RAM SS P-Delta analysis Load Factors 2

[SteveGregory]

The Frame program in RAM SS allows you set a factor either for the Mass or the Dead and Live loads. If I understand the literature and the program correctly, you should use 1.2 for the dead load and 1.6 for the live load almost always. This would apply regardless to your design method (ASD or LRFD) for AISC 360-05.

Straighten me out, please.

 

[strguy11]

Do you mean the scale factor in the frame analysis?

 

[SteveGregory]

Yes!! The new version 14.02 give the mass scale factor option or the Dead/Live load scale factors option.

Attached is a page from the manual that refers to it.

 

[strguy11]

Ok. I actually had a conversation with the head engineer for RAM SS at the steel conference a couple of years ago, and here is what he told me about the scale factor:

Since RAM SS does not do true P-delta analysis, it performs an approximate analysis. To do this the scale factor is used. Basically, it is a ratio of the (DL+LL)/DL. The LL can be the reduced live load from the code. If the DL and LL are about equal, the scale factor would be 2. Say for like a single story warehouse. As you get more floors, say for an office building, the LL's would be larger than the DL, and at the bottom, after summing all of these, the factor would be lower, say like 1.25.

If you are using LRFD, the equation should be (1.2DL+1.6LL)/1.2DL.

I should note that from my experience, we use about 1.5-1.7 for multistory, and 2.8 like is shown for single story warehouses.

If you havent been doing it right, dont stress too much. For braced frames, or shearwall buildings the delta is so small, that it really doesnt affect things too much. For a moment frame building, however, it does matter. But more than likely, it is only about the last 5-10% of the total moment that should be applied, at least that is what the RAM guys told me when i was doing it wrong! lol

 

[SteveGregory]

It looks like you must be using the scale factor associated with the mass.

Now if you use the new scale factors that are associated with the Dead and Live loads

Does this mean Use:
ASD Scale Factors > DL = 1, LL = 1
LRFD Scale Factors > DL = 1.2, LL = 1.6

It looks lke the new scale factors are more straight forward if I understand it right.

I attached another Ram Manual reference, this time from the Analysis section.

 

[strguy11]

It depends on your building, but I think for the mostpart you are right. I can think of a time when you would need to use a higher factor than 1.2 for LRFD. I will try to explain.

After looking at the way (and why) the mass option works it seems that the program used the mass properties you entered in for the loads as well as the material self weights for calculating the forces used. Since the masses you enter are based on DL, there is no "mass" for a LL, thus the need for the scale factor to account for the LL forces.

With the new load method, this appears to make it simpler, however, you still need to make sure you have accounted for any "extra mass" that might need to be included that could contribute to the lateral forces (and thus the P-Delta analysis), particularly in a seismic event.

Take for example a small, single or 2-story office building where the cladding (say mtl studs with brick veneer) are supported directly by the foundation. Therefore, there is no need to add a gravity DL to the beams. However, I always try to add a line load of only MASS to the perimeter of the building to account for this brick for use in the seismic weight and force generation. When I do this, the DL is zero, but the mass is something else.

Doing this in the old mass method for P-delta is fine. However, with the gravity option, you would need to scale the DL to account for this, since you would not have had a DL gravity force included in the line load, only mass.

If the brick was supported off the structure, then you would have an acutal DL gravity force called out, and then there is no need to scale the DL.

Does this make sense???

 

[SteveGregory]

My building is a 4-story building with a brick ledge supported by a spandrel beam at every level including a parapet at the roof.

So, I am thinking that I have everything covered for ASD using: Scale Factors > DL = 1, LL = 1

I might bump it up a bit for some glass curtainwall in a couple of areas.

 

[WillisV]

I will also try to shed some additional light on the subject:

RAM requires the user to input both dead, live, and "mass" loads. It uses dead and live loads for member design checks, and uses the mass loading for eigen value analysis (mode shapes etc).

For second order effects, RAM uses an approximate "one-shot" geometric stiffness based analysis as opposed to a rigorous iterative analysis. This is fine for most regular type building stuctures which RAM excels at designing - however - since it is only essentially running one load combination, the user is required to choose the correct "P" for it to use in the "P-Delta".

Prior to version 14.02, the P was based on a ratio to MASS loading - with 14.02 it can be based on MASS loading or load factors on dead and live (which is much easier to use).

I disagree with strguy11 in that it is my opinion you do not need to account for any extra mass in seismic events in this factor (i.e. it does not need to relate to seismic weight "W" in any way). You simply need to have the correct ratio of mass to controlling load combination - or with the new option just controlling load combination.

To make it slightly more confusing, technically you should vary this factor depending on which load combination is controlling the elements being designed, for instance if you are looking at seismic design vs. wind design.

For example, if your mass loads (seismic weight) were the equivalent of 1.1D, your live loads were about 0.75D, and the load combination 1.2D+0.5L+1.6W controlled the interaction ratio on your columns, the appropriate P-Delta factor using the "mass" option would be (1.2D+0.5*0.75D)/1.1D = 1.43. Using the new load combo option you would just enter 1.2 for D and 0.5 for L.

However, for that same building if 1.2D+0.5L+1.0E controlled the moment frame designs, your P-Delta factor would technically be (1.0D+0.5*0.75D)/1.1D = 1.25. Why 1.0 instead of 1.2 for the D? Because ASCE 12.8.7 allows seismic p-delta to be based on load factors not exceeding 1.0. With the new load combo option you would just enter 1.0 for D and 0.5 for L.

Clear as mud right? In general, using the new load combination values with 1.2D and 1.0L will always be conservative for most frame designs, and unless you have moment frames the effects will be negligible. With flexible moment frames you might want to consider looking at separate factors for wind and seismic as indicated above to squeeze every bit of capacity you can get out of the frame, but most likely drift will control anyway.

 

[WillisV]

A follow up point that SteveGregory's last post reminded me of - if you are using Direct Analysis your P-Delta loads should always be based on ultimate load combinations (which can be estimated at 1.6*ASD per Appendix 7), not ASD load combinations. So you would still use LRFD type combos for calculating your P-Delta factor.

 

[SteveGregory]

I got an email from Bulent Alemdar, SR Software Product Research Engineer\Senior Developer at Bentley that I will pass on to everyone. For ASD, it looks like 1.6 is the factor to be used for both DL and LL in the new version.

"When P-delta is performed by an iterative method the specification requires that you perform the analysis with the combinations, with all of the loads multiplied by an additional 1.6 factor, then the analysis results are divided by 1.6. Since the P-delta effects are nonlinear, this results in higher design moments than if you didn’t multiply the loads by 1.6 and then divide the analysis results by 1.6.

Note however that because RAM Frame uses the Geometric Stiffness method, and uses Superposition to combine the analysis results, there is a different process. The 1.6 factor needs to be applied (by the user) to the loads used in calculating the Geometric Stiffness reduction. Technically when the user intends to design using ASD he should use a factor of 1.6 on both the Dead and the Live Load; in the General Criteria command if the user selects the Use Gravity Loads option the scale factor should be 1.6 for both Dead and Live, if the user selects the Use Mass Loads option the scale factor should be whatever he would normally use (to scale the mass numbers to represent both the Dead and Live loads) multiplied by 1.6. It is not then necessary to divide any results by 1.6 because in RAM Frame the 1.6 factor is not applied to all of the loads, it is only applied to the P-delta stiffness modification. In this way the P-delta represents an ultimate P-delta, and this is what allows RAM Frame to use Superposition to combine load case results after the analysis."

 

[WillisV]

Yes - see my previous post...technically it is 1.6 X ASD combo so if you are looking at a load combo with 1.0D+0.75L+0.75W you could use a 1.6 factor on the D and a 1.2 (1.6*0.75) factor on the W.

 

[WillisV]

...err on the L not W.

 

[strguy11]

Willis V -

It looks as if you are correct, regarding the mass option. you would not need to include the mass weight of the cladding (unless supported by the structure).

I found this article which addresses the calculation:

Some programs allow specification of what is called “P-Delta scale factor”. This factor multiplies the effect of the weight of the floor that causes the P-Delta effects. The weight of a floor that participates in a secondary effect can be increased to account for live loads not included in the “mass dead load” by modifying the P-Delta “scale-factor”. The mass dead load is the total dead load weight that is supported by a floor diaphragm. The same scale factor can also be modified to account for factored load effects if LRFD code is used. It is instructional to note that the interaction of factored load effects in P-Delta needs to be considered as specified by the LRFD code [AISC, 1994].

For example, if 100% of the dead load is specified as the weight of the diaphragms and the applicable live loads are assumed to be 25% of that of dead load, the scale factor that includes factored load effects is given as below:

P-Delta Scale Factor = 1.2 + 1.6(0.25) = 1.6.

Since the mass weight is only for the total dead load weight that is supported by a floor diaphragm, if I place a mass line load to account for this in the EQ loads, I would need to "back this out", which would be the opposite of what I said above.

Thanks for the clarification.

As a side note, the mass option appears to be more straightforward to me! lol