Scaling of Response Spectrum Base Shear

Scaling of Response Spectrum Base Shear

1 of 6

Home

PostGen.

http://forum.civilea.com/thread-14897.html

ImageHost

FB!

Search

Members

Help

Advertisements:

Current time: 04-05-2012, 01:14 AM

Scaling of Response Spectrum Base Shear Subscribe to this thread

View a Printable Version Send this Thread to a Friend User(s) browsing this thread: timpalok

Pages (2): 1

2

Author:

Last Post:

Replies:

Views:

burhan1985

chrono146

14

1219

Vote on Google+: Recommend this on Google

Next »

[Problem] Scaling of Response Spectrum Base Shear 08-03-2010, 12:19 PM (This post was last modified: 04-03-2012 09:20 PM by usman. Edit Reason: N/A) (Print Post)

Threaded Mode | Linear Mode Post: #1

burhan1985 Junior Engineer


In ETABS (and other structural analysis softwares), we have a scale factor for scaling the Response Spectrum Base Shear to a factor of ELFP base shear according to several building codes. UBC97:1631.5.4 allows to equate the RS-BS to 90% or 100% of ELFP-BS, no matter, whether to increase or decrease. Whereas, what I understood from ASCE7-05:12.9.4 is that, it allows only to increase RS-BS if needed and not to reduce it if it comes greater than ELFP-BS. I want to ask whether one can define a scale factor less than unity, which is clear from UBC97 but not from ASCE7-05. I mean if RS-BS (with scale factor = 1) is greater than ELFP-BS, then shall we reduce the RS-BS to a factor of ELFP-BS ? Please keep in mind the mentioned clauses of both codes. Regards, BURHAN.

The following 1 user says Thanks/Agree You to burhan1985 for this post:

08-03-2010, 05:01 PM (Print Post)

Post: #2

04/04/2012 22:52


2 of 6


benny Semi Senior Engineer

RE: Scaling of Response Spectrum Base Shear

Using a scale factor equals to 1 has two purpose:

1. The response spectra need not be scaled to match the site specific spectra and structure specific criteria. 2. To obtain the unscaled RSA elastic response base shear.

The reasons for having to calculate scaling factors (sometimes less than unity) are two fold. 1. If a “normalized” or “acceleration” spectra was used to calculate the spectral results, the spectral results must be scaled to match the site specific spectra and structure specific criteria. 2. The UBC97 sets minimum values for the design base shear.

08-03-2010, 05:23 PM (This post was last modified: 08-03-2010 05:50 PM by kowheng. Edit Reason: ) (Print Post)

Post: #3



From where to obtain the site specific spectra? Will it be equal to ELFP base shear?

08-03-2010, 08:55 PM ((Print Print Post Post))

Post: #4

Semi Senior Engineer

Shout: EQ/Struct. Eng. RE: Scaling of Response Spectrum Base Shear

The response spectra (RS) used for getting your so-called RS-BS is based on general values for regions of a given area. This is a code-based RS and is different from the site specific spectra (SSS); if you want to obtain an SSS for your proyect, you should get them by either probabilistic or deterministic means. Note tha SSS is not provided by the code. On the other hand, ELFP-BS is based on a number of simplifications, which in turn are based on a number of hypothesis; you have to know and satisfy all of them so that your resulting analysis is reliable. RS-Dynamic Analysis will give you a more approximate look at the expected behaviour of the structure, both in terms of forces and displacements, than the ELFP-Analysis. So, if I am getting your doubts correctly, it sounds quite dangerous trying to scale down the results of your RS-Dynamic Analysis to those of the ELFP. As for the other way around, scaling up your results from an RS-Dynamic Analysis to a percentage of the calculated ELFP-BS is partly based on the fact that most codes accept the concept that an RS-Analysis is more complex and needs more expertise than what is required by an ELFP-Analysis, and accordingly a designer would be more exposed to make an error. In summary, most (but not always) of the time the code will not allow you to take full advantage of the RS-Dynamic Analysis, as most (but not always) of the time the RS-BS will be less than the ELFP-BS. If I misunderstood your question please let me know.

Information is not knowledge. Albert Einstein

The following 1 user says Thanks/Agree You to faros3000 for this post: RafikCGS

04/04/2012 22:52


3 of 6

08-03-2010, 09:27 PM (Print Post)


Post: #5



Then why UBC97 allows to scale down RS-BS to 90% or 100% of ELFP-BS. Refer to 1631.5.4? It clearly says, one can reduce RS-BS to 90 or 100% of ELFP-BS.

08-04-2010, 12:48 AM ((Print Print Post Post))

Post: #6

Dan888 Junior Engineer


burhan1985, From my understanding of the UBC 97 clause, you can scale down the design base shear of a RSA to 90% or 100% of a ELFP "...with the limitation that in no case shall the elastic response parameters be reduced such that the corresponding design base shear is less than the elastic response base shear divided by the value of R." - clause 1631.5.4. In other words, the minimum design base shear is the RSA base shear divided by the overstrength and ductility factor. I believe that performing a RSA will give you a more accurate design base shear based on having considered the modal effects on the structure. Because of this, i don't think you should be able to reasonably reduce the base shear obtained from a RSA to that of the ELFP if the RSA is higher.

Regards

08-04-2010, 07:34 AM (Print Post)

Post: #7



Dan888, you are saying that "...the minimum design base shear is the RSA base shear divided by the overstrength and ductility factor...". From where to get this RSA base shear so that i can obtain the minimum design base shear by dividing it by R? I mean with what scale factor in ETABS? Lets have an example: I modeled a building in ETABS, and now i want to design its members by RSA. How should I scale the RSA base shear in this case?

Post) 08-04-2010, 10:55 AM ((Print Print Post)

Post: #8

Batman07 Semi Senior Engineer

Shout: Keep sharing =)


Just to have a simple example but may help on your concern. It is recommended that the scale factor in the response spectrum function is defined as (g/R). In the example (see below figure extracted from ETABS),

04/04/2012 22:52


4 of 6


the scale factor is set to 9.81/5.5 = 1.78. Any subsequent scaling will be carried out in a separate load combination. Therefore a separate load combination is also defined for each spectrum case with an appropriate scaling factor, which is a function of the code required minimum static base shear. The two orthogonal scaled respinse spectrum combos can now be combined as a load combination using SRSS method (100% EQSX combo & 100%EQSY combo) and thus positive and negative output is automatically generated in the results from the combination. The code requires that the spectrum case results are scaled up to a minimum percentage of the equivalent staic force base shear. Note all results from response spectrum analysis are scalar as the vector effect is lost through the modal combination, in other words, for example, the correlation of sign convention of axial load and moment is lost. For this reason care should be taken in coupled shear wall design and also pile reaction distribution when using response spectrum analysis results.

"If I had eight hours to chop down a tree, I'd spend six sharpening my ax. "- A. Lincoln

The following 4 users say Thanks/Agree You to Batman07 for this post: kristoffer_89, joamirose, nander jr. Post) 08-04-2010, 04:56 PM ((Print Print Post)

Post: #9


04/04/2012 22:52


5 of 6



Dear All, Thanks for the the replies. But I m still unclear about the issue that UBC97 allows to scale down the RSA-BS while ASCE7-05 does not allow to scale down but only scale up if required. Why this huge difference is there in the two codes. Or if I m understanding the clauses of codes incorrectly then please correct me. I've summarized steps for RSA for ASCE7-05. If something is incorrect then please correct it.

1. Define a static quake case in desired direction in Define > Static Load Case. 2. Define a Response Spectrum Function in Define > Response Spectrum Functions. 3. Define a Response Spectrum Case in Define > Response Spectrum Cases > Add New Spectrum. a. Define Damping of system, generally equal to 0.05 for concrete buildings. b. Set Modal Combination to CQC. c. Set Directional Combination to SRSS. d. In Input Response Spectra section, enter the Function defined in Step 2 in desired direction (direction shall be the same as defined in Step 1), and enter an initial Scale Factor of gI/R. Where ‘g’ is acceleration due to gravity (9.81m/s2 or 32.2ft/s2), ‘I’ is importance factor, and ‘R’ is Response Modification Factor. e. Put excitation angle equal to zero. f. Put Eccentricity equal to zero. 4. Set number of modes to 10 initially, in Set Analysis Options. 5. Run Analysis. 6. In Show Tables, check the table: ANALYSIS RESULTS > MODAL INFORMATION > BUILDING MODAL INFORMATION > MODAL PARTICIPATING MASS RATIOS. Check whether the SumUX and SumUY of last period are above 90% or not. If not, then increase number of modes and re-run the model. 7. In Show Tables, check the table: ANALYSIS RESULTS > BUILDING OUTPUT > STORY SHEARS, for spectrum case defined in Step 3, and the static seismic case defined in Step 1. 8. Check Story Shear at bottom of base story for both cases (ignoring signs). This is the Base Shear for both cases. 9. ASCE7-05 defines a minimum value of spectrum base shear, which is equal to 0.85 times the static base shear. a. If spectrum base shear is less than the minimum one, then increase the scale factor defined in Step 3d by linear extrapolation, and repeat the analysis. Continue iterating unless spectrum base shear reaches the minimum value. b. If spectrum base shear is greater than the minimum one, then response spectrum analysis is complete in the considered direction.

I've problem with step9. UBC97 says to equate the RSA-BS to 0.9 or 1.0 ELFP-BS, no matter RSA-BS is greater or lesser than 0.9 or 1.0 ELFP-BS. Please clarify, have i understood correctly and whether my steps for RSA are correct?

08-09-2010, 10:37 AM (Print Post)

Post: #10

RafikCGS Semi Senior Engineer


Dear burhan1985 there is two ways for RSA-BS: scale up or not scaled, and never scale down. to accept this fact we need to understand WHY seismic codes order us to compare RSA-BS to ELFP-BS? I think a good answer is given in the second part of faros3000's post so for step 9. *************************** if you are using ASCE7-05: if V(RSA) < 0.85*V(ELFP) then multiply RS by the factor 0.85*V(ELFP)/V(RSA). this factor is greater then 1. if V(RSA) >= 0.85*V(ELFP) don't do anything if you are using UBC97: if V(RSA) < (0.85 or 0.9 or 1.0)*V(ELFP) depending in which Item of 1631.5.4 you are. then multiply RS by the factor (0.85 or 0.9 or 1.0)*V(ELFP)/V(RSA) if V(RSA) >=0.85*V(ELFP) don't do anything ******************************* you don't need to extrapolate and iterate calculations, when you need to scale up the RS there is a unique value. in this web page you can find an example with detailed steps for your problem

04/04/2012 22:52


6 of 6


http://www.structsoftlab.co.cc/main/reference/showpage.php?pageid=8

The following 3 users say Thanks/Agree You to RafikCGS for this post: theanhpe, RANA WASEEM, WASEEM ptqc06

« Next Oldest | Next Newest » Pages (2): 1

2

Enter Keywords

Search Thread

Next »

Quick Reply Message Type your reply to this message here.

Font

Text Size

Signature Disable Smilies Smilies

Post Reply

Preview Post

Go to Full Reply

View a Printable Version Send this Thread to a Friend Subscribe to this thread

User(s) browsing this thread: timpalok Contact Us | Forum | Return to Top | Return to Content | Lite (Archive) Mode | RSS Syndication Powered By MyBB, © 2002-2012 MyBB Group. Theme redesign and modified by MohsenG, origin by Audentio Design., © 2011. Proudly Using a Yaldaram.com Plugin.

04/04/2012 22:52

Scaling of Response Spectrum Base Shear

Recommend Documents