Serviceability Considerations 1

[KBVT]

I have been involved in a discussion with a coworker about the requirements for serviceability design. There is some confusion about what the required load combinations are for the checking of deflections and building drifts. I have a few questions that I was hoping to get some help on. All questions are referencing ASCE 7 load combinations.

1. In an LRFD designed building, are ASD load combinations applicable for serviceability checks (i.e. deflections, vibrations, etc.)?

2. In calculating building drift due to wind, it is it acceptable to use the ASD load combinations and substitute 0.7W for 1.0W in order to reduce the wind pressures from a 50-year return event to a 10-year return event?

3. In calculating buiding drift due to seismic, ASCE 7 explicitly states to use strength level seismic forces. Is it acceptable to substitute 1.0E for 0.7E in the ASD load combinations in order to use service level loads for other loading types besides seismic loads?

4. ASCE 7-02 has a commentary on serviceability considerations. In the commentary, several load combinations are presented in order to achieve a 0.05 annual probability of being exceeded:

D + L
D + 0.5S

D + 0.5L (creep or long-term considerations)

D + 0.5L + 0.7W (drift for walls and frames)

Are engineers utilizing these load combinations in the commentary for the calculation of deflections, vibrations, drift, etc.?

 

[JAE]

1. In an LRFD designed building, are ASD load combinations applicable for serviceability checks (i.e. deflections, vibrations, etc.)? [red]Yes[/red]

2. In calculating building drift due to wind, it is it acceptable to use the ASD load combinations and substitute 0.7W for 1.0W in order to reduce the wind pressures from a 50-year return event to a 10-year return event?
[red]Yes, I think you can use 0.7W per the IBC drift limit tables[/red]

3. In calculating buiding drift due to seismic, ASCE 7 explicitly states to use strength level seismic forces. Is it acceptable to substitute 1.0E for 0.7E in the ASD load combinations in order to use service level loads for other loading types besides seismic loads? [red]To determine seismic drift, you use 1.0E and multiply your deflections by the Deflection Amplification Factor Cd. For all other deflections use the service level loadings. So your max. deflection in a seismic event would be 1.0E x Cd plus other deflections. Keep in mind that seismic drift limits are different than wind drift limits.[/red]


4. ASCE 7-02 has a commentary on serviceability considerations. In the commentary, several load combinations are presented in order to achieve a 0.05 annual probability of being exceeded + LD + 0.5SD + 0.5L (creep or long-term considerations)D + 0.5L + 0.7W (drift for walls and frames)Are engineers utilizing these load combinations in the commentary for the calculation of deflections, vibrations, drift, etc.? [red]I don't know about this one.[/red]

 

[steellion]

4. For parapets and NLB stud walls, we use a 50-year recurrence (0.02 probability) for strength criteria and a 10-year recurrence (0.10 probability) for deflection criteria. That's probably what the ASCE "serviceability" load cases are for.

 

[Lion06]

1.) Yes, but vibrations has different loadings than service loading. See DG #11.

2.) Yes.

 

[KBVT]

So in review, the load combinations that are being used for wind drift calculations are as follows according to ASCE 7-05:

D + 0.7W
D + 0.525W + 0.75L + 0.75S
0.6D + 0.7W

For seismic drift calculations, the load combinations used are:

D + 1.0E
D + 0.7E + 0.75L + 0.75S
0.6D + 1.0E

The seismic deflections inserted into the load combinations above would include the Cd factor added to the deflections due to the other service loadings.

 

[Stazz]

FYI (and because I just figured this out a minute ago):

0.7W reduction = 0.84^2 * W

where

0.93 = wind speed conversion from 50 year to 10 year.

^2 = to proportionate wind speed to a force e.g. deflections are proportional to the applied forces which are proportional to the square of the wind speed.

 

[Stazz]

oops...

where

0.93 = 0.84

 

[slickdeals]

See Table C6-3 of ASCE 7-02 (Page 308). The conversion factors are approximate for V > 100 mph.

 

[asixth]

I will express my opinions on serviceability considerations but I am not using ASCE codes so there may be some minor discrepancies. However the engineering principles should be the same nonetheless.

1. Wind Service...My national wind code was updated in 2002 to a limit state format. The previous wind code provided separate wind speeds for both working design and service design. The 2002 (limit state) code specifies that a 2000 year wind event is used for ultimate design and a 25 year wind event is used for serviceability checks. The multiple (Vserv^2/Vult^2) typically comes out to 0.5 for the majority of designs. Converting from ultimate wind (2002 code) to working wind (previous code) Vwork^2/Vult^2) is 0.66 so service wind would have a similar load factor to what would be applied for a live load.

Not knowing ASCE-7 to convert from a 50 year wind event to a 10 year wind event would be roughly 0.7 if a similar model was assumed for wind speed calcaulations (AS/NZS vs ASCE).

2. Seismic Drift...Depends on how you are analyzing your structure for seismic events but for using the equivalent horizontal static forces, the force applied to the structure is reduced depending on the level of ductility and energy absorption of the structural system, so the calculated drifts should be increased by the inverse of co-efficient that was used to reduce the forces for this absorption.

3. Creep and shrinkage are permanent actions and their actions should not be reduced when making checks for service. They should be treated as a dead load and load factor of 1 applied when checking service. Regardless of how variable creep and shrinkage may be in nature.

 

[Lion06]

stazz-
That 0.84 factor is only for 90mph winds right? If you have higher winds, the conversion factor is different.

 

[slickdeals]

SEIT,
The ASCE 7-02 code has the conversion factor for various recurrence intervals and various wind speeds. See Table C6-3

 

[WillisV]

ASCE 7-10 will take away the conversion factors all-together by providing maps rated for different return intervals. They will be providing strength level return period maps to be used in strength level load combinations with a 1.0 factor just like seismic. Serviceability will be checked with velocities pulled directly off of 10 or 25 year maps (they are providing multiple ones).

 

[apsix]

asixth

Please refer to new thread I've started in 'AS/NZS Code' section.

 

[asixth]

Just to clarify my original post, the ultimate wind for 'normal' buildings should be for a 500 year wind, not 2000 year. When considering creep, 0.5*LL (50% of the design live load) should be considered as sustained load which contributes to long-term deformation.