CANDE-2007 Culvert Analysis and Design

[pjlefeb]

I am a structural engineer, trying to find some help with CANDE-2007. It is a FEA program designed to study soil-structure interaction with respect to culverts. It is a public-domain finite element program written to satisfy a FHWA grant, so I would expect to see some forums on the web. No such luck.

Does anyone know if people actually use this program? I'm not going to bother writing up my question if no one knows the program, so let's start there.

Thanks,

Paul

 

[IDS]

Paul - I haven't used CANDE (although I almost certainly have downloaded it and have it sitting on my hard disk somewhere ), but I do use other FEA programs for this sort of analysis all the time. If your question is not specific to the software then I may be able to help, and even if it is I'll have a look at the software and see if I can help if I have time.

So please post the question.

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[pjlefeb]

IDS,

Thanks for the response. The issue is that I am trying to model 3-sided culvert with an arched top deck and arched elbows, but with straight legs. The analysis type I am using is Level 2. I can successfully model an arch culvert as well as a culvert with straight legs and deck, but the problem may be with joining the two types of geometry. The program tells me the arch geometry is not self-consistent, but I know it is because I can model it in Autocad with the same dimensions and it works fine.

I'm happy to post the input file if anyone wants to see it.

Thanks,

Paul

 

[pjlefeb]

IDS, another thing, you mentioned using other programs that can model the soil-structure interaction? Can you recommend a couple?

 

[IDS]

Paul - I have downloaded Cande now and will be having a look through it, but it's a complex program, and I don't have a lot of spare time at the moment, so it will take some time for me to become familiar with it. I'll post here if I find anything useful for your problem, but don't hold your breath! But thanks for drawing my attention to this program, it should be useful as a cross check to output from my other programs.

The main program I use is Strand7. The latest version has a module allowing easy modelling of a construction sequence, so you can place fill in layers. It's soil models are fairly basic, but if necessary I adjust soil properties at each step with an external spreadsheet through the Strand7 API. The latest version has a new Cam-Clay model, but I haven't tried using that for serious work yet. The other program I use is an in-house product of one of my clients, so is not available for general use.

The most popular FEA programs for soils applications (at least in Australia and Europe) seem to be FLAC and Plaxis. Both are well-suited to soil-structure interaction problems, but I prefer Strand7 because it includes everything within one package at a reasonable price (including 3D, dynamic analysis, buckling, thermal, etc), and with the API has the flexibility to do almost any task.

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[IDS]

Paul - by the way, could you post a sketch of your cross section.

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[IDS]

Paul - the documentation refers to the source code for CANDE being available, but I can't find it anywhere. Do you know if it is available for download, or if there is anyone connected with the project I could e-mail?

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[namklof]

Paul - from your description of the model, I suspect your model would require a level 3 analysis, were every detail of the model needs to be specified. For a new user, that would take a great deal of effort.

Doug - I don't believe the source code for the current version is available for download. The previous version was available, but it lacks many enhancements and has no user interface.

 

[pjlefeb]

namklof and IDS,

The most current version of Cande was released as part of NCHRP report 15-28. It is available for download here:

http://144.171.11.40/cmsfeed/TRBNetProjectDisplay.asp?ProjectID=408

Namklof, I ended up trying to do a level 3 analysis. The model passes the initial Cande compiling, but a fortran error stops the process. I've attached an autocad file of the culvert cross section I'm trying to model.

 

[namklof]

Did you check the Output Report (under the view menu) and the CANDE Log File? It appears there is an error in the imput but without some program diagonistics, finding it may be difficult.

 

[pjlefeb]

I believe I had checked, but now I'm not sure. I will check when I get back to the office on Monday.

Thanks, I'll let you know.