Autodesk Robot Structural Analysis

[#1]

Hi guys,

Any thoughts on the best way to model a suspension bridge?  I was thinking of modelling a cable element as segments that define a parabolic curve.  At the joins between elements I will add droppers to pick up the deck.  I've had a quick try but seem to be coming unstuck. 

It seems to me that this sort of model is actually not really possible, and maybe I should be looking at other options?

Tony

[#2]

Hi Tony,

attached the model of some example of suspension structure shown below:

The main things to remember when modeling and calculating such structures in Robot:

1/ avoiding or at least minimizing definition of cables by force or by stress - using definition by elongation or by length

2/ activating the 2nd and 3rd order geometric non-linear effects (activating "Non-linear analysis" and "P-delta analysis" check-boxes in parameters of load cases) to avoid instabilities for planar nets of cables or for collinear cables

3/ in case of convergence problems activating "Matrix update after each iteration" to use full Newton-Raphson method

---------------------------------------------
If this post answers your question please click "Accept as Solution". It will help everyone to find answer more quickly!

Regards,

[#3]

Love it. 

If I had a pylon at each end and the cable then draped back down, still possible to model like this?

[#4]

Yes, of course it is possible.

If you attach your model I will take a look.

Regards,

[#5]

Hi Pawel,

Thanks,

I havn't actually spent a lot of time developing this model as I got a bit stuck on the whole cable thing.  So basically there are probably many things wrong with it.  

Tony

[#6]

I have set it to the strating point of further modifications:

1/ "Cable" bar sections replaced by cables with "main cable" label

2/ "12mm bar" sections replaced by cables with "tendon 12mm" label, tension-only truss attibute removed from these elements

3/ then it was necessary to use additional 2 of 3 "commandments" mentioned in my previous answer (the first one was used when defining above cables with zero relative elongation)

With these modifications model is convergent. The displacements under DL1 (selfweight load and UDL on middle span) are exceeding 300mm upwards in middle span and 300 downwards in side spans.

Recommended:

a/ modification/completing of loads: adding UDL on side spans/ separating different loads in different loads cases. Remember that in structures with any cables the first load case ("assembling" load case) is automatically added to all consecutive load cases and combinations - the effect of it can be easily seen in sums of forces and reactions in the table of reactions

b/ all cables were defined with zero relative elongation - it means unloaded length of each cable is equal to the distance of appropriate nodes. When necessary it is possible to define and assign more labels of cables shortened or elongated (with negative or positive elongation respectively)

---------------------------------------------
If this post answers your question please click "Accept as Solution". It will help everyone to find answer more quickly!

Regards,

[#7]

Pawel thats great.  In the time since I sent you the model I have come to the same conclusion as what you've posted.

I've now set it up with the two draped cables 1.8m apart and all of the cross members on the deck.  Its working well now 

Thanks a lot for your help

Tony