Odd results from SAP2000 non-linear direct time history 1

[tubsdua]

Hello All,

I am working on a project to build a numerical model of a power transmission line which is being monitored. The model has been provided with wind load time history. The mean wind component is as per EN and the turbulent part has been generated by a random process using Shinozuka approach.

I have applied the load time history and performed the non-linear direct time history analysis. I first idea was to compare the extreme value of axial force (found by appropriate statistical method for probability distribution) in the insulator to the axial force due to the quasi static loads provided by the EN. On completion of the analysis it was noticed in the time history of axial forces in insulator were mostly in compression. This is quite unexpected.

I am attaching a very simple model with 2 spans of 1 conductor which is supported in the middle via an insulator. In this simple model also the insulator is in compression.

Can someone explain how can this be possible when the wind load is acting in x direction and the conductors displacement suggests that the insulator will be in tension?

 

[crisb]

Did you solve your problem? I couldnt get the file link to work, is it the $2k file? Maybe you need to upload again.

 

[tubsdua]

No there is no explanation still. I tried to simplify the problem more and made another model to test this.

A vertical beam element pinned at the top and a time history of loads applied at the free end. Chose the cosine function as time history available in SAP. Also applied a staic live load at the free end to check. The static live load gives tension in the member. Whearas the time history still produces a compression. This is strange. I am sorry but my version is SAP v 15 and I cannot save it as a lower version file. May be you can make this simple model above to test.

I also found a workaround to the problem. Instead of the beam element if I use a cable element in the previous model attached in original post (with conductor cables), the results are as expected. The insulator (cable element now) is in tension in the time history plot. The model is attached below.

Do share your findings.

thanks

 

[crisb]

Thanks for the second model. I couldn't download the first model, maybe you can post it again. I notice some of the cables are slightly offset from the yz plane.
In the cable element model the insulator appears to be fully fixed at one end, translation and rotation.
In the frame element model have you applied moment releases in all the frame elements and at the supports? Ideally you would also apply frame tension/compression limits although probably not necessary if always in tension.

 

[tubsdua]

Thanks for looking into my problem. I am attahcing the previous model again.
You are right that the cables are slightly offset from the yz plane. That is because in the complete model the arms in the supporting towers are not of the same length. It is a real time model being monitored with accelerometers and anemometers. Due to this same reason in the cable element insulator I had to fix the top as in actuals it is pinned to the tower. I could not find a way to give moment release to cable elements. As the cable elements is not supposed to take any moment it should not be a problem.
When I modeled the insulator as a beam element I had given moment releases in M3 and M2 directions.
However most surprising is a model with just one beam element pinned at top and with time history loads at other end as I explained in my last post. Here too the beam element is in compresion which is completly absurd. I am attaching this simple model also.
Hope you can assist me.
Best.

 

[tubsdua]

The simple beam element pinned at top and time history a other end is here.
Check the time history plot for the element.

 

[tubsdua]

I am sorry the file I posted at 12.02 is one with cable element insulator.
The file attached here is the one with beam element insulator.

 

[crisb]

Interesting, and strange. I wonder if your time step is too big, its interesting to try it smaller, plus it converges a lot quicker. You will see at some stages there are very big deflections.
I think the cable element is the best element here anyway, and if you put compression limits in frame sections it will have similar effect (stopping compression). There is also an option to overide individual section properties. Unfortunately until you have bottomed this out, I guess you don't have any faith in any of the other results. Personally I would be tempted to try a simple identical model, in another program and compare notes. SAP2000 can be very buggy at times, it may be worth asking them for a second opinion. Sorry I don't have a proper answer for you. It's possible the results are right and the bar is just springing back into compression, but just looks a bit odd.