response modification factor for concrete connection

[penpal97]

I am designing a bridge pier with columns and a cap at the top. The system works as a moment frame during an EQ. I perform an elastic seismic analysis based on the response spectrum to get axial, moment, and shear. Time to design the rebar...

The column rebar will be governed by the moment at the connection. Code says I can divide the moment by R=5.0 (lets say). However, connections must be designed w/ R=1.0. (AASHTO 3.10.7) I believe it is the same principal in ACI and ASCE 7, I just can't identify the appropriate section numbers.

How do you design the rebar? In the past, I've just use R=1.0 for everything. Now I'm trying to be a little more economical.

 

[ishvaaag]

According to

AASHTO LRFD Bridge Design Specifications, 4th Edition (2007)

table 3.10.7.1-2 response modification for (connections...)

Columns, piers, or pile bents to cap beam or superstructure

and

Columns and piers to foundations

R=1

TO is the key word here. My understanding of this TO is the shear friction transfer of forces and longitudinal forces in rebars for the connection, or if you find it more proper following the case, the elements of the strut and tie in the transfer need be all able for R=1

 

[penpal97]

Thank you for your thoughts. I generally agree with you. The sentiment of the code seems to be that connections should be capacity protected. Shear friction capacity definitely falls under that criteria.

The problem comes w/ the longitudinal rebar for the connection under moment. If I design the rebar @ M_eq/R for the column, but then have to check the rebar @ M_eq/1.0 for the connection, the rebar will be too small by a factor of R.

I think the code just doesn't have enough verbiage to explain what it wants. But since I have to design to what the code says, I am at a loss.

Everyone's thoughts are appreciated.

 

[ishvaaag]

For the longitudinal rebar this means that in the length of splice for the bigger force and rebar at the connection, the splice needs be designed as per the code; then that means that you will have in piers 2 rebar end regions and 1 central, the end ones meeting the capacity for R=1, that is my view. For tall enough piers, that shouldn't be a problem, nor for long enough piles.

On splices,

pier end with pile cap, as rebar with full capacity at R=1

pier end with central region of pier, as with worse at pier with R=5 or corresponding R, or maybe with engineering judgement you need not even take the max value at end with R=5, maybe even less is sufficient, given the point where the splice is to be made, more inwards than the ends.

Ans so on in piles with pile caps, or piers with pile bents.

 

[penpal97]

If I am understanding you properly...

For dowel bars that come up from the footing and into the column, design the dowel bars for bending w/ R=1.0.

Then for the smaller main column bars that lap w/ the dowels, design main column bars for bending w/ R=5.0.

The lap will be governed by the smaller (column) bar. I guess since the dowel is the connection "TO" the footing, it is OK if the lap fails from R=1.0?

And if I have a short column with a large dowel, then i just end up carrying the large dowel the entire height of the column?

Does anyone else also have an opinion?

Thanks in advance.

 

[ishvaaag]

Yes, you could dimension as you say the splices "on a permissive understanding" of the clause code. If restrictive, for short piers you should go with the larger section for the whole height, and for tall enough piers then define 2 end sections with sections with demands in accord with R=1 (the larger), and with splices for the inner section as you say, dimensioned with the forces for R=5 and at point of splice, since already in the column and "no" notional connection. This of course if the pier tall enough to make such changes worth.