Post-tensioning tendons - proximity to opening

[hope9010]

There is an existing opening 51" x 24" thru an existing post-tensioned slab. They want to increase this by 1.5" each of the 4 sides, at each of a number of floors, for a larger size duct, so that the opening will become 54" x 27" (see attached). We have had x-rays taken of the tendons. We have to allow some tolerance for the tendon location, when interpreting the x-ray.

Question:

How close to the opening can the tendon be, given the curvature of the tendon? Is there any industry standard for this?
(We realize that we should try to maintain at least the required fire protection, but our question is asked from the perspective of the tendon breaking out of the concrete due to its change in direction).

 

[hokie66]

KootK,
Thanks for the laugh. As for acronyms, I hate them. By they way, what does GOBS stand for?

 

[Ingenuity]

In the interest of full disclosure...

...and based upon some recent attachment/sketches you have posted on company calc pad you currently (or previously) work for Canadian consulting company that I used to work for back in the late 80's/early 90's.

Small world, eh!

I don't miss the Canadian winters!

 

[KootK]

@ Hokie: You won't think it's so funny when I fly down under and land on your doorstep looking for work! Gobs = lots and lots. It was a mistake to capitalize it for emphasis. Very confusing when discussing TLA's.

@ Ingenuity: Totally not fair. I demand to know your location and employer this instant! Don't make me call HR. Due to my promiscuous nature, I have calc pads at home from pretty much every major structural firm north of the 49th. A few US outfits too. But yeah, you got me.

The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[Ingenuity]

@KootK - I hear Peter has let the reigns go to Mike V. Is Bob (the founder) still around?

I reside these days in much, much warmer climates...

 

[KootK]

Nope. No founders left. Did you know that we now have a rather largeish parent company?

The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[Ingenuity]

I did...good to see the original founder's name is still 'front and center'.

Sorry for hijacking the thread hope9010/ajk1...

 

[BAretired]

KootK,

I dislike acronyms too. Attached are a few definitions for SCM:

http://www.acronymfinder.com/SCM.html

When a person uses an acronym he really should define it the first time he uses it. I often find that I know the term intended but can't think of it off the top of my head. And some people use acronyms incorrectly. One example I can think of is RCC which hokie and I have discussed before. It should stand for Roller Compacted Concrete but some people think it stands for Reinforced Cement Concrete. Is there any other kind? But when you look up the acronym RCC you get this:

http://www.abbreviations.com/RCC

which suggests they are both right. I give up!

BA

 

[BAretired]

And by the by, KootK, I have no objection if you can convince the administration to allow anonymous posting. You have provided some good reasons in this thread for trying it out.

BA

 

[KootK]

Right? How was I to know that we weren't discussing Snow Cone Machine admixtures? Eschew Obfuscation I say!

The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[rowingengineer]

KK,
Happy to have a discussion on number 1 of your list, but the term that comes to mind is "average bond" strength. Also strut - tie analysis can provide some insight into the nodal points that the code often have requirements for that aren't clear in a moment analysis.

http://www.nceng.com.au/

"Programming today is a race between software engineers striving to build bigger and better idiot-proof programs, and the Universe trying to produce bigger and better idiots. So far, the Universe is winning."

 

[KootK]

Thanks Rowing. As much as I'm dying to get into this, it deserves it's own thread. Someday soon, I'll get my thought ducks in a row and post that thread. I'll give it a title that you'll recognize in the hope that you'll participate. For now, here are some, teaser questions for you:

1) In traditional, non-STM, models of corbel design, the primary tension steel is only required to be embedded ld or ldh into the column. If the column were only ld+cover wide, and it's often close, this arrangement would never satisfy an STM check. The primary strut would be too steep. The same is true of end span beam/column joints. How can that be okay?[pre][/pre]

2) Imagine a large block of unreinforced concrete on the ground with a single piece of rebar sticking out of it. Now use a crane to grab the rebar and lift the assembly off the ground. Size the rebar to handle the tension. What will be the required anchorage length here? Most folks will say ld. I think that's wrong. It's an important case because that's exactly the condition at shear wall footings at the tension zones.

3) In an elevator shaft coupling beam, how far into the walls should the beam reinforcement extend? Many folks will again say ld. I say more, much more. Seismic codes seem to agree with me here in that they often specify some multiple of ld. But then, that seems like an arbitrary improvement.



The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[hokie66]

KootK,
Look forward to your new thread. There have been some threads where the distinction between anchorage and development has been brought into the discussion, but I can't remember one exclusively devoted to that topic.

 

[rapt]

KootK,

In a corbel, I would never stop the bars ld + cover into the column. Always to the back of the column and then cogged around a transverse bar as is required by strut tie logic. Maybe people are being misled by shear friction logic. Interesting that Shear friction logic does not appear in other design codes other than ACI!

In 2) above, the problem is in the "unreinforced". The bar you are talking about needs to develop Ld past the transverse bars that should be at cover to the face of the concrete. Unfortunately if it is unreinforced, you do not have transverse bars. So now it becomes an exercise in bar pull out and a cone of failure and is relying on concrete tension. This is not a standard development length calculation.
There is a similar problem with all fixings and how they connect to concrete members. The fixing people do it by test. But this only gets the load into the bottom of the member. All concrete design is based on loading at the top of the member. So reinforcement perpendicular to the face is required to transfer the load from the fixing to the top (far) face and it needs to fully lap with the fixing to transfer the force correctly. One important requirement in this case is that the fixing fully overlap with a transverse bar at the face where it is connected and that hanging shear reinforcement also overlap with that same transverse bar to ensure that there is not a zone where the force transfer is relying on concrete tension. i.e. to ensure that there is a reliable tensile load path from the load to the far face (compression face) of the member.

 

[KootK]

Rapt,

In a corbel, I will also run the reinforcing to the far side of the column and turn it down a fair ways. Most people realize that it needs to be more than just a standard hook. The point that I'm trying to make is that, using traditional methods, the thing that you check is development length past the face of the column. Why check that if it's not sufficient to satisfy the STM model? And why don't we check the hooked end of the bar for it's ability to contain the strut that it so obviously contains? It should be a curved bar node STM with a radius on the hooks that doesn't result in excessively high bond stresses.

I disagree with you on #2 Rapt. Even with top steel in the concrete block running parallel with the top face of the block, it is still very much a concrete breakout situation.

I also disagree regarding the fixings. At least, if I understand you correctly, I do. Fixing tests should be based on un-reinforced -- and in may cases cracked -- concrete. They should ultimately be capable of converting tension into some form of pullout cone that works. Once a pullout cone is established at the bottom that works, all pullout cones further up the member would work as well and any tension will have been effectively transmitted to the main body of the member.

For what it's worth, shear friction appears in the Canadian concrete code as well as ACI.

KootK

The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[rapt]

KootK,

I hope if you have a significant load being applied at the bottom of a concrete, you are providing "hanging reinforcement" to transfer that load to the top of the member. Similar to the way you would in a steel member.

The whole concept of the truss analogy for shear requires this. It assumes that all loads are applied at the top surface and if not then they have to be carried to the top surface using reinforcement. It cannot be carried there by concrete in tension.

 

[hope9010]

I think this has wondered far off my topic, but thanks for the comments about my original question. Much appreciated.

 

[hope9010]

Please no one take offence, as none is intended, but may I make a gentle request - if you are going to continue with this string, but along lines that are not relevant to my question, perhaps you could start a new string, as every time you send a new message it shows up on my email. Thanks. Much appreciated.

 

[KootK]

@Hope9010: no offence taken. Rather, please accept my apologies as the thread drift is mostly my fault.

@Rapt: I'd like to continue our discussion. I'll start a new thread right now.



The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[hope9010]

Kootk - not a problem.