  However, if tangent approximation for soil load-displacement curve is true, there is limitation that the resultant multilinear spring can not be negative.  The negative tangent line can exist for strain softening soil.  

    With respect to the input of springs to the negative deflection,  I did it in my calculation and compared with assigning springs to positive deflection.  It showed that assigning springs to both directions resulted in a higher deflection of pile head and the result better agreed with LPILE's result.

  • Hello All

    I am not sure Bogdanovl is discussing Shallow foundation of Deep foundation.

    I think STAAD Pro processes the analysis based merly on mathematic FEM for stress-train/defection normally of elastic state of material. STAAD Pro helps users with reduced time of crazy calculation while application of it to the realistic structures is by the user.

    For soil-structure interaction of Deep foundation (eg. Pile), we may adopt the researches by prof. Lymond C. Reese and by Cox 1974 to build p-y curves for soil stratum, then use the modulus of subgrade reaction (p/y)*(width of member) as input in STAAD Pro. This one is an iteractive procedure until the unit reaction and deflection resulted by STAAD Pro converge to the p-y curve. The LPILE program recommended by Forrest0 is a good one. Another one of its family is ALLPILE.

    Rock cheers

    Tuanhai