Precast Slab - Transverse Reinforcement 2

[astructurale]

I am designing a 12" deep precast slabs that span over a creek at a residence to resist HS20-44. (2-10ft wide, 20ft span, 2" composite topping) I have designed it for flexural capacity and shear, and it meets the deflection and crack criteria. How do I specify the transverse steel? Is there a minimum required by ACI 318-99? The only examples I can find are for transverse slab design when you have stems supporting the slab, and this is not a stemmed member. Or is stripping and handling the only case I need to design for in the transverse direction?

 

[JAE]

You would probably just use the minimum temp/shrinkage reinforcing required in ACI. 0.0018 x Ag if I remember correctly. Look in Chapter 7 or 8 for it.

 

[jike]

The transverse steel requirement is based upon the bending moment required to spread out the concentrated load(s) laterally. Using the shrinkage temperature steel requirement of ACI for transverse steel would be way underdesigned!

If I recall correctly, the transverse steel requirement is as much as 70% to 80% of the main reinforcement.

 

[JAE]

jike....huh?

I've designed hundreds of slabs for a multitude of loads with temp reinf perp. to the main reinforcing without any problems....but is this a requirement due to the fact that its a "bridge" - with wheel loads?

 

[jike]

The transverse reinforcement is to resist the bending moment which develops laterally as the concentrated loads try to dish the slab. I would not use heavy transverse steel for uniform loads or small concentrated loads.

Westergaard studied this problem in the 30's as wheel loads on bridge decks started to become significant. The same principal holds for forklifts on building slabs but is generally not well understood by building engineers.

When I get back to work next week, I will calculate the required transverse steel by the AAHSTO formulas and the Westegaard method and post the results. If you have an FEM program, you can see what transverse bending moment develops. I am confident that it is more than the T & S reinforcement.

 

[astructurale]

I used the AASHTO formula 3-21 in section 3.24.10.2. Someone on another thread gave me that tip. I came up with ~22% since my main reinforcement is parallel to traffic. I ended up with #7 at 12" for my transverse and #10 @5" for my main steel. My design included a 1.3 impact factor. I found that the worst case was the lane loading (0.64klf) with the point load of 18kip as specified by AASHTO design requirements. An experienced production manager said he thought that my slabs were over designed, but I figured I would rather be over than under for a case like this. (FYI-slabs are connected by grouted keyway.)

I really appreciate all of the discussion on my question. Also, I am interested to see the results that jike will be posting.

 

[JAE]

Your maximum reinforcing for your section is about 3.5 sq. inches per foot. (that's at 75% rho balanced). This is based on

f'c = 5000 psi
fy = 60 ksi
d = 11.5"

With a uniform load of 640 plf and a midspan point load of 18 kips spread out over a 12' lane, I get a service live load moment of 13.3 ft-kips/ft and a dead load moment of 8.75 ft-kips/ft.

For the d=11.5" case, I get a required As = 0.70 sq. in/ft.

Your #10 at 5" provides 3.05 sq. inches per foot! Way more than you need.

 

[astructurale]

I think I know where my difference is. I understood the 0.64klf as per foot of load lane, so for a 10ft slab I used 6.4klf. Isn't that incorrect for bridge design?

So I get (1.3)*410kipft for lane loading plus the concentrated load, plus topping 12.5kipft = 545.5kipft.

Then adding in the 1.4DL and 1.7LL factors,
923.6kipft for superimposed moment. (It adds up.)
For a one foot strip that would be, 92.4kipft which is way more than your 13.3kipft.

f'c=6000psi
d=11in
fy=60ksi

With this design cracks in the slab is also something that I am checking, and it is at 0.13in.

 

[JAE]

The 640 plf lane load is for a single 10 foot lane - so for a 10 ft wide slab you would use the full 640 plf. The concentrated load is also for the full lane width (not one foot).

If you are designing a single 1 ft wide strip of slab, then you would use the following:

Lane Load
Uniform - 640 PLF / 10 ft = 64 lbs/ ft
Concentrated - 18 kips / 10 ft = 1.8 kips

Dead Load
14 inch total thickness = 175 psf
Uniform dead = 175 plf (for one ft strip)

Moment
from uniform LL - .064 x 20^2 / 8 = 3.2 ft-kips
from concentrated LL - 1.8 x 20 / 4 = 9 ft-kips
from dead load - .175 x 20^2 / 2 = 8.75 ft-kips

Total Factored moment (using the 1.4 and 1.7)

(3.2 + 9)1.7 + (8.75)1.4 = 33 ft-kips

Much lower than your 923 ft-kips. You should read through the AASHTO spec about the lane and recheck also the truck loads as these may control.

 

[jike]

There are a couple of different ways to approach the design of this structure depending upon the governing code. One way is to design it is with the ACI code load factors and the AASHTO loads. Another way is to design it all by the AASHTO (loads and load factors). I do not have the current AASHTO but would be surprise if this section has changed much. Another is to use Westegaard's method with either ACI or ASSHTO load factors.

I agree with JAE that the 640 plf and 18 kip concentrated load is to be distributed over the full lane width of 10 feet. According to AASHTO for this span the live load should have an impact factor of 1.3.

By various design methods, the area of main (bottom) steel that I calculate (assuming no mistakes) is (based upon
F'c =6000 psi and Fy = 60 ksi):

AASHTO Loads with
ACI load factors 0.89 sq. in. per foot of width

AASHTO Loads with
AASHTO load factors 1.14 sq. in. per foot of width

Westegaard method with
ACI load factors 0.83 sq. in. per foot of width

Westegaard method with
AASHTO load factors 1.06 sq. in. per foot of width

Please note that with the Westergaard method, I have only used the concentrated wheel loads and have not included any uniform lane load.

By various design methods, the area of distribution (bottom) steel is:

AASHTO Loads with
ACI load factors 0.20 sq. in. per foot of width

AASHTO Loads with
AASHTO load factors 0.26 sq. in. per foot of width

Westegaard method with
ACI load factors 0.37 sq. in. per foot of width

Westegaard method with
AASHTO load factors 0.46 sq. in. per foot of width

The distribution steel for the 20 foot span by AASHTO is 22% of the total main reinforcement (DL + LL). The distribution steel by the Westegaard method (for this particular problem) is 67% of the steel required for the main bending moment due to the concentrated loads (only) or 44% of the total main reinforcement (DL + LL). Please note that the distribution steel will vary with span.

I understand that the modern AASHTO code was based upon the information that came out of Westegaard's report for the U.S. Bureau of Public Roads in the early 30's. I do not know why the difference in distribution steel.

Westegaard's report is an excellent reference for anyone designing for concentrated wheel loads on one way slabs, whether it is for buildings or bridges. It's title is "Computation of Stresses in Bridge Slabs Due to Wheel Loads" by H. M. Westegaard, Professor of Theoretical and Applied Mechanics, University of Illinois, Urbana, Il. Your local library should be able to order a copy for you.

I hope this helps!

 

[astructurale]

Thank you! I have learned a lot from both of your postings. I also appreciate the book reference.

In this line of work there's not a day that goes by that I don't learn something new. Thank you for speeding up my learning curve on bridge slab design.

 

[jike]

I just realized that I forgot to take into account the DL on just the 12" slab before the topping is poured....so the area of steel for the main reinforcement should be a bit larger than that listed in my previous post.

Have you considered using a pre-topped slab similar to what is done in parking garages? This would prevent possible delamination of the topping over the years.

 

[astructurale]

Since you two are good at giving advice...
Are either of you familiar with Strut-and-Tie design?

I've been working through the example in the ACI318-02, and I would like to know if the method can be done for a uniform loading instead of the point load like the example shows. I was going to split it up into 3 loads. One at midspan for 1/2 total load and one at each end over the supports for 1/4 of total load each. Where I've gotten hung up is the node at each of the supports. Having that extra load coming in over the supports makes the face of the node at the support for the strut larger than the face of the node for the strut required at midspan at the top.

I seem to be running into a lot of new things lately, and I appreciate the advice.

Thanks again for all of your helpful comments on the Precast Slab design for HS20-44.

 

[jike]

Sorry, I am not up on the strut and tie modeling.

 

[JAE]

Me neither...sorry.

 

[tomcor]

I have a related question. I'm putting in a residential bridge that will also need to be hs 20. It is 30' x 12' wide. I want to use a prestressed hollow core, which they rate at psf. From your post here it seems I need a (640/12) 54 psf plank. Is this right?

 

[JAE]

No - a properly designed bridge is based on the more significant of two loading conditions. One is the truck wheel loads that are specific axle forces at various spacings, moved across the span - all this per lane.

The lane load is a 640 psf load over a 10 ft lane along with a concentrated load of a specified amount depending on whether you are designing for shear or moment.

You can't simply take a part of the HS20 lane load and convert it to psf. It takes a few more calcs to get there.

 

[jike]

tomcor:

The precaster's engineer should be able to design the hollowcore plank for the HS20-44 loading. Just tell them what loading you want it designed for.