elvetos
Civil/Environmental
- Apr 18, 2011
- 2
Good evening, I'm modelling a retaining wall 20m high, with 5 steel anchor rows and a 0.80m circular beam concrete section for the pile. I'm using the Mohr-Coulomb soil model. The excavation takes place in 5 4m steps. I'm also using two contact pairs to model the interaction between the pile beam and the soil elements adjacent to it on both sides, using Pressure-Overclosure=Exponential and Small Sliding.The anchors are formed by T3D2 elements and are composed of tho parts (grouted and free).
The steps are:
1. Geostatic initial conditions
2. Geostatic step where the beam, contact pairs and full anchors are removed
3. Static step where the beam and contact pairs are added and the first 4m of soil are removed
4. Static step where the grouted part of the first anchor is added and two forces are introduced to one end of the grout and on a node on the beam to account for the prestress of the anchor
5. Static step where the free part of the anchor is added with a corresponding initial stress and the previous two forces changed to zero
Etc.
The problem occurs at step 4. The MSG file gives a negative eigenvalue error. With smaller anchor loads the analysis stops at a later step but the previous steps still have negative eigenvalues. I have ran previous analyses of retaining walls with no anchors and with contact pairs and analyses with anchors and no contact pairs with no problems.
If anyone could give me an opinion I would greatly appreciate it, thank you!
The steps are:
1. Geostatic initial conditions
2. Geostatic step where the beam, contact pairs and full anchors are removed
3. Static step where the beam and contact pairs are added and the first 4m of soil are removed
4. Static step where the grouted part of the first anchor is added and two forces are introduced to one end of the grout and on a node on the beam to account for the prestress of the anchor
5. Static step where the free part of the anchor is added with a corresponding initial stress and the previous two forces changed to zero
Etc.
The problem occurs at step 4. The MSG file gives a negative eigenvalue error. With smaller anchor loads the analysis stops at a later step but the previous steps still have negative eigenvalues. I have ran previous analyses of retaining walls with no anchors and with contact pairs and analyses with anchors and no contact pairs with no problems.
If anyone could give me an opinion I would greatly appreciate it, thank you!