mbqueens
Civil/Environmental
- Feb 24, 2011
- 10
I’ll apologize in advance for this long winded post, but I also want to thank anybody that’s willing to share their knowledge and give me some advice.
I’m looking for the best way to model the interaction between a cured-in-place liner and its cast iron host pipe. The cast iron pipe is approx 10mm thick has an inner diameter of approx 155mm. The liner is approx 5mm thick and is blown up to the inside of the cast iron pipe (providing a tight fit). A resin is used in this process which helps form the liner and provides a weak bond to the inner surface of the cast iron pipe.
For my particular problem, I’m assuming that the cast iron pipe has cracked circumferentially (completely through its thickness). I want to apply a moment to the damaged cast iron section and obtain stresses in the liner due to this moment. It’s also worthy to note that this is a water main and therefore the system is under internal pressure. Here’s what I’ve done so far:
***Please see the attached image for a better understanding of the description below***
It’s a 3D analysis. I’m using C3D10M tets (which after reading this forum might be a big mistake) for both the cast iron and the liner. It’s a 3 step analysis. First step: apply boundary conditions – symmetry bc’s on the liner at the circumferential break (left side of the image) as well as on the top and bottom face along the longitudinal direction. I’ve applied a rigid constraint to the cast iron pipe and have completely fixed it at the reference point (the X in the image). Second step: Apply internal pressure to the inner surface of the liner. Third step: apply a moment to the cast iron pipe about the reference point (after removing the UR1 bc at the reference point), while continuing to apply the internal pressure.
Here are some issues:
- The compressive stresses that occur in the liner at the invert of the cast iron pipe (at the circumferential break) increase almost linearly with increasing mesh refinement. I believe this is due to the sharp edge of the cast iron “cutting” into the liner at this point. I’m looking for a way to explain this clearly.
- I’d like to model the “weak bond” between the liner and the cast iron pipe using cohesive surfaces (I’m using version 6.9) but I’m having difficulty choosing the Knn, Kss, Ktt values and I’m not exactly sure which value to use for total displacement in the damage evolution. I’ve performed push out tests on the liner and I know the force required to break the bond but I don’t know how to translate that into the traction-separation definition. (Once the bond breaks I’m using a friction model to define the behavior between the liner and the cast iron.)
I’m just wondering if I’m even going about this the right way. Is there a better way to be performing this type of analysis? Should I be using linear hex elements like many of the posts in this forum mention? I’ve basically been learning FEM on my own. I’ve read the documentation and feel that I have a good grasp of the program itself but I keep second guessing myself when it comes to an accurate and efficient model.
Help is greatly appreciated.
Mike
I’m looking for the best way to model the interaction between a cured-in-place liner and its cast iron host pipe. The cast iron pipe is approx 10mm thick has an inner diameter of approx 155mm. The liner is approx 5mm thick and is blown up to the inside of the cast iron pipe (providing a tight fit). A resin is used in this process which helps form the liner and provides a weak bond to the inner surface of the cast iron pipe.
For my particular problem, I’m assuming that the cast iron pipe has cracked circumferentially (completely through its thickness). I want to apply a moment to the damaged cast iron section and obtain stresses in the liner due to this moment. It’s also worthy to note that this is a water main and therefore the system is under internal pressure. Here’s what I’ve done so far:
***Please see the attached image for a better understanding of the description below***
It’s a 3D analysis. I’m using C3D10M tets (which after reading this forum might be a big mistake) for both the cast iron and the liner. It’s a 3 step analysis. First step: apply boundary conditions – symmetry bc’s on the liner at the circumferential break (left side of the image) as well as on the top and bottom face along the longitudinal direction. I’ve applied a rigid constraint to the cast iron pipe and have completely fixed it at the reference point (the X in the image). Second step: Apply internal pressure to the inner surface of the liner. Third step: apply a moment to the cast iron pipe about the reference point (after removing the UR1 bc at the reference point), while continuing to apply the internal pressure.
Here are some issues:
- The compressive stresses that occur in the liner at the invert of the cast iron pipe (at the circumferential break) increase almost linearly with increasing mesh refinement. I believe this is due to the sharp edge of the cast iron “cutting” into the liner at this point. I’m looking for a way to explain this clearly.
- I’d like to model the “weak bond” between the liner and the cast iron pipe using cohesive surfaces (I’m using version 6.9) but I’m having difficulty choosing the Knn, Kss, Ktt values and I’m not exactly sure which value to use for total displacement in the damage evolution. I’ve performed push out tests on the liner and I know the force required to break the bond but I don’t know how to translate that into the traction-separation definition. (Once the bond breaks I’m using a friction model to define the behavior between the liner and the cast iron.)
I’m just wondering if I’m even going about this the right way. Is there a better way to be performing this type of analysis? Should I be using linear hex elements like many of the posts in this forum mention? I’ve basically been learning FEM on my own. I’ve read the documentation and feel that I have a good grasp of the program itself but I keep second guessing myself when it comes to an accurate and efficient model.
Help is greatly appreciated.
Mike