# IQN fails for quickly changing forces during wetting in a free-surface FSI problem

**URL:** <https://precice.discourse.group/t/iqn-fails-for-quickly-changing-forces-during-wetting-in-a-free-surface-fsi-problem/248>\
**Category:** Using preCICE\
**Tags:** inactive\
**Created:** [May 28, 2020, 4:36pm UTC](https://precice.discourse.group/t/iqn-fails-for-quickly-changing-forces-during-wetting-in-a-free-surface-fsi-problem/248 "2020-05-28T16:36:15Z")\
**Posts on this page:** 19\
**Page:** 1

<div class="post-metadata">

**Author:** ![uekerman](https://yyz2.discourse-cdn.com/free1/user_avatar/precice.discourse.group/uekerman/32/2199_2.png) [@uekerman](https://precice.discourse.group/u/uekerman)\
**Post date:** [May 28, 2020, 4:36pm UTC](https://precice.discourse.group/t/iqn-fails-for-quickly-changing-forces-during-wetting-in-a-free-surface-fsi-problem/248/1 "2020-05-28T16:36:15Z")

</div>

Continuing the discussion from [Error in writing watch-point for received mesh](https://precice.discourse.group/t/error-in-writing-watch-point-for-received-mesh/242/7):

 ![lastposition_before_divergence_0235s](https://global.discourse-cdn.com/free1/uploads/precice/original/1X/b222c4d91e6284cf4e603a0bcc1a797c91d6206f.jpeg)

> [@Error in writing watch-point for received mesh](https://precice.discourse.group/t/error-in-writing-watch-point-for-received-mesh/242/7):
>
> I have increased the accuracy in FASTEST to less than 1e-6 normalize error, and I used the parameters “50, and 10” that you suggested, but the simulation is not stable. Divergence occurs when water wets the structure.  
> Do you have any suggestions about these types of cases when the load changes abruptly?

> [@Error in writing watch-point for received mesh](https://precice.discourse.group/t/error-in-writing-watch-point-for-received-mesh/242/6):
>
> ```
> <max-used-iterations value="50"/>
> <time-windows-reused value="10"/>
> 
> ```

I am afraid there is not easy answer to this. Your setup is really quite challenging.

- When using much smaller values for allowed columns and reused time windows does it converge then?

```
<max-used-iterations value="15"/>
<time-windows-reused value="0"/>

```

- Using a `<coupling-scheme:parallel-implicit>` and then

```xml
<acceleration:IQN-ILS>
   <data name="Displacements" mesh="Structure_Nodes" />
   <data name="Forces" mesh="Structure_Nodes" />
   <preconditioner type="residual-sum"/>
	...
 </acceleration:IQN-ILS>

```

could maybe help

- So could be playing with the filter

```xml
<filter type="QR2" limit="1e-2"/>

```

Try “QR1” with “1e-5” as well for example

- Is there any particular reason why you use

```xml
<relative-convergence-measure limit="1e-4" data="Displacements" mesh="Structure_Nodes"/>
<relative-convergence-measure limit="1e-2" data="Forces" mesh="Structure_Nodes"/>

```

?  
Why not using the same limit for both?

- Or maybe `acceleration:IQN-IMVJ` and then

```xml
<max-used-iterations value="15"/>
<time-windows-reused value="0"/>

```

is worth a try here.

- We could also experiment with a restart of the QN acceleration or a tuning of the number of columns. But this is sth we have not yet implemented. Could be an interesting research question for @KyleDavisSA ?

---

<div class="post-metadata">

**Author:** ![KyleDavisSA](https://yyz2.discourse-cdn.com/free1/user_avatar/precice.discourse.group/kyledavissa/32/108_2.png) [@KyleDavisSA](https://precice.discourse.group/u/KyleDavisSA)\
**Post date:** [May 29, 2020, 7:54am UTC](https://precice.discourse.group/t/iqn-fails-for-quickly-changing-forces-during-wetting-in-a-free-surface-fsi-problem/248/2 "2020-05-29T07:54:34Z")

</div>

Regarding the value at top of the beam being very small, in this case the bottom of the beam seems to be pushed first, with the top following after, but the simulations ends before the top has a chance to move? That is something I think could be happening, but am not entirely sure.

Looking at the log files, FASTEST seems to diverge. I am not familiar with FASTEST, but in the log file it says:

> preparing geometry for Flow …  
> Check SCL (second order in time) …  
> ±------------------------------------------+  
> | Current time is: Fri May 22 15:32:47 2020 |  
> ±------------------------------------------+  
> ±------------------------------------------+  
> | Current time is: Fri May 22 15:32:49 2020 |  
> ±------------------------------------------+

when everything seems to converge, and:

> preparing geometry for Flow …  
> Check SCL (second order in time) …  
> ±------------------------------------------+  
> | Current time is: Fri May 22 15:32:49 2020 |  
> ±------------------------------------------+  
> \*\*\* !!!Iterations diverged!!! \*\*\*

for the final time step. Is FASTEST running second order in time? It could help stability to run with first order in time, and see if that works. But I would first try different filters and the `IQN-IMVJ` method Benjamin suggested. Also, for strongly coupled problems, sometimes the coupling stability gets better with increasing the time step, at the expense of stability in each solver. If that is not a problem with the solvers, you could try that.

---

<div class="post-metadata">

**Author:** ![JessicaMarino](https://avatars.discourse-cdn.com/v4/letter/j/3ab097/32.png) [@JessicaMarino](https://precice.discourse.group/u/JessicaMarino)\
**Post date:** [June 8, 2020, 10:03am UTC](https://precice.discourse.group/t/iqn-fails-for-quickly-changing-forces-during-wetting-in-a-free-surface-fsi-problem/248/3 "2020-06-08T10:03:20Z")

</div>

- Is there any particular reason why you use

```
<relative-convergence-measure limit="1e-4" data="Displacements" mesh="Structure_Nodes"/>
<relative-convergence-measure limit="1e-2" data="Forces" mesh="Structure_Nodes"/>

```

No really, but if I use more strict precision for the forces, the simulation takes a lot of time and the results are not better.

On the other hand, the parallel coupling is less stable than Serial coupling for this case. I was using IQN-ILS with “0” time-windows-reused and 25 max-used-iterations, making the simulation more stable but not enough to finish the required time (0.85). See the video

[https://drive.google.com/file/d/1GUvul8tab72HvaYIqBgiMpLdvkIIXzlR/view?usp=sharing](https://drive.google.com/file/d/1GUvul8tab72HvaYIqBgiMpLdvkIIXzlR/view?usp=sharing)

The conflict starts when the beam should return and begins to oscillate.

Concerning the IQN-IMVJ acceleration, I tested it but this appears less effective than IQN-ILS.

For this point with the acceleration methods. How can decide correctly the value for the filter, the number of reused iterations, and time reused?

I think this test case is very tricky if I change something in the coupling configuration I have the following problems:

1. If the time is bigger, Calculix converges very fast and the simulation is more stable but Fastest diverges because the multiphase model does not support very large time-steps.
2. If the time is smaller and appropriate for the multiphase fluid, Calulix diverges after a few coupling interactions.
3. If the grids are coarse in the two programs, I can simulate the test case but the results are not enough accuracy. Therefore, I am trying to simulate the case using finer grids. How can be a good proportion between the fluid and structure elements at the fluid-structure interface?
4. This test case is a 2D case (0.8mx0.8m fluid domain, 0.004mx0.09m rubber beam), I only have one volume on the z-axis (for FASTEST) and in the mapping setup z-dead=true. However, I do not know what is the appropriate width for the grid in the z-direction. On the one hand, if I use a very small value, for example, 0.002 or 0.006, the grid aspect ratio for the fluid domain is okay, but the RBF mapping diverges and to works requires at less 2 elements (C3D20R or C3D8i) are defined in the z-direction of the structure. On the other hand, if the width value is greater e.g. 0.2m, the RBF mapping works only with one element in the z-direction of the structure, but I notice in the vtk-files that the displacement is different for the parallel points, which is not logical if the case is 2D. My question is how can I decide the appropriate width for a quasi2D case? and how can I avoid these unequal displacements? To clarify my question, I attach the force and displacement values for 2 parallel points. The forces received are equal and the displacement unequal results in a complicate deformation of the fluid grid.

 ![problem with forces](https://global.discourse-cdn.com/free1/uploads/precice/original/1X/a9b04d2d718413a018cb1630235a6b9832ff1e3e.jpeg) ![dambreak_problem_grid](https://global.discourse-cdn.com/free1/uploads/precice/original/1X/3390292f1a0ecca516ff26a67c20286f15e357ed.jpeg)

1. Finally, in this case, the structure is a rubber beam (E=3.5MPa, nu=0.5), I am still using the elastic model (with nu=0.49) and the non-linear effects are considered (NLGEOM). Do you think that it is adequate? or the hyperelastic model should be used? However, for the hyperelastic model, I do not have the complete parameters that require these models. Maybe do you have more information about the constants for a rubber material?

---

<div class="post-metadata">

**Author:** ![uekerman](https://yyz2.discourse-cdn.com/free1/user_avatar/precice.discourse.group/uekerman/32/2199_2.png) [@uekerman](https://precice.discourse.group/u/uekerman)\
**Post date:** [June 10, 2020, 12:15pm UTC](https://precice.discourse.group/t/iqn-fails-for-quickly-changing-forces-during-wetting-in-a-free-surface-fsi-problem/248/4 "2020-06-10T12:15:53Z")

</div>

> [@JessicaMarino](#):
>
> See the video
> 
> [https://drive.google.com/file/d/1GUvul8tab72HvaYIqBgiMpLdvkIIXzlR/view?usp=sharing](https://drive.google.com/file/d/1GUvul8tab72HvaYIqBgiMpLdvkIIXzlR/view?usp=sharing)

The link is unfortunately private. I guess you have to adjust some sharing option in Google Drive.

> [@JessicaMarino](#):
>
> For this point with the acceleration methods. How can decide correctly the value for the filter, the number of reused iterations, and time reused?

There is no simple to answer to this. Some guidelines:

- The more past information you use the better the performance in general (as long as you don’t use too much, then the condition gets bad). But then the coupling scheme has also more troubles to react to sudden changes. An interesting future direction of research would be to make IQN more adaptive.
- The filter should work, meaning not filter out too little, but also not too much. You can check the number of filtered out columns per timestep in the “iterations” file of the second participant.

> [@JessicaMarino](#):
>
> - If the time is bigger, Calculix converges very fast and the simulation is more stable but Fastest diverges because the multiphase model does not support very large time-steps.
> - If the time is smaller and appropriate for the multiphase fluid, Calulix diverges after a few coupling interactions.

This sounds like an intersting use-case for the waveform relaxation we want to implement in preCICE, @BenjaminRodenberg’s PhD project. Preliminary results [here](https://onlinelibrary.wiley.com/doi/abs/10.1002/nme.6443?af=R), but might still take a year till we have this directly in preCICE. At the moment, there is no good solution for this.

> [@JessicaMarino](#):
>
> How can be a good proportion between the fluid and structure elements at the fluid-structure interface?

Well, hard question again. The black-box philosophy would say that both solvers should use what is best for them, so to balance the error between both. And then the data mapping has just to cope with those meshes. But, of course, the meshes also have an influence on the mapping error.  
Maybe some hope here: we are currently working on an artificial testing environment for data mappings. So that you can test mappings between your meshes without needing to run the complete coupled simulation. Because in the end, things like this simply need a lot of testing and comparison.

> [@JessicaMarino](#):
>
> This test case is a 2D case (0.8mx0.8m fluid domain, 0.004mx0.09m rubber beam), I only have one volume on the z-axis (for FASTEST) and in the mapping setup z-dead=true. However, I do not know what is the appropriate width for the grid in the z-direction. On the one hand, if I use a very small value, for example, 0.002 or 0.006, the grid aspect ratio for the fluid domain is okay, but the RBF mapping diverges and to works requires at less 2 elements (C3D20R or C3D8i) are defined in the z-direction of the structure.

This is odd. If you have `z-dead="true"` the RBF mapping should not see the width in z-direction at all. Could CalculiX have troubles with too small aspect ratios?

> [@JessicaMarino](#):
>
> My question is how can I decide the appropriate width for a quasi2D case? and how can I avoid these unequal displacements?

You do give force values in z direction. Then, CalculiX also computes a deformation in z direction and, thus, you get different displacements along the z axis. Could an easy solution be to set the z components of the forces to 0?

> [@JessicaMarino](#):
>
> Finally, in this case, the structure is a rubber beam (E=3.5MPa, nu=0.5), I am still using the elastic model (with nu=0.49) and the non-linear effects are considered (NLGEOM). Do you think that it is adequate? or the hyperelastic model should be used? However, for the hyperelastic model, I do not have the complete parameters that require these models. Maybe do you have more information about the constants for a rubber material?

That is, unfortunately, beyond my expertise. Maybe @KyleDavisSA knows more. Or you could ask on the CalculiX mailing list.

---

<div class="post-metadata">

**Author:** ![Alphaoo1](https://yyz2.discourse-cdn.com/free1/user_avatar/precice.discourse.group/alphaoo1/32/497_2.png) [@Alphaoo1](https://precice.discourse.group/u/Alphaoo1)\
**Post date:** [September 15, 2020, 7:29am UTC](https://precice.discourse.group/t/iqn-fails-for-quickly-changing-forces-during-wetting-in-a-free-surface-fsi-problem/248/5 "2020-09-15T07:29:42Z")

</div>

@uekerman Pointed out in the [post](https://precice.discourse.group/t/fsi-with-interfoam-and-calculix/277/10) that I might have a similar issue as discussed here.

@uekerman I tried out the suggestions you have in the first post in this thread.  
I will roughly mention the time steps at which my simulation ends.  
For the following implicit coupling scheme

```
 <coupling-scheme:serial-implicit>
            <time-window-size value="1.e-3" />
            <max-time value="0.5"/>
            <participants first="Fluid" second="Calculix"/>
            <exchange data="Forces0" mesh="Solid" from="Fluid" to="Calculix"/>
            <exchange data="Displacements0" mesh="Solid" from="Calculix" to="Fluid" />

            <max-iterations value="100"/>
            <absolute-convergence-measure limit="1e-6" data="Displacements0" mesh="Solid"/>
            <relative-convergence-measure limit="1e-2" data="Forces0" mesh="Solid"/>
            <extrapolation-order value="2"/>
 
            <acceleration:IQN-ILS>
                <data name="Displacements0" mesh="Solid"/>
                <preconditioner type="residual-sum"/>
                <filter type="QR1" limit="1e-6"/>
                <initial-relaxation value="0.01"/> 
                <max-used-iterations value="100"/>
                <time-windows-reused value="10"/> 
            </acceleration:IQN-ILS> 

    </coupling-scheme:serial-implicit>

```

For this “default” set-up, my simulation runs until `t=0.2`  
For `QR-2` Filter it runs until `t=0.0019`  
For `time-windows-reused value="0"` it runs for `t=0.003`  
For `<max-used-iterations value="15"/> <time-windows-reused value="0"/>` it runs for `t=0.0039`  
For `<max-used-iterations value="50"/> <time-windows-reused value="0"/>` it runs for `t=0.0010`  
For convergence, measure both set to `1e-6` it runs for `t=0.002`.(no particular reason to choose different, just using the same as what we have in the FSI flap tutorial).  
And lastly for the “default” scheme just changing from serial to parallel it runs for `t=0.0013’  
After this, I tried again the settings mentioned above, but it did not improve the results.

For serial explicit coupling scheme, initially, I had `<time-window-size value="1e-3"/>`, with which I was able to run until `t=0.28` (time of wetting).  
After increasing the deltaT for preCICE to `1e-4`, I am able to run the simulation until `t=0.52`, but as you can see the results are really bad.

 ![mesh_overlap_alphaW](https://global.discourse-cdn.com/free1/uploads/precice/original/1X/21bd320d693ca61a514fbbc457acc60eeaf3b0f5.jpeg)

 ![mesh_overlap](https://global.discourse-cdn.com/free1/uploads/precice/original/1X/b08ef0903c90bd52f300c310b496c23ea08a2571.jpeg)

 ![solid.5266](https://global.discourse-cdn.com/free1/uploads/precice/original/1X/42ab473a4035564737766b7d1422a57163a268d0.png)

Edit: Log files  
file:///home/prasad/01\_PhD/3\_Tutorials/3\_trials/0\_OpenFOAM-CalculiX\_interFoam/Fluid.log  
file:///home/prasad/01\_PhD/3\_Tutorials/3\_trials/0\_OpenFOAM-CalculiX\_interFoam/Solid.log

---

<div class="post-metadata">

**Author:** ![KyleDavisSA](https://yyz2.discourse-cdn.com/free1/user_avatar/precice.discourse.group/kyledavissa/32/108_2.png) [@KyleDavisSA](https://precice.discourse.group/u/KyleDavisSA)\
**Post date:** [September 15, 2020, 10:08am UTC](https://precice.discourse.group/t/iqn-fails-for-quickly-changing-forces-during-wetting-in-a-free-surface-fsi-problem/248/6 "2020-09-15T10:08:52Z")

</div>

Hi,

over what ranges did you test the filter limit for both QR1 and QR2? The `time-windows-reused` should stay above 10 for a case like this, as well as `max-used-iterations` above 50. I think the performance could be improved with the filtering.

---

<div class="post-metadata">

**Author:** ![Alphaoo1](https://yyz2.discourse-cdn.com/free1/user_avatar/precice.discourse.group/alphaoo1/32/497_2.png) [@Alphaoo1](https://precice.discourse.group/u/Alphaoo1)\
**Post date:** [September 15, 2020, 1:01pm UTC](https://precice.discourse.group/t/iqn-fails-for-quickly-changing-forces-during-wetting-in-a-free-surface-fsi-problem/248/7 "2020-09-15T13:01:15Z")

</div>

> [@KyleDavisSA](#):
>
> over what ranges did you test the filter limit for both QR1 and QR2? The `time-windows-reused` should stay above 10 for a case like this, as well as `max-used-iterations` above 50. I think the performance could be improved with the filtering.

For both the filters, initially, I was using

```
<max-used-iterations value="100"/>
<time-windows-reused value="10"/>

```

I did not play with the filter limit much outside the suggestions from the [Accelaration](https://github.com/precice/precice/wiki/Acceleration-Configuration#quasi-newton-schemes) wiki page.

---

<div class="post-metadata">

**Author:** ![KyleDavisSA](https://yyz2.discourse-cdn.com/free1/user_avatar/precice.discourse.group/kyledavissa/32/108_2.png) [@KyleDavisSA](https://precice.discourse.group/u/KyleDavisSA)\
**Post date:** [September 17, 2020, 2:57am UTC](https://precice.discourse.group/t/iqn-fails-for-quickly-changing-forces-during-wetting-in-a-free-surface-fsi-problem/248/8 "2020-09-17T02:57:12Z")

</div>

Hi,

would you be able to try running the problem with `parallel-implicit` coupling and include the `Forces0` data into the acceleration? I wonder if the lack of knowledge about the forces on the interface is causing the QN method to have problems. Perhaps start with:

```
<acceleration:IQN-ILS>
                <data name="Displacements0" mesh="Solid"/>
                <data name="Forces0" mesh="Solid"/>
                <preconditioner type="residual-sum"/>
                <filter type="QR2" limit="1e-2"/>
                <initial-relaxation value="0.1"/> 
                <max-used-iterations value="100"/>
                <time-windows-reused value="10"/> 
            </acceleration:IQN-ILS> 

```

It was mentioned above that parallel coupling is more unstable, however maybe this might help with the larger time-windows reused and a larger max-iterations

---

<div class="post-metadata">

**Author:** ![Alphaoo1](https://yyz2.discourse-cdn.com/free1/user_avatar/precice.discourse.group/alphaoo1/32/497_2.png) [@Alphaoo1](https://precice.discourse.group/u/Alphaoo1)\
**Post date:** [September 17, 2020, 6:40am UTC](https://precice.discourse.group/t/iqn-fails-for-quickly-changing-forces-during-wetting-in-a-free-surface-fsi-problem/248/9 "2020-09-17T06:40:59Z")

</div>

@KyleDavisSA I tried the setting, I think including `Forces0` in the acceleration surely helped.  
Before including the forces in the acceleration, the forces applied on the flap were in the magnitudes of \*e+03.  
Now they have come down to the magnitudes of \*e+02. I think I have to modify my OpenFOAM adapter further since this seems to be a case similar to the post by [Ulrich](https://precice.discourse.group/t/fsi-with-interfoam-and-calculix/277/5).

I have made the modifications implemented, but not sure.  
I will have a look at my adapter again.

For `<time-window-size value="1.e-3" />`, it run until almost `t=0.03`, which was similar to what I got before.  
For `<time-window-size value="1.e-4" />`, it run until `t=0.538698`, which is better than the explicit coupling with similar `<time-window-size`.

Unfortunately, the actual results are the same as before.

> [@Alphaoo1](#):
>
> ![mesh_overlap](https://global.discourse-cdn.com/free1/uploads/precice/original/1X/b08ef0903c90bd52f300c310b496c23ea08a2571.jpeg)
> 
> mesh\_overlap1509×804 257 KB

I am trying a case where the solid is a kind of a ramp, might have better results.

 ![interFoam_ramp_case](https://global.discourse-cdn.com/free1/uploads/precice/original/1X/2ccd41aff035683832f6d496c56d2ff1a07aeca8.png)

---

<div class="post-metadata">

**Author:** ![Alphaoo1](https://yyz2.discourse-cdn.com/free1/user_avatar/precice.discourse.group/alphaoo1/32/497_2.png) [@Alphaoo1](https://precice.discourse.group/u/Alphaoo1)\
**Post date:** [September 18, 2020, 6:44pm UTC](https://precice.discourse.group/t/iqn-fails-for-quickly-changing-forces-during-wetting-in-a-free-surface-fsi-problem/248/10 "2020-09-18T18:44:24Z")

</div>

Summary: With help of @KyleDavisSA, I was able to run the interFoam+CCX dam breaking simulation.  
Initially, the Young’s Modulus of the material was 10 times less than what it actually should be.  
I was also using the filter limit for `QR2` too high. `Forces` were included in the acceleration.

In the [interFoam case](https://github.com/Alphaoo1/tutorials/tree/master/FSI/flap_perp/OpenFOAM-CalculiX_interFoam), you can see the implicit coupling applied, this time including the `Forces` in the acceleration helped the simulation a lot.  
With decreasing the heavy fluid density to 100kg/m3 and increasing the Young’s modulus along with `IQN-ILS` acceleration, you can see the [results](https://github.com/Alphaoo1/tutorials/blob/master/FSI/flap_perp/OpenFOAM-CalculiX_interFoam/ani/interFoam-CCX_rho%3D100_t%3D5s_E%3D2.1MPa.mp4) for the simulation.

Next, I increased the density to match the density to that of water, keeping the rest of the settings the same.  
The simulation runs only until `t=1.33s`, but the [results](https://github.com/Alphaoo1/tutorials/blob/master/FSI/flap_perp/OpenFOAM-CalculiX_interFoam/ani/interFoam_E%3D2.1MPa_waterRho%3D1000_t%3D1.33.mp4) are sensible in terms of mapping.

The results do not differ much when using `IQN-IMVJ`

I think the simulation is failing on the OpenFoam side, due to excessive Courant number. (initially, it ran until `t=1.2`, with some modifications in `fvSchemes` and `fvOptions`, I was able to push it a bit farther.)

 ![](https://global.discourse-cdn.com/free1/uploads/precice/original/1X/19b73733d8b5e910f60a3522b1698ace52a21143.jpeg)

P.S: I could not upload the video here due to the large size file. But the links in the results have the video.  
The following are the 2 settings I used for which the simulation works.

```
<acceleration:IQN-ILS>
                <data name="Displacements0" mesh="Solid"/>
                <data name="Forces0" mesh="Solid"/>
                <preconditioner type="residual-sum"/>
                <filter type="QR2" limit="1e-2"/>
                <initial-relaxation value="0.1"/> 
                <max-used-iterations value="100"/>
                <time-windows-reused value="10"/> 
            </acceleration:IQN-ILS>

```

```
<acceleration:IQN-IMVJ>
                <imvj-restart-mode truncation-threshold="0.0001" chunk-size="8" reused-time-windows-at-restart="8" type="RS-SVD"/>
                <data name="Displacements0" mesh="Solid"/>
                <data name="Forces0" mesh="Solid"/>
                <preconditioner type="residual-sum"/>
                <filter type="QR2" limit="1e-2"/>
                <initial-relaxation value="0.1"/> 
                <max-used-iterations value="100"/>
                <time-windows-reused value="10"/> 
            </acceleration:IQN-IMVJ>

```

---

<div class="post-metadata">

**Author:** ![Alphaoo1](https://yyz2.discourse-cdn.com/free1/user_avatar/precice.discourse.group/alphaoo1/32/497_2.png) [@Alphaoo1](https://precice.discourse.group/u/Alphaoo1)\
**Post date:** [September 22, 2020, 8:55am UTC](https://precice.discourse.group/t/iqn-fails-for-quickly-changing-forces-during-wetting-in-a-free-surface-fsi-problem/248/11 "2020-09-22T08:55:22Z")

</div>

> The following are the 2 settings I used for which the simulation works.
> 
> ```
> <acceleration:IQN-ILS>
> <data name="Displacements0" mesh="Solid"/>
> <data name="Forces0" mesh="Solid"/>
> <preconditioner type="residual-sum"/>
> <filter type="QR2" limit="1e-2"/>
> <initial-relaxation value="0.1"/> 
> <max-used-iterations value="100"/>
> <time-windows-reused value="10"/> 
> </acceleration:IQN-ILS>
> 
> ```
> 
> ```
> <acceleration:IQN-IMVJ>
> <imvj-restart-mode truncation-threshold="0.0001" chunk-size="8" reused-time-windows-at-restart="8" type="RS-SVD"/>
> <data name="Displacements0" mesh="Solid"/>
> <data name="Forces0" mesh="Solid"/>
> <preconditioner type="residual-sum"/>
> <filter type="QR2" limit="1e-2"/>
> <initial-relaxation value="0.1"/> 
> <max-used-iterations value="100"/>
> <time-windows-reused value="10"/> 
> </acceleration:IQN-IMVJ>
> 
> ```

Is it possible to fine-tune this [case](https://github.com/Alphaoo1/tutorials/tree/master/FSI/flap_perp/OpenFOAM-CalculiX_interFoam) further, when the density of heavy fluid is increased to 1000kg/m3?

As for the density of 100kg/m3, the case runs just fine.

---

<div class="post-metadata">

**Author:** ![uekerman](https://yyz2.discourse-cdn.com/free1/user_avatar/precice.discourse.group/uekerman/32/2199_2.png) [@uekerman](https://precice.discourse.group/u/uekerman)\
**Post date:** [September 25, 2020, 8:33am UTC](https://precice.discourse.group/t/iqn-fails-for-quickly-changing-forces-during-wetting-in-a-free-surface-fsi-problem/248/12 "2020-09-25T08:33:40Z")

</div>

> [@Alphaoo1](#):
>
> Summary: With help of @KyleDavisSA, I was able to run the interFoam+CCX dam breaking simulation.  
> Initially, the Young’s Modulus of the material was 10 times less than what it actually should be.  
> I was also using the filter limit for `QR2` too high. `Forces` were included in the acceleration.

Great!

> [@Alphaoo1](#):
>
> Is it possible to fine-tune this [case](https://github.com/Alphaoo1/tutorials/tree/master/FSI/flap_perp/OpenFOAM-CalculiX_interFoam) further, when the density of heavy fluid is increased to 1000kg/m3?

You settings look already pretty good. Of course, you can always play around with them but I guess that there is no more significant improvement to expect.

The fundamental problem remains: due to the sudden wetting of the coupling interface, the history information that the quasi-Newton coupling uses becomes invalid from one timestep to another.  
@KyleDavisSA wanted to look into developing a kind of automatic restart mechanism of quasi-Newton, meaning to throw away all history if the coupling condition changes drastically. This could be a very nice future solution, but nothing for right now.

What you can always do is to manually restart you coupled simulation when the wetting happens.

---

<div class="post-metadata">

**Author:** ![uekerman](https://yyz2.discourse-cdn.com/free1/user_avatar/precice.discourse.group/uekerman/32/2199_2.png) [@uekerman](https://precice.discourse.group/u/uekerman)\
**Post date:** [September 25, 2020, 8:34am UTC](https://precice.discourse.group/t/iqn-fails-for-quickly-changing-forces-during-wetting-in-a-free-surface-fsi-problem/248/13 "2020-09-25T08:34:50Z")

</div>

> [@Alphaoo1](#):
>
> P.S: I could not upload the video here due to the large size file. But the links in the results have the video.

I guess that you can include YouTube videos here.

---

<div class="post-metadata">

**Author:** ![Sun](https://avatars.discourse-cdn.com/v4/letter/s/49beb7/32.png) [@Sun](https://precice.discourse.group/u/Sun)\
**Post date:** [June 22, 2021, 1:48pm UTC](https://precice.discourse.group/t/iqn-fails-for-quickly-changing-forces-during-wetting-in-a-free-surface-fsi-problem/248/14 "2021-06-22T13:48:25Z")

</div>

Dear Alphaoo1,  
I have the same issue as you mentioned. The simulation crashed when I set the density of water 1000kg/m3. But case can be run fine when the density of 100kg/m3. Did you solve this problem?  
 ![interfoam_calculix](https://global.discourse-cdn.com/free1/uploads/precice/original/1X/a902fd6de2db65b677f2f9360f64e2b8ba2e4aa5.png)

Best regard  
Sun

---

<div class="post-metadata">

**Author:** ![Alphaoo1](https://yyz2.discourse-cdn.com/free1/user_avatar/precice.discourse.group/alphaoo1/32/497_2.png) [@Alphaoo1](https://precice.discourse.group/u/Alphaoo1)\
**Post date:** [June 22, 2021, 2:59pm UTC](https://precice.discourse.group/t/iqn-fails-for-quickly-changing-forces-during-wetting-in-a-free-surface-fsi-problem/248/15 "2021-06-22T14:59:19Z")

</div>

Dear @Sun,

Yes, so I think the current issue is that the flap material is too weak.  
Check [post 10](https://precice.discourse.group/t/iqn-fails-for-quickly-changing-forces-during-wetting-in-a-free-surface-fsi-problem/248/10) in the current thread.

You can increase the Young’s Modulus in `/Case/Solid/flap.inp`

```
*MATERIAL, Name=EL
*ELASTIC

```

---

<div class="post-metadata">

**Author:** ![O\_K](https://yyz2.discourse-cdn.com/free1/user_avatar/precice.discourse.group/o_k/32/1340_2.png) [@O\_K](https://precice.discourse.group/u/O_K)\
**Post date:** [August 2, 2022, 5:11am UTC](https://precice.discourse.group/t/iqn-fails-for-quickly-changing-forces-during-wetting-in-a-free-surface-fsi-problem/248/16 "2022-08-02T05:11:18Z")

</div>

Hi @Alphaoo1,

I get your tutorials form git-hub and run it well. But when I set initial U into 0, it doesn’t run well. Do you know why and have any suggestion to solve this problem?

---

<div class="post-metadata">

**Author:** ![Alphaoo1](https://yyz2.discourse-cdn.com/free1/user_avatar/precice.discourse.group/alphaoo1/32/497_2.png) [@Alphaoo1](https://precice.discourse.group/u/Alphaoo1)\
**Post date:** [August 23, 2022, 8:10am UTC](https://precice.discourse.group/t/iqn-fails-for-quickly-changing-forces-during-wetting-in-a-free-surface-fsi-problem/248/17 "2022-08-23T08:10:54Z")

</div>

@O_K Sorry for the late reply.

Can you please elaborate on what you mean by “doesn’t run well”?

---

<div class="post-metadata">

**Author:** ![O\_K](https://yyz2.discourse-cdn.com/free1/user_avatar/precice.discourse.group/o_k/32/1340_2.png) [@O\_K](https://precice.discourse.group/u/O_K)\
**Post date:** [September 5, 2022, 4:52am UTC](https://precice.discourse.group/t/iqn-fails-for-quickly-changing-forces-during-wetting-in-a-free-surface-fsi-problem/248/18 "2022-09-05T04:52:26Z")

</div>

@Alphaoo1 Sorry for late reply.

It means that the calculation will be crashed down when it ran about 0.1s.

---

<div class="post-metadata">

**Author:** ![O\_K](https://yyz2.discourse-cdn.com/free1/user_avatar/precice.discourse.group/o_k/32/1340_2.png) [@O\_K](https://precice.discourse.group/u/O_K)\
**Post date:** [September 13, 2022, 5:48am UTC](https://precice.discourse.group/t/iqn-fails-for-quickly-changing-forces-during-wetting-in-a-free-surface-fsi-problem/248/19 "2022-09-13T05:48:21Z")

</div>

I’d like to know why the Young’s Modulus of the material needs to be 10 times larger than what it actually should be. Now, I was validating the FSI case about dambreak in the thesis of Vilmar Fuchs. The Young’s Modulus of material was 10^6 in the thesis, but the calculation crashed down when it ran about 0.15s. No matter how I adjust it, the calculation example can only calculate 0.15s. But the case ran about 0.36s when I change the Young’s Modulus to be 10^7. And there is the other problem, the Poisson’s ratio can’t be set as 0.5. If it is 0.5, Calculix will not run.