AASHTO Load Distribution - Spread Boxes

[whatname]

Section 3.28 of the AASHTO Standard Specifications defines the methods for calculating the live load distribution factor for a prestressed concrete spread box beam superstructure.

Equation 3-33 of this section consists of a number of variables, two of which have defined upper and lower bounds.

The variable S is defined as the beam spacing, for which only the following values (in feet) are applicable: 6.57<=S<=11.00.

The variable W is defined as the curb-to-curb width of the roadway, for which only the following values (in feet) are applicable: 32<=W<=66.

Unfortunately, the code does not specify how to calculate your distribution factor should either or both of these variables fall outside these limits.

One option could be to use the method described in Table 3.23.1 for prestressed concrete girders (S/5.5), since they are in fact prestressed concrete girders, just with a box shape. However, because of the wide widths of the boxes, I can see where this might not be applicable.

Alternatively, it seems that the most conservative approach would be to calculate the distribution factor assuming the deck slab acts as a simple beam spanning between the box beams. This method (footnote f in Table 3.23.1) is defined for most longitudinal beam bridges with widely spaced beams.

My firm is currently working on the design of a small bridge with curb-to-curb widths and beam spacings outside of the lower bounds of both variables in question. A comparison of the S/5.5 method to the simple span lever method for this particular bridge results in a 23% higher distribution factor for the lever method.

Incidentally, we are using CONSPAN by LEAP Software to assist with the design, and the program (confirmed by LEAP support engineers) is using the lever method.

Have you run into this situation before? Any comments on this?

 

[Qshake]

You can find more about distribution factors by visiting the AASHTO library.

Regards,
Qshake

Eng-Tips Forums:Real Solutions for Real Problems Really Quick.

 

[broekie]

I had this exact case a couple of years ago on a bridge that we were designing. We met every critieria of the Standard Specification except the beam spacing. Our spacing was on the order of 6.4' if I remember correctly. We chose to use the equation anyway based on our engineering judgment. We checked the formula with both the 6.57' spacing and the 6.4' spacing, and the distribution factors were almost identical.

That being said, you didn't say how close you were to the lower bounds of those values. If you are on the order of 5' spacing, I might hesitate to use those equations.

FYI, the new LRFD code does not limit the width of the bridge for spread box beam distribution factors, and the beam spacing must be between 6 and 18 feet. However, it does say "For beam spacing exceeding the range of applicaabillity as specified in tables in Articles 4.6.2.2.2 and 4.6.2.2.3, the live load on each beam shall be the reaction of the loaded lanes based on the lever rule unless specified herein."

If I were you, and you were far outside the bounds of the standard spec equations, I would use the lever rule. True, it is very conservative, but I'd obviously rather err on that side than the other. I don't think the S/5.5 can be used. That assumes much more of an I-girder shape.

 

[edward1]

I designed a bridge with prestressed concrete box beams and composite concrete deck about one year ago. We used Staadpro 2005 to model the deck using elements and the graphic input to model the box beams. We used beam offsets to get the results of total LL+I moment on the beams and the individual LL+I moments on the beams. The results were that moments from the model were less than the moments using AASHTO distribution factors. We used the moments from Staadpro to design the prestressed box beams and the bridge is working just fine.