Model Truss Bars as Pins?

[David Deck Guy]

First, I need to buy an old 1950s textbook on reinforced concrete so I can figure out what sort of design considerations these guys, or women, were making.

More immediately important though, I'm having trouble getting a 1950s building to pencil out for some new loads. The beams I'm looking at have typical truss bars, bending at about the quarter points, as well as bottom bars that go all the way across. Some of the beams have top bars as well, but the top bars typically only go as far the quarter points.

Here is my issue. My beams are failing at the location of the bend. One obvious option is to try to reinforce the beams, maybe with FRP or channels on each side. But my question to eng tips is can I just let the beam fail at this location and model it as a pin, basically just forcing the inflection point where the original designers probably expected it to be?

 

[DaveAtkins]

Instead of allowing it to fail at the quarter point, can you redistribute moments as permitted by ACI?

DaveAtkins

 

[David Deck Guy]

To be honest, I was black boxing this a little and chugging it into RAM Concept, but I will look into moment redistribution and see if that allows me to model and consider this RC roof as the original designers considered it.

 

[phamENG]

If you're trying to figure out why those bars are there, it's for shear. Bottom reinforcing is less useful near supports (lower positive moment or negative moment if continuous or fixed), so they'd bend half of it up to resist shear near the supports. The bent bar intersects the diagonal shear failure planes at roughly a 90 degree angle, holding it together.

But no, you can't let it fail. You can try redistribution as Dave suggested, though I don't know enough about that to speak on it.

 

[David Deck Guy]

There are beams in the schedule with no stand alone top reinforcement, so I assumed the "bent bars" were to provide bottom reinforcement in the middle, top reinforcement at the ends, and have the bends located at where they anticipated inflection points to be. The issue is that with state of the art FEM, as well as adding some roof deck loads, the considerations about locations of the inflection points are not correct.

Maybe another way to ask the question: If a part of the beam (bent bar location) is detailed with no moment capacity but high shear capacity, should I model that point as a pin? This is just forcing the inflection point assumption.

 

[jayrod12]

I agree with your most recent post in regards to intent of the bent bars, they're to add additional bottom strength where needed, and additional top strength where needed over the support.

I agree a moment redistribution review will solve your issues.

 

[phamENG]

Yes, the bent bars also provide top steel over supports. But no, you would not model that as a pin. A pin assumes an idealized free rotation at ultimate loads. You can do that with steel with a properly designed and proportioned connection, but to assume that in the middle of a concrete beam is to assume a nearly complete failure of the beam.

 

[David Deck Guy]

Thank you for the feedback. I'm trying to think outside the box on this one, but don't want to stamp something that stands under ultimate loads but has MASSIVE cracking and serviceability failures. I'll probably explore attaching channels to each side of the beam.