Slenderness In Etabs

[WR1]

May be my questions seem stupid but just for the sake of disucssion today I have posted this to Mr. Arshad Khan, Civilea forum and SEFP forum. Looking forward to arrange all the answers by you guyz and write a tutorial

Dear all, I have following question regarding slenderness in ETABS

1) Why moment magnification method and 2nd order elastic method both are used by ETABS. I mean in ACI there are 3 methods to do (any of them) then why we put modifiers and then ETABS calculates internally the moment magnification factors.

2) If structure is braced then how to stop etabs from calculating moment magnification?

3) If drift limits are defined for example in ASCE H/400 then why is the need to perform P-delta. If we control the drifts to be H/400 still we need to perform p-delta?

4) We take wind drifts for 10 year return period. that is multiplying 0.7 to W of 50 year year..to convert this into service level....What type of modifiers are required to be put in ETABS to take drift or deflections on this 10 year Wind..

Serive modifiers Beam=0.5 & Column=1.0 or Ultimate modfiers Beam=0.35 & Column=1.0

I mean as per ACI we can multipy the deflections from ETABS to 1.43 if we use strength level modifiers?

[Badar (BAZ)]

first point:

Both methods addresses same behaviour. Moment magnification is a term used when we are taking second order effects due sway of structural system while elastic second order method(Euler) addresses slenderness of individual members. ACI code directs us to reduce stiffness of members when computing forces due to second order effects so ETABS requires us to specify stiffness modifiers and then it can calculates theses effects .

Second point:

Code uses Stability Index or Pu/Pc as tool to estimate magnified moments and second order effects can still be there in braced frame as there might be some slender member (Euler's equation is used in that case).

Third point:

Yeah one must check for slender members. I think, limiting values for drift control are not there to control structural damage instead they are for non structural damage(citation needed): hence it does not mean that there are no second order effects.

Cant comment on remaining material.

[WR1]

thanks baz, another confusion

In ACI R8.8 it states two options for use of stiffness modifers...

option a & b

Option a is what we use in ETABS

And in commentary it states that its intended for strength design but at the same time the deflections or drifts from wind or quake can also be read from the same model. So no need to make another model or to use another set of modifiers...Is my interpretation right?

[Badar (BAZ)]

Idea is to use the stiffness properties of crossection based on its expected behaviour in particular situation, as mentioned in commentary. Commentary gave the example of analysis under wind and earthquake situations.

In case of wind structure is intended to remain elastic under service loads so use properties based on gross crossection or 1.4 times the reduced crossection properties which are based on section 10.10.4.1)

In case of earthquake reduced crossection properties are required at service level to estimate drifts and related p-delta/second effects.

So in case of wind load you can check drifts based on gross crossection properties against unfactored loads. But P-delta effect may dictate final configuration of member and for that reduced crossection properties are needed. Remember P-delta effect is increased in the values of moments and shear forces due to gravity loads acting on deformed member shape. P-delta analysis estimates the increase in these values.

Another situation where one must use reduced properties is long term deflection of flexural members.

[WR1]

So it means for wind drifts one must use! Ig for beams = 0.5 and for columns =1.0

If seismic then use according to chapter 10?

[WR1]

k also tell me what is the load combination to check drift on both seismic and wind?

[Badar (BAZ)]

Yeah, except Ig=0.5 for beam is not correct value for checking drifts against wind loading. It will overestimate drifts; 1.0 (for beams) is allowed by code but one can use value closer to one, like 0.8 to have conservative estimates of drifts.

Drifts should be checked against all load cases involving lateral loads (9-3 to 9-7 which ever is applicable).

[WR1]

but those combinations are strength combination using 50 year return period. In ASCE commentary the equation is D+0.5L+0.7W

[WR1]

and what about seismic drift....what combination to use..i believe its only static load case that is only EQx, EQy etc..no combination....

[Badar (BAZ)]

I need to dig again not sure about load combination but I think you are right about only considering EQX and EQY with eccentricity.

[asadishaq]

In UBC section 1630.9.1 it is written that where allowable stress design method is used and whre drift is being coputed the load combinations of section 1612.2 shall be used..what it means that we shold check the drift for all load caes?? am i right??

[asadishaq]

I am doing the vetting of a project they made a seperate model for deflection check in which they use uncracked member is it the right approach???

[UmarMakhzumi]

You should go with cracked section properties as Section 1910.11.1 of the UBC provides modified properties to be determine lateral story drift. If your lateral force resisting system includes a shearwall, here is what yo should do, determine if the tensile stresses in the shearwall are greater than the modulus of rupture (7.5 *sqrt f'c) of the concrete (these would be the S22 stresses if the local x axis is in line with the global x axis). If so then the section is cracked. Use a value of 0.7Ig for serviceability. For the same model, apply crack section properites to shearwalls(use 0.35Ig or whatever is agreed upon) for strength design as they would reflect a true redistribution of loads.