precast stadium bleachers 5

[pepperoni]

Dear all,
my firm has a project where we are assessing the condition of the existing stadium structure. The stadium consists of precast bleachers supported by CMU walls. The building, in general, has a lot of issues due to the age and exposure to weather but the main concern is the excessive, noticeable deflection of the planks. The worst deflection due to self weight is approx. 1.5in and it varies throughout the structure. The plank spans 28'. I'm concerned that the planks are losing the prestress force. All planks are exposed from below, and not a single crack was observed during the inspection.
From your experience, is loss of prestress possible and would the planks deflect beyond the limit without cracking of the concrete cover? What would be the other reason for a 1.5in deflection due to self weight.
How would you reinforce or repair those planks?

 

[jayrod12]

I would consider engaging a local pre-caster to review. Perhaps you may even be able to determine who supplied them in the first place and have them review. .5" of deflection over 28ft is pretty close to L/240. I wouldn't be surprised if that is their response.

 

[TehMightyEngineer]

Wouldn't be surprised if the age (normal sag) combined with normal loss of prestress resulted in that kind of deflection. As jayrod indicated that's just shy of meeting L/240. So it's a large deflection but not too unreasonable. I'm surprised it didn't crack a little but with the prestressing it's likely holding any cracking together. How is the live load deflection of the bleachers? That's likely far more critical to evaluating the condition of the bleachers.

Best approach would be as jayrod said; try to find the original precaster or someone more experienced locally with stadium precast bleachers. Hopefully you can identify the reinforcement and prestressing used and calculate what the theoretical deflection should have been and compare it to actual to verify whether there is an issue or not.

Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

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

 

[pepperoni]

Measured 1.5in deflection is only due to self weight, the seating is currently unoccupied so no contribution from the live load. My understanding is that the camber in precast construction should account for deflections due to self weight so they are flat when installed. Actual 1.5in self weight deflection + 1" allowable live load deflection (L/360) would result in 2.5in which is way above any limit.
No existing drawings or design data exist.

 

[JAE]

pepperoni,

I had a very similar task some years ago with a series of precast double tees that spanned between load-bearing masonry walls and formed a small outdoor baseball bleacher structure.

My task was for me to verify that the precast tees were "safe". There was no historical source for the tees so we performed a load test per ACI.

This involved setting up a series of water-filled steel drums along the span, to load the tees above and beyond the designated live load and measure the various deflections from underneath:
Prior to test
During loading of the tees
After full loading of the tees
After 24 hours loading
Immediately after removal of the test load
After 24 hours from removal.

See Chapter 20 of ACI 318-11.

I didn't have any significant deflection issues like you have but this at least would offer a load-test for structural strength.

Check out Eng-Tips Forum's Policies here:
faq731-376

 

[TehMightyEngineer]

There is typically some camber in stadium risers but it's not much from my understanding. Plus prestressing camber undergoes sag as well as normal loss of prestress, so it can lose that camber over its service life.

Remember that for deflection per ACI 318, the total deflection check is for "that part of the total deflection occurring after attachment of nonstructural elements...". I would agree that this isn't a good design for a stadium riser but it may in fact meet code for total deflection. I'd load up a bleacher with a known live load and measure the deflection before and after that loading; that will establish the condition of the bleachers for deflection more accurately in my thinking.

Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

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

 

[BridgeSmith]

It seems everyone is assuming precast means prestressed. Do we know this is the case? I would be very surprised that a prestressed beam would sag under selfweight alone. Typically, the creep in the concrete would cause a prestressed beam to camber up over time, rather than sag.

In either case, unless this is a very old structure, and the planks are over-reinforced (concrete crushing occurs before steel yielding), the structural adequacy of the beams would not seem to be compromised. Because of the ductility of the steel, be it strand or mild steel, the concrete would be severely cracked before a tension failure. Using dimensions of the plank and an estimate of the loading and required strength, the reinforcement ratio can be estimated to determine if a possible over-reinforced condition exists. It's unlikely unless there are severe depth limitations (where a really shallow depth is required to meet physical size constraints). The most economical RC sections typically do not approach an over-reinforced condition, and I don't believe it has been allowed by code in situations where failure could cause injury for at least several decades.

That said, the sagging could present problems of functionality or perception. If the sagging is noticeable, it could deter people from using it, because it looks unsafe.

As far as repair strategies, I can think of only one that would not exacerbate the aforementioned visual perception problem, but it would likely exceed the cost of replacement - external post-tensioning. Replacing the planks is likely the least expensive option other than the "do nothing" option. If there are a large number to replace, precast beams can become very inexpensive.

 

[pepperoni]

HotRod10,
the planks are prestressed, there are two visible anchorage access holes, patched with concrete. This is typical for all planks in the structure. My concern is if there is a progressive loss of prestress force, how would one establish the rate of the loss? The load test would definitely help to confirm the current capacity of the planks but to predict the useful lifespan of the structure would be very difficult and the planks would need to be constantly monitored.
The structure is approximately 30 years old.

 

[TehMightyEngineer]

HotRod: You're almost certainly not getting a normal section of stadium bleacher to span 28 ft without prestressing. On thin prestressed slabs (which I would argue stadium risers approximate) the long-term deflection often outweighs the camber increase I've found. Plus, due to the asymmetrical design of these risers your camber wont be entirely vertical. It also depends how they stored the concrete prior to install, we sometimes yard precast pieces in specific ways to increase or decrease the install camber.

Agreed on all your other points.

Another possible repair/strengthening might be shoring up the panels and installing a composite reinforcement. Either that or installing new supports to try to cut the span down.

Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

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

 

[TehMightyEngineer]

Pepperoni: Do you have any drawings or ability to identify the reinforcement of the planks? Might be worth scanning the panels and/or performing a little destructive investigation to find the strand quantity and type.

Heck, if this is a big stadium might be worth taking out a existing riser and destructively measuring the strain in the strand as you de-tension the strands. It would also allow you to test their deflection in a controlled environment before and after the de-tensioning and let you destructively identify all the rebar in the panel. You could even core sections to establish the compressive strength of the concrete.

Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

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

 

[Ingenuity]

the planks are prestressed, there are two visible anchorage access holes, patched with concrete.

"Anchorage" as-in POST-tensioning anchorage, or patched strand 'torch-marks' after cutting, as-in PRE-tensioning?

Only 2 tendons in the precast unit?

Do you have a dimensioned sketch of the section you can share?

and destructively measuring the strain in the strand as you de-tension the strands.

TME: your thinking is that it is unbonded POST-tensioned?

 

[pepperoni]

TehMightyEngineer,
no, unfortunately there is no documentation regarding the design or installation.

 

[hokie66]

How thick are the planks? I assume they are pretensioned, but what are these 'access holes'? Any photos you can share?

 

[pepperoni]

please see attached sketch with dimensions and a picture.

http://files.engineering.com/getfil...5-b97a-6a09b3f03605&file=plank_dimensions.pdf

 

[TehMightyEngineer]

Only 2 tendons in the precast unit?

I'm assuming it's two "pockets" of strands. Similar to this PCI detail:

These pockets can be grouted and sealed over to protect the ends of the strands from corrosion. You see this a lot in double-tees:

http://www.oldcastleprecastspokane....ecastspokane/uploads/images/Pics/DSC_0001.JPG

TME: your thinking is that it is unbonded POST-tensioned?

Pretensioned.

No, I'm assuming you could chip away enough concrete cover on a strand to attach a strain gauge. You could then carefully cut the strand section adjacent to the strain gauge to detension the strand. I will fully admit I've never attempted to measure the strain in a strand during cutting and it may not be feasible. However, it shouldn't be any different procedurally then when we cut the strands between precast sections.

Scratch my idea about checking the post-detensioned deflection. You'll likely damage the section too much to make this check worth anything.



Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

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

 

[structSU10]

probably worth it to run though a calc, assume some initial prestress, and see what the long term deflection comes out to. PCI has coefficients and such to figure long term deflection. If you cant get close to the field measured value, then you can worry.

 

[Mjkkb2]

loss of pretension or creep and most likely both. Are the sections a double L-type like in the picture above or a sigle L-shape?

 

[TehMightyEngineer]

Ran some rough numbers with a lot of guesses. Assumed f'c = 5000 psi, (2) 1/2" dia. 270 ksi low-lax 7-wire strand stressed to a target of 31 kips per strand, assumed 60% relative humidity. Probably not conservative enough but it's just a quick check.

Looks like they prestressed very close to the centroid; so vertical camber will be minimal. The non-prestressed principle axis are only 7 degrees off of the horizontal so I'm mostly neglecting that effect for expediency, the adjacent risers and attachments will probably restrain the section to some degree. I roughed in the non-rectangular section pretty quickly and may have goofed on that in my calcs.

Unless there's something I'm missing the section is compression controlled. That tiny 6" wide block at the top doesn't give much area for compression in the section so the c/d_e value I calculated was much greater than 0.375. Again, rough calcs but I wouldn't be surprised if it was indeed compression controlled. I also don't think there's enough strength here if it is compression controlled. This would explain the lack of cracking if it was losing prestressing and would be very dangerous.

As for deflection, I came up with the following using PCI prestressing loss estimates:

Initial camber at release = 0.298 in
Deflection at erection = 0.340 in
Long-term dead load deflection = 0.422 in
Total D+L load deflection = 0.687 in

I didn't calculate the deflection using an effective moment of inertia as I should. But, if I'm anywhere close, there's something strange going on here if you're getting that large of a deflection.

Definitely worth more than my quick look. I fully admit I may have goofed in the math as I threw this together quickly. Still, hopefully this helps.

Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

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

 

[Ingenuity]

Looks like they prestressed very close to the centroid; so vertical camber will be minimal.

Assuming vertical-axis bending only, CG is 4" from bottom of section. Prestress looks to be in center of bottom flange, so 3" from bottom, so eccentricity of prestress, e = 1".

Assuming P_ef = 54 kips (say 2 x ¹/₂" strands), I calc an uplift due to prestress ONLY of 0.11" (upwards: ie P*e*L²/8EI). Compare that to a SW deflection of 0.45" (downwards: ie 5*w*L⁴/384EI), and the net result, based upon GROSS section properties is 0.34" (downwards). So there is NO camber.

P/A is 250 psi, based upon gross section - so it has some precompression. Bottom tension fibre is about 240 psi (tension), so theoretically NOT cracked under SW.

BUT, the significant SW of the section adds another 1000 psi of compression to the top fibre, and with such a small compression area, creep is significant.

So creep after 30 years, applied to a downward deflecting section, would be say 4 to 5 x elastic deflection and 0.34" becomes 1.5" over time.

Pepperoni: Do you know if the concrete is light-weight?

Also, PCI 7^th edition design aid table 3.12.11 on stadium risers, may be of some use:

...and note the bottom flange thickness of only 3.5" compared to your 6"...you have a lot more SW!

 

[BridgeSmith]

Ok, I apparently made a few false assumptions. I was thinking these were deeper rectangular beams with tension steel near the bottom. It seems the prestressing may be closer to the middle (unless the strands are harped so that they are closer to the bottom at midspan).

Assuming TME is correct about the section being compression controlled, I retract my previous assertion that a lack of cracking indicates it is not in danger of failing. As TME points out, in a compression controlled section, brittle failure without warning is a possibility.

Given this configuration, where the sections rest on each other, midspan shoring of every 4th or 5th section may be adequate.

Without any info on the design or reinforcing of the sections, doing a load test as JAE suggested, would be a prudent move.