Range-add load combinations and load cases

[dslyecix]

Hi everybody.

I'm working with a structure where I need to have a supported load of 1000 lbf anywhere. I will be converting this requirement into a 50 psf load over a 20 ft^2 area. My structure is such that the shell elements between secondary beams are essentially 20 ft^2, and so I will apply, in succession, a 50 psf load to each shell object along the length of my structure.

Now, I am interested in the moment and shear envelope of this analysis. That is, the maximum and minimum effects of puting the load on shell1, then on shell2, then on shell3, etc, instead of designing the entire structure to withstand 50 psf at one time.

According to this wiki entry this is possible. "For example, the bays of a multi-story building may be subjected to many different vertical load patterns. The beam in Figure 1 is divided into eight equal-length segments to simulate the structural configuration of an eight-bay system. Any combination of bays may be loaded. A separate load case is defined for the loading of each individual bay. Range-add load combination is then implemented to envelope the response from all possible combinations."

I highlighted the important part in bold. The way I interpret this paragraph, is that I create DEAD LOAD and LIVE LOAD load patterns, then create load cases like Load1 = 1.25DL + 1.5LL on Shell1 only, Load2 = 1.25DL + 1.5LL on Shell2 only, etc. Finally, I create the envelope load combination that will find the max/min values by running the various combinations, as described farther down the article: "For pattern-type loading, all possible load combinations are considered. An envelope is generated for response from the following combinations:".

The problem is, I cannot find anywhere how you "assign load cases" to individual shell objects.

I create the (seemingly global) DEAD and LIVE load patterns. I can define a load case as a combination of those patterns. I can define a load combination that is a combination of those cases... but nowhere in that chain can I say "THIS CASE APPLIES ONLY TO THE SHELL1 OBJECT".

The only way I can see getting around this is to create load patterns, one dead and one live, for each member of the structure (DL1 & LL1, DL2 & LL2, etc). Then I can have a load case Load1 defined as "1.25DL1 + 1.5LL1" and Load2 defined as "1.25DL2 + 1.5LL2". Finally, a load combination set to envelope can give me the envelope of Load1, Load2, etc. The problem is to do this, I must define twice the load patterns as shell objects, and then manually tell SAP that Shell1 is carrying DL1 and LL1, Shell2 is carrying DL2 and LL2, and also define EACH of those patterns individually across each shell, when in fact the loadings are all identical (DL1 = DL2 = DL3 etc).

I guess my main question is how do I accomplish "A separate load case is defined for the loading of each individual bay."?? I cant specify any individual object, location, or anything for the load case. If I load up all my shell objects with a global DL and LL, nowhere can I specify that they should be handled as seperate load cases.

PLEASE HELP! This is driving me crazy.

 

[dslyecix]

Another way to phrase this is that it seems you can only differentiate the loads applied to different shell objects at the pattern stage. I click on a shell object, and tell SAP which load patterns should applies. This is directly NOT what the wiki says, where you are told to define global load patterns, and then use load cases to let SAP know which parts of the structure are loaded for each case, all using the same load patterns previously defined.

 

[Guest262653]

According to CSI Wiki, automatic pattern live loading is not possible in SAP2000 and the only way is to do it manually, creating separate live load cases for each load configuration and then combining them as needed:

https://wiki.csiamerica.com/display/kb/Pattern+live-load+factor

However, SAFE has an automatic feature that accomplishes this:

https://wiki.csiamerica.com/display/safe/Auto+pattern+live

Hope it helps.

 

[dslyecix]

Thank you avscorreia, at least I know now I must do it manually. It's not too much of a hassle, it just seemed like there was a way to avoid it based on that wiki entry. A bit tedious but it works. For anyone else following along in the future, here's how I got it to work:

Load patterns:
Dead1 (I have a superimposed dead load that will be on the entire structure no matter where the live load is, hence I only need one pattern)
Live1
Live2
Live3, etc.

Load cases:
DeadC1 = 1.4*Dead1
LiveC1 = 1.25*Dead1 + 1.5*Live1
LiveC2 = 1.25*Dead1 + 1.5*Live2
LiveC3 = 1.25*Dead1 + 1.5*Live3, etc.

Load combination:
Comb1 = Envelope of DeadC1, LiveC1, LiveC2, LiveC3, etc.

I apply the live loads to my structure according to the defined load patterns and where I want them to be located. If this were a 3-span beam for instance, every shell/frame object gets Dead1, the left bay is loaded with Live1, the middle bay with Live2, and the right bay with Live3. These Live values are the same in my case (hence my original confusion about why I must define them separately) but they do not have to be.

To use the steel design function, you must remove the default design combinations and use Comb1 only (or others if you want). If the default combos are left on then you get compounded factors, at least the way I have done things. I believe it is necessary to define things this way, but perhaps I'm mistaken and you can introduce the factors later.