Substantial Openings in Double Tee 4

[KootK]

I'm seeking guidance and/or references for dealing with substantial openings in precast double tee flanges. I'd think this to be a pretty common thing but I've not yet hit the jackpot on google. Some starter thoughts:

- strut and tie as always. 2" flanges though. Not getting a lot of bar in there.

- design the stems across the opening pretending that there's no flange at all.

- even designing the member sans flanges, I still worry about the vestigial flange bits absorbing compression and buckling etc.

As with all things that get prefabbed, shipping and handling stresses inducing cracking is proabably as important as anything else.

Recommendations?

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.

 

[TehMightyEngineer]

If you can find a document giving extensive guidance on large openings in precast structures (wall panels, double tees, slabs, etc.) I would be a very happy engineer. I've found none in my searches other than basic design information and recommendations.

Regarding lifting and handling; the times I've done similar have been basically guided by PCI Design Handbook (PCI MNL-120) section 5.3.3.2 for members without cracking. This requires tensile stresses be kept well below the modulus of rupture during lifting and handling. Using this I've basically assumed that the reinforcement will not be utilized sufficiently and that I can consider only the plain concrete as an isotropic material. I then construct a 2D (or 3D) FEA mesh model of the element and get principle stresses in the material. If there is prestressing I approximate it in the FEA model by applying compression and moments to the mesh to simulate the prestressing forces. I then use pin connection supports with the self-weight + dynamic loading for the various lifting configurations I have to move the piece through. Keep in mind that the stripping strength is lower than design strength unless you tell the precaster otherwise.

For in-service loads where you have much higher stresses but no longer need to keep the panel 100% crack-free, I agree that strut and tie models are likely the best to model the stresses as they flow around the opening. I'd also try modeling the reduced section elements as individual (single "tee") prestressed members of just the stems with a bit of flange and see how they react for the design flexure and shear forces in the partial section regions. I wonder whether you'll exceed the allowable stresses for a prestressed member in these areas. I agree that buckling and weird things could occur which could cause some cracking at the sharp transitions.

I suppose some other general thoughts would be to put a good 45° chamfer at the corners to avoid such a sharp stress-riser. I'd also look at reinforcing the top slab with WWR mesh (assuming you weren't already) which will be much easier to torch out around the opening and also much easier to get lots of reinforcement in and around the opening without the usual grumbling about engineers putting too much bar in a piece (#6 bars @ 4" o.c. is terrible; but D25 @ 2" o.c and the complaints vanish...).

Have you ever done a 3D strut-and-tie model? If I ever got into something like this I'd probably look at picking up a copy of AStrutTie

http://astruttie.aroad.co.kr/

Ian Riley, PE, SE
Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

https://www.facebook.com/AmericanConcrete/

 

[EZBuilding]

Can you locate the opening so that it falls between two double-tees? This allows you to keep the majority of the flange intact and minimizes the impact to the double-tee.

 

[KootK]

Can you locate the opening so that it falls between two double-tees? This allows you to keep the majority of the flange intact and minimizes the impact to the double-tee.

Sadly no.

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.

 

[TehMightyEngineer]

Also, I'm assuming your picture was just an example and you're not actually planning to lift so close to an edge like that. If you're like me, the little ACI Appendix D devil that sits on your shoulder would be loudly complaining.

Ian Riley, PE, SE
Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

https://www.facebook.com/AmericanConcrete/

 

[KootK]

It is just an example but yes, we are planning to lift close to an edge like that. In the past we've put steel struts in between the tees to deal with some of the concern. Perhaps you can help me out with a couple related things:

1) Is it a bit deal to arrange lifting with a spreader bar are something to achieve vertical-ish ift loads?

2) How big of a deal would it be to lift closer in on the plank? Would that create control problems?

I'll ask these questions of my client too of course...

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]

When I've specified similar openings and reviewed the calculations from the precast flooring designer its just been treated by using the actual tributary width available for the compression block at the locations of the opening (compression located wholly within the in-situ topping anyway). Prestressed rib is then designed as it would be for a normal full unit. Apart from that, there is no special treatment required, its no different to designing a concrete beam with an opening adjacent which reduces the effective width of the slab at a particular location along the beam.

Perhaps if you need something to make you feel better, looking at similar treatment of openings in in-situ concrete or steel composite beam/slab construction as the same mechanisms are at play. I'm not aware of any special precautions for steel composite beams for example, you just take the available slab effective width and check the strength and percentage composite action at the critical opening location.

Just need to be a bit more careful when lifting not to twist the units as they are more susceptible to cracking longitudinally at the inside of the web and flange interface (dependent on how much flange remains).

 

[TehMightyEngineer]

I'd expect you are going to be lifting on an anchor set into the stem down to the prestressing strands (something like these: Patterson quicklift DTA anchors). They're very strong in tension but I've no good metric for how much of that depends on the flanges. I'd contact Patterson (or your precasters supplier) and try to find someone in engineering who is brave enough to help you come up with an acceptable capacity.

1) Depends on the precaster/erector. Likely not an issue but call it out in bold letters and design it for a little wiggle room off of vertical.

For one of our shops we would need to use a large spreader bar (or both shop cranes) and 2 small bars to achieve a perfect vertical lift on a piece this long (without fabricating a custom lifting setup). For our other shop we have a Combilift which lets them lift all sorts of crazy things.

Both shops would prefer a single bar to span the length of the piece with slings down to the lifters making a small off-vertical angle that unfortunately pulls toward the opening edge.

In short, spell out what you need but try to give the precaster whatever you reasonably can give them for leeway.

2) It wont be bad at all in my experience if you lift further in (Link and Link best photo examples I could find, I've done tighter). You have to really start getting close to the CG to get control issues. You'll pick up more negative flexure though which (combined with your prestressing) may cause excessive tensile stress on the top of your double tee. Also, your piece won't self-level as nicely and will likely lift a few degrees off level. This wont be too bad for a long double tee but other pieces can be quite sensitive to this during stripping (they pinch and snap off chunks of concrete if they bind up in the form during stripping).

Ian Riley, PE, SE
Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

https://www.facebook.com/AmericanConcrete/

 

[JoelTXCive]

For the lifting operation... Can things be done with temporary steel bracing/beams?

Or, (and this is really outside the box)....but what if you cast the section with a smaller hole and then saw cut once it is placed in position?

 

[hokie66]

I'm not a precast guy, but would think that 2" flanges are the biggest part of your problem. Make it like the one in the picture, and the issues largely go away.

 

[KootK]

@Agent666: thanks for you input. A solid, non-panicky perspective is always appreciated. That said, I'll take exception to a few things:

When I've specified similar openings and reviewed the calculations from the precast flooring designer its just been treated by using the actual tributary width available for the compression block at the locations of the opening (compression located wholly within the in-situ topping anyway).

- Keep in mind that there may be some calculation work happening behind the scenes that your precaster hasn't shared with you. I do this all the time. Approvals come easier if what I show the EOR is kept as simple as possible. And any voodoo that needs to happen prior to setting the plank isn't usually relevant to the EOR.

- At ultimate, I agree with the compression block tweaking business more or less. As hokie has suggested, it may be my wimpy 2" flanges that make that non-viable. I'll have to find out the real state of affairs with the compression block and report back.

- As TME has intimated, much of what matters here will be shipping and handling stresses. The compression block business won't resolve that and losnig large swaths of flanges will have a substantial impact on Sx_top etc. This is part of what you' may not be getting shown as EOR reviewer.

Perhaps if you need something to make you feel better, looking at similar treatment of openings in in-situ concrete or steel composite beam/slab construction as the same mechanisms are at play.

I'm gonna call a llittle BS on that. Not your BS so much as the industry's BS:

- the insitu concrete and steel systems typcially are much more continuous and redundant that precast. They are, therefore, more forgiving.

- frankly, the insitue concrete and steel openings probably should get more design attention than they do. To my knowledge, there's no special "don't worry about it" code clauses that apply to these systems. Trim bars 2' beyond the opening are really just detailing, not design.

- given a [40% x span] long opening next to an insitu steel or concrete beam, many designers will flat out disregard all of the slab on that side of the beam for composite action rather than deal with the complexity.



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]

Make it like the one in the picture, and the issues largely go away.

I agree but it's just not going to happen here. The original building and three additions all went up with 2" flanges. I can't allow the optics of the situation to be "choose to work with KootK LTD and life gets harder". It probably does get harder but that needs to remain opaque.

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.

 

[TehMightyEngineer]

Approvals come easier if what I show the EOR is kept as simple as possible. And any voodoo that needs to happen prior to setting the plank isn't usually relevant to the EOR.

This is part of what you' may not be getting shown as EOR reviewer.

KootK, giving away our trademarked secrets!

Ian Riley, PE, SE
Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

https://www.facebook.com/AmericanConcrete/

 

[Agent666]

Hi KootK, well I won't take exception to you overthinking anything. Anyone's free to do that if they feel its justified. Often its the small things that no one considers that end up being the Achilles heel of many structures when it comes to real world performance.

In my view the precast flange is essentially glorified form work, needing only to support the wet concrete load, and function as part of the final construction in a composite manner to develop the ultimate capacity required of the final system. Also needs to support any construction loads and act as a slab unitl such time as the topping might be placed.

In this part of the world the compression block depth is limited to the insitu topping depth only and we do not use untopped units (whole different kettle of fish if this is your case). We often put spiral reinforcement in the rib to transfer the rib compressive force into the topping, and don't end up with reliance on the sometimes smooth top surface of the unit like you posted.

Interestingly and as an aside, in a recent 2016 earthquake we had some frame elongation in one particular modern (circa 2005/6 from memory) structure and several double tee units at different levels lost their seating and collapsed directly onto the floor below. They completely delaminated from the topping which remained above spanning across the opening, after seeing that I think typing the two together is a must with MRF's in seismic areas (even if your codes don't prescribe looking at effects of frame elongation or loss of seating, it was only introduced within our codes as a direct response to this collapse).

To take the compression out of the rib into the insitu flange (as longitudinal shear) a strut and tie truss mechanism in plan can be used and essentially provided reinforcement is the same in both directions within the topping then it works with a 45 degree strut arrangement (i.e. any tension or compression forces in the topping reinforcement is balanced by the same force orthogonal/across the beam axis). You can come up with similar strut and tie arrangements for working around openings one side or both if you like and justify reinforcing things. For a beam I think this falls into the main building designers court, for precast the forces are even lower than supporting beam moments typically (dependent on relative spans obviously), so naturally the topping reinforcing usually just works for the lower shear flows in making the precast composite with the topping.

I doubt in this part of the world the precast guys even consider it part of their design to justify forces around the slab openings, they simply neglect the slab width as required and design the rib accordingly.

If you work out the shear flow (the amount of load/distance being transferred out into the overhanging flanges), its pretty manageable for minimum levels of topping reinforcement for typical floor beams/ribs given the rate of change of the moment in say a WL^2/8 type moment profile.

Steel framing was just an analogy to help find relevant info on a similar subject, however I suspect there isn't too much on that either as you elude to. I think this is because there is nothing particularly complicated going on that cannot be justified by an appropriate consideration of the flow of forces around the opening if the magnitude of forces warrant it. I'd be interested if you found otherwise.

One other question, what would you do if you had to come along later on and put an opening in in that position if that opening had never been considered by the precast guy? You'd just check the unit/rib with a narrower compression block, what else would you possibly check? This type of thing happens all the time and I've never heard of any issues.

In this part of the world the precast guys just put in a bit of left over strand anchored to the bottom of the ribs and use it to lift the units. Might not be some pretty proprietary lifting point but it works I guess.

 

[TehMightyEngineer]

KootK: I took a look through the references in the PCI and CPCI Design Handbooks and found a few things. I've not read these so I have no idea if there's anything relevant in them but figured I'd share:

Johnson, Ted, and Ghadiali, Zohair, Load Distribution Test on Precast Hollow-Core Slabs with Openings, PCI JOURNAL, Vol. 17, No. 5, Sept - Oct 1972.

Savage, J. M., Tadros, M. K., Arumugasaamy, P., and Fisher, L.G., Behavior and Design of Double Tees with Web Openings, PCI JOURNAL, Vol. 41, No. 1, Jan - Feb 1996.

Saleh, M. A., Brady, P. A., Einea, A., and Tadros, M. K., Design and Performance of Prestressed Precast Reinforced Concrete Double-Tee Beams with Web Openings, U.S. Army Corps of Engineers, USACERL Technical Report 97, April 1997

I also found this in the CPCI Design Manual and the PCI Design Handbook:

The PCI equivalent is B_t + 2 inches.

Ian Riley, PE, SE
Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

https://www.facebook.com/AmericanConcrete/

 

[KootK]

Or, (and this is really outside the box)....but what if you cast the section with a smaller hole and then saw cut once it is placed in position?

Didn't see it coming but the contractor likes this. Thanks. The lifting loop anchorage business is settled nicely this way which is going to save me some effort.

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.

 

[Teguci]

I seem to remember a standard detail we used for the cut through stair detail in my parking garage days. The precast stairs were usually supported by a tube steel spanning between the tee stems at either end of the opening. On top of that, the perimeter of the opening had site cast concrete to form up the curb which would cover over any handling cracks. It'd also cover over the torsional cracks since we'd typically warp the end tees to/over the maximum recommended since drainage is more important than a silly little nonmoving crack that's getting covered over. If the opening is for something else though then I wouldn't want to put any exposed steel in there.

You could probably get away with an unreinforced flange area to be cut away in the field (perforate the edges - tear here), but I'd think a spreader bar for the lift would be cleaner and safer.

 

[KootK]

we do not use untopped units (whole different kettle of fish if this is your case)

Unfortunately it is an untopped system. I can definitely see how the topping would improve matters:

1) Straight up additional thickness.

2) More continuity and therefor stability to the flange "remnants".

3) Ability to utilize the flanges of neighboring tees to move stress around penetrations reliably.

Thanks again for all your input so far.

One other question, what would you do if you had to come along later on and put an opening in in that position if that opening had never been considered by the precast guy? You'd just check the unit/rib with a narrower compression block, what else would you possibly check?

In an untopped system, I would be inclined to check the following in addition to compression block:

1) That prestress doesn't overstress the tension side concrete evaluated at 28 day strength for the reduced cross section.

2) That compression can flow around the flange opening without excessive stress in the remnants or those remnants becoming unstable (just cantilevered from the stem in my case).

3) Possible impact on deflection.

I may well be overthinking this as I'm prone to doing that. Still though, I feel that prestressed concrete is a somewhat more dangerous mistress than CIP and, as a result, there are a few more ways that one can get themselves into trouble.


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]

@TME: thanks for the references. While I appreciate the effort, I'm not sure that any of them really address this:

Johnson, Ted, and Ghadiali, Zohair, Load Distribution Test on Precast Hollow-Core Slabs with Openings, PCI JOURNAL, Vol. 17, No. 5, Sept - Oct 1972.

Plank's really a different beast owing to it's high torsional to flexural stiffness ratio. I'd be reluctant to apply any of that to TT.

Savage, J. M., Tadros, M. K., Arumugasaamy, P., and Fisher, L.G., Behavior and Design of Double Tees with Web Openings, PCI JOURNAL, Vol. 41, No. 1, Jan - Feb 1996.

There are a few things out there dealing with web openings but, curiously, seemingly none on flange openings which I'd expect to be much more common.

I also found this in the CPCI Design Manual and the PCI Design Handbook:

But that's basically just saying that one ought not be a moron and attempt to put an opening into the stem, right? Surely, the figures don't imply acceptability of the 2" from a design perspective?



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.

 

[TehMightyEngineer]

While I appreciate the effort, I'm not sure that any of them really address this.

Agreed, but figured it wouldn't hurt if you had ready access to them. I agree that I would think flange openings to be much more common in TT construction.

Surely, the figures don't imply acceptability of the 2" from a design perspective?

Your guess is as good as mine (I would guess it does not imply acceptability, only guidance). As I said in my original reply, I've only ever been able to find generalized guidance on this topic and otherwise have had to fend for myself.

Thankfully I've only dealt with things like this on non-prestressed panels or composite prestressed bridge deck panels. Most normal designs keep the opening size small or just stop the panel short and use a header. I've never had to deal with large openings on an un-topped, prestressed system and I agree with your thoughts about how there's less room to fudge it with an untopped system. I concur with your three items to investigate and would expand on #3 to consider not only deflection but differential camber with adjacent planks.

Ian Riley, PE, SE
Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

https://www.facebook.com/AmericanConcrete/