Heavily Loaded Pipe-Arch Culverts 2

[bsantana]

I have a problem concerning very heavily loaded pipe-arch culverts under which normal assumptions appear rather conservative. This is based upon comments from a researcher who tested large span culverts with shallow cover. He related a test on a 24 ft span culvert with 16" cover supporting a 108 kip axle load without failure or serious deformation.

I have a 6 ft pipe-arch, 55" rise with 2 ft of cover and 110 psi tire pressure (46"x70" contact area, 4 tires) and a total axle load of 1406 kips. At the top of the culvert, using 2:1 footprint spread of the load, the resulting pressure is about 7.7 ksf uniformly distributed over the full span and length of the culvert. I am looking for a way to better determine pressures at the soil-structure interface using a concrete mat over the culvert area at the surface. It works fine if I use a slab-on-elastic foundation linear analysis, but to start I have just used a simple approach to determine the spring constants and done some bounding of the problem. I have varied the K's over the culvert from 5-10 ksf/in above the culvert to 67 ksf/in at a diameter away from the culvert, using 30 ksf/in in between.

Pipe-arch manufacturer's are not inclined to deviate from their normal approach of adding cover depth. This is not a practical option in this case, therefore the concrete mat approach. Additionally, a worst case scenario has 2 of the pipe-arches side-by-side with about 3 ft of space between them.

I have found reference to CANDE, a culvert analysis and design program, which is much more rigorous theoretically and appears to have some value for my analysis. I purchased the user's manual from McTrans and it will work for the single culvert case. With appropriate boundary conditions, I believe that it will suffice for the double culvert case as well. My question is: does anyone have comments pro or con about the software and experience with the somewhat esoteric soil and interface elements (for a stuctures type) used? Also, is there something out there that is better either for a manual solution or computer solution. Thanks for any help.

Barry Santana

 

[mcwebb]

Have a look at new finite element software called CandeCAD which is an update to CANDE.

The web site address is:

http://www.ssismint.com

I have used CANDE for over 8 years now and I know it very well. It is very powerful but time consuming to use. CandeCAD is much faster and easier to use.

I have been involved with conducting full-scale tests on a metal arch culvert under shallow soil cover and subjected to heavy axle loads. We found it difficult to model the live load effects accurately especially at shallow soil cover.

 

[BabbageGeo]

Barry,

I suggest that you check out the support conditions for the pipe arches and the potential for movement of the fill between and supporting the arches. I suspect for the heavily loaded situation you are looking at minor variations in field performance and control of the backfill and the ground to which the loads are being transferred will significantly affect the deflection of the pipe arches ad teh slab.

If the arches are more flexible than the mat/slab the deflected shape of the slab could be considered as an indicator of the variation in loading imposed on the arches.

Regards

 

[bsantana]

Thanks for the reference to CandeCAD. I am going to check it out. Looks promising. Thanks.
Barry

 

[bsantana]

I have been searching for CandeCAD. The web site seems to have changed. I am trying to determine current availability of the software and how it integrates with AutoCAD.