Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.
Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.
4 Point Bending of a Multi-Layer Bar with simultaneous phase transformations
Posted 2013年2月27日 GMT-5 09:34 Structural Mechanics Version 4.3a 2 Replies
Please login with a confirmed email address before reporting spam
Hello all,
I have been trying to simulate a 4-point bending test of a multi-layer bar with a geometrically simple 2-d model on COMSOL 4.3a.
For the better understanding of the case,
The bar consists of 3 horizontal layers, the nethermost one consists of a 5 mm thick Ni-base superalloy PWA 1484, the middle layer is an intermetallic compound NiAl with 80 micrometer thickness and the uppermost is Al2O3 with 5 micrometer thickness.
In order to simplify the 4 point bending test, I have adjusted two points on the top of the bar and assigned point loads in the negative y-direction, since the case is a bending under compressional load. 2 other points were also assigned for this stationary case at the netherrmost surface of the bar in order to fix the bar in the y-direction, in other words to contribute the deformation and bending of the bar. The model is quite simple, even for a beginner like me, but efficient indeed. As attached you might find this quite simple model, in order to understand the problem.
Now, I would like to take another step and couple this model with a time-dependent concentration change. Since my middle layer is a diffusional coating which would break down with time due to the Al depletion and instead of NiAl, in the middle layer Ni3Al would form with proceeding time. This case effects also the mechanical properties, therefore the 4-point bending mechanics, since the Elastic Modulus of my 2nd layer should be concentration and time dependent. Therefore the attached model should be somehow modified to compute a time-dependent and isothermal concentrational change and couple it with the initial 4 point bending under constant load.
More accurately, the concentration of the 2nd layer at the beginning should be pure 100 % NiAl, but as the time proceeds the NiAl should be completely consumed and the concentration should drop linearly (for simplification) to 0 % at the end, while the concentration of Ni3Al should start from 0% and increase to 100% linearly. Time dependence of the concentration and this phase transformation causes the elastic modulus of my second layer to be also time and concentration dependent.
Unfortunately, I couldnt achieve the so-called concentrational gradient with a time dependent study of transport of diluted species under the chemical species transport module.
Can someone suggest me a real beginner-friendly and simple approach to this problem? In reality the case is much more struggling since the growth rate of Ni3Al is not linear and affected by the deformation also. As a start, I thought that it would be much more simple for me to try a linear approach. Every idea for a better approach is welcome and appreciated.
With sincere regards,
Ceyhun
I have been trying to simulate a 4-point bending test of a multi-layer bar with a geometrically simple 2-d model on COMSOL 4.3a.
For the better understanding of the case,
The bar consists of 3 horizontal layers, the nethermost one consists of a 5 mm thick Ni-base superalloy PWA 1484, the middle layer is an intermetallic compound NiAl with 80 micrometer thickness and the uppermost is Al2O3 with 5 micrometer thickness.
In order to simplify the 4 point bending test, I have adjusted two points on the top of the bar and assigned point loads in the negative y-direction, since the case is a bending under compressional load. 2 other points were also assigned for this stationary case at the netherrmost surface of the bar in order to fix the bar in the y-direction, in other words to contribute the deformation and bending of the bar. The model is quite simple, even for a beginner like me, but efficient indeed. As attached you might find this quite simple model, in order to understand the problem.
Now, I would like to take another step and couple this model with a time-dependent concentration change. Since my middle layer is a diffusional coating which would break down with time due to the Al depletion and instead of NiAl, in the middle layer Ni3Al would form with proceeding time. This case effects also the mechanical properties, therefore the 4-point bending mechanics, since the Elastic Modulus of my 2nd layer should be concentration and time dependent. Therefore the attached model should be somehow modified to compute a time-dependent and isothermal concentrational change and couple it with the initial 4 point bending under constant load.
More accurately, the concentration of the 2nd layer at the beginning should be pure 100 % NiAl, but as the time proceeds the NiAl should be completely consumed and the concentration should drop linearly (for simplification) to 0 % at the end, while the concentration of Ni3Al should start from 0% and increase to 100% linearly. Time dependence of the concentration and this phase transformation causes the elastic modulus of my second layer to be also time and concentration dependent.
Unfortunately, I couldnt achieve the so-called concentrational gradient with a time dependent study of transport of diluted species under the chemical species transport module.
Can someone suggest me a real beginner-friendly and simple approach to this problem? In reality the case is much more struggling since the growth rate of Ni3Al is not linear and affected by the deformation also. As a start, I thought that it would be much more simple for me to try a linear approach. Every idea for a better approach is welcome and appreciated.
With sincere regards,
Ceyhun
Attachments:
2 Replies Last Post 2013年2月28日 GMT-5 03:33
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
first of all your units are in meters, I doubt that your Si wafer is that Thick and big ;)
I have several times suggested to COMSOL to add the default graph units on the graph scales, but so far it has not arrived, unfortunately I see dozen of these errors (and I still do them myself often)
It's strange as COMSOL has units almost everywhere now, that's really great and essential for validation, but NOT on the graphs and it's damn easy to implement that by software !!
But I still hope it will arrive once
This said, you should also avoid using "point loads" use rather boundaries, or small section (lines in your 2D case) to apply BC, the solution will be that much better
--
Good luck
Ivar
first of all your units are in meters, I doubt that your Si wafer is that Thick and big ;)
I have several times suggested to COMSOL to add the default graph units on the graph scales, but so far it has not arrived, unfortunately I see dozen of these errors (and I still do them myself often)
It's strange as COMSOL has units almost everywhere now, that's really great and essential for validation, but NOT on the graphs and it's damn easy to implement that by software !!
But I still hope it will arrive once
This said, you should also avoid using "point loads" use rather boundaries, or small section (lines in your 2D case) to apply BC, the solution will be that much better
--
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hello Ivar,
Thank you very much for your interest and answer. I will give it a shot with boundary load and lets see what happens.
Sincere Regards,
Ceyhun
Thank you very much for your interest and answer. I will give it a shot with boundary load and lets see what happens.
Sincere Regards,
Ceyhun