No, not safe to assume walls take all the load, frames go along for the ride seismically and will have seismic induced actions in them. Depending on the configuration of the structure and how stiff walls/frames are seismic load cases could be critical for the frame design.

No you cannot separate walls like that, if they are built integrally they must be analysed integrally for all concurrent loading (potentially from both orthogonal directions simultaneously as required by your design code.

If you are designing a 6 storey structure via simplified 'hand' methods and doing a computer model to check this hand analysis, in my opinion you are going about it backwards. I'd design to the computer modelling, and check results by hand.

You imply a single core, this wouldn't be a desirable configuration (it probably fails any irregularity checks that might be required for you code (whatever that might be)).

Agent666 and I differ on some aspects of this one.

Quote (OP)

Is it safe or a common practice to let the core-wall take all the lateral load while the frames only carry DL+LL?

It's quite common and safe when executed correctly. A lot of high-rise buildings are constructed this way, albeit usually with closed section cores. As Agent666 mentioned, it's a torsionally sensitive system so that requires some attention. And, in high seismic jurisdictions, the non-lateral elements of the structure should be designed for a level of movement consistent with the drift of the lateral system again, per Agent666.

Quote (OP)

can we divide the C-section and analyse as if it were three planar walls where torsional strength is ignored?

Yes, and again, this is pretty common in my experience. In doing this, you're choosing to ignore the composite flexural behavior of the three walls. To an extent there's actually some technical merit in this approach for shorter buildings. Where walls are dominated by shear deformation instead of flexural deformation, the walls really will behave as three independent walls. At six stories, though, you're probably closing in on an elevator shaft shape that will have meaningful flexural response. It's also important to note that, just because you've chosen to treat the walls non-compositely, that doesn't mean that the system will have no torsional strength. In this case, torsional resistance will be produced by the two flanges of the channel acting independently but forming a torsional resisting couple.

Quote (OP)

what are the approximate formulas/equations used to roughly determine V(shear), M(moment) and T(torsion)?

With a concrete deck slab, I'd assume rigid diaphragm behavior and distribute your loads based on relative rigidity of the lateral resisting elements. This is a pretty common method that can be found in a lot of text books. If you're unfamiliar, let us know and we'll steer you towards something suitable. With the loads distributed, you're just back to statics to find the shears and moments on your individual walls.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

Agent666, I agree that the C-section should be analysed compositely for better results [when analysed by software] but would it be possibly practical for hand calculations to crosscheck computer output?
A friend in the team has told us that ASCE allows shearwall to carry 100% base shear while the frames are permitted to be designed for forces not less than 25% of total base shear. Not sure if we have interpreted the code properly. The building has regular plan with the core-wall centrally positioned. Thus torsional irregularity may be small.

Kootk, I am glad you have brought up your experience to this conversation. I am aware that codes require the frames to be designed in such a way that they meet compatibility requirement but, honestly, I haven't done anything this way. Probably I should this time!
I have done analyses with load distribution to each floor levels based on relative rigidity [of columns: SMRF] but I believe it requires a little different approach in this case. So please share a bit of your experience as guidance.
We thought it would be more complex to have closed section with elevator door openings.

Quote (Rick)

I haven't done anything this way. Probably I should this time!

If it's high seismic then yeah, you probably should. It usually just comes down to making sure that you're columns don't shear off under drift.

Quote (Rick)

I have done analyses with load distribution to each floor levels based on relative rigidity [of columns: SMRF] but I believe it requires a little different approach in this case.

Why? What you're doing by hand is merely a check on the actually design, right?

Quote (Rick)

We thought it would be more complex to have closed section with elevator door openings.

Agreed. I wasn't suggesting that you should go closed core. I was merely pointing out that, when single cores are used to laterally support very large buildings, they're often closed core-ish. Precision in language is important here. If you're sloppy, somebody will usually call you out on it.

Quote (Rick)

A friend in the team has told us that ASCE allows shearwall to carry 100% base shear while the frames are permitted to be designed for forces not less than 25% of total base shear.

That sounds like the dual system approach permitted under ASCE. It's real.. and good. But it would complicate matters as far as trying to replicate the numbers with an independent hand calculation.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

Hi Kootk, I am from Asia and learning English through these conversations at the same time...
So could you suggest any reference or share something related for our guidance?
The building is going to be in one of Indonesia's provinces where PGA = 0.3g-0.5g. We usually have about 5.0-6.8 earthquake magnitude.
Although computer outputs are technically better and close structure's actual behavior and resulting forces, we hope to be able to use approximate hand calculations to crosscheck if we have inputted and modeled everything properly.

Quote (Rick89)

So could you suggest any reference or share something related for our guidance?

I'm afraid that I can't really think of anything that would be targeted towards your particular questions.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

No worries Kootk. I have just collected some related references.

Whatcha found?

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

I found a sample calculation of shearwall-frame of low-rise r.c. structure in one of my old books. This example seems to use the approach allowed by ASCE as pointed out above. It is a good reference but rough analysis of C-section wall remains challenging without experience. Taranath book is also excellent but doesn't have detail coverage on c-section wall.

Maybe I do know of a useful reference. This delves extensively into hand methods for lateral and came out prior to the wholesale adoption of ETABS etc as our design tool for this kind of thing.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

Thanks Kootk, I just got this excellent book too.

Hopping in a little late... I haven't seen the building plan... any chance there is a stairwell at either end? Where are your stair exits located? and, how are they enclosed? In a high seismic area having a rigid core in the middle may cause some issues by introducing some torsional forces; a C shaped core may cause some other issues. Are you in a high seismic area? Is the building more or less symmetrical? and, more 'square' shaped rather than rectangular?

Dik

Quote (The building is going to be in one of Indonesia's provinces where PGA = 0.3g-0.5g. We usually have about 5.0-6.8 earthquake magnitude.)

Hi Dik,
It is a 20m x 22m building plan with 5m o.c. on x-axis, 5m o.c. y-axis and the remaining 2m is just in between 10m of the 22m long side. The corewall sits at the centre (x=10m, y=11m from any corner point). We do have two small exit stairs on each far-end (x=0, y=11m and x=20m, y=11m).
Could you please elaborate on the torsional issues and some practical solutions to those?

Any structure has both stiffness center and mass center, two of these do not coincide. Therefore a torsion will be introduced to the structure when horizontal ground movement occurs. When the seismic load is parallel to the two “flanges” of the “C” shape concrete wall, the torsion will be resisted by the two “flanges” (an equal and opposite direction force in each “flange”). When the seismic load is parallel to the “web” of the “C” shape concrete wall, there will no structural element to resist the torsion.It is really an not good general layout if the seismic design category is high. You may consider to build some interior wall parallel to the "web" of the "C" shape core.

Please note: this conclusion is based on the assumption that concrete wall can only take in plane shear.

Thank you so much Shu Jiang, what you shared has given me an insight to start doing it roughly.
Would this approach neglect torsional strength provided by the composite wall sections?
It would be really helpful if you could share a simple example for guidance or a reference that covers this approach.

The excerpt is from 2006 IBC seismic design manual. Hope it help.

Quote (SJ)

When the seismic load is parallel to the “web” of the “C” shape concrete wall, there will no structural element to resist the torsion.

I don't agree with this statement. The flanges of the channel shape are still available to resist torsion.

Quote (SJ)

It is really an not good general layout if the seismic design category is high.

I do agree with this one though.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

KootK, Could you present more detail why the wall can resist force perpendicular to the wall plane

Can the torsion not be split into a force couple that gets transferred into the "flange" walls?

I have read some papers discuss the torsional resistance of C-section (T-section) to lateral torsional effects but they seem to all use finite element (software) when composite action is considered in the analysis.
The books mentioned above are really helpful plus the IBC seismic design manual. By reading through these documents, I would say we are still on the safe side even the composite action is neglected. So the walls (3 walls) would be assumed to resist lateral forces independently, given that detailing follows properly.

Does your elevator core have concrete between the top of the door opening and slab soffit? In other words, will your C shaped core have a coupling beam to 'close' the core section.

I looked at a very similar problem a few years ago.

Link

Replying to the OPs original questions

- Yes, it is as long as the detailing and allowance for movement follows and is compatible with your assumptions
- Yes you can, as that's how many of the buildings were designed when it had to be done "by hand". Neglecting the composite effect of the "flanges" of the C is ok. It is done in my practice when we do things by hand.
- Others cited good sources on this one.

And just to add, I wouldn't expect that you get similar results with the SAP (or any other software). AS some say 'there are many ways to skin the cat'. Only because the software got you X results it doesn't mean that you made a mistake if you get Y results doing it on paper with a pencil. AS long as you make reasonable assumptions and have a direct load path with all applicable loads accounted for there's nothing wrong with your hand calc being off.

good luck