Damping problem

Numerical methods and mathematical models of Elmer
Post Reply
Sayan
Posts: 112
Joined: 06 Dec 2020, 12:44
Antispam: Yes

Damping problem

Post by Sayan »

Hiiii,
My file is not considering damping though I have defined rayleigh damping in my file...Is there anything need to modify? The original result is also attached as jpg
Attachments
IMG-20210324-WA0005.jpg
(106.03 KiB) Not downloaded yet
case.sif
(2.43 KiB) Downloaded 163 times
undamaged.msh
(30.51 KiB) Downloaded 163 times
kevinarden
Posts: 2312
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Damping problem

Post by kevinarden »

I did some testing and it appears Rayleigh does not work with
Element = "P:2"
Sayan
Posts: 112
Joined: 06 Dec 2020, 12:44
Antispam: Yes

Re: Damping problem

Post by Sayan »

Ookk...I checked after eliminating that line...now the result is quickly coming to periodic motion after that no damping effect is there .also from.result I checked it's not showing non-linear effect..whether damping and elastic solver together doesn't work ?
Attachments
IMG-20210324-WA0011.jpg
IMG-20210324-WA0011.jpg (12.4 KiB) Viewed 2545 times
kevinarden
Posts: 2312
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Damping problem

Post by kevinarden »

Still doing some further testing using an earlier damping verification model I have.
mika
Posts: 236
Joined: 15 Sep 2009, 07:44

Re: Damping problem

Post by mika »

Sayan wrote: 24 Mar 2021, 14:10 My file is not considering damping though I have defined rayleigh damping
The sif file has

Rayleigh Beta=Real 0.0001

ElasticSolver however tries to read "Rayleigh Damping Alpha" or "Rayleigh Damping Beta", so try with

Rayleigh Damping Beta = Real ...

-- Mika
kevinarden
Posts: 2312
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Damping problem

Post by kevinarden »

Mika is correct. I noticed when I went back to my test case it was working. I missed that the Rayleigh Damping Beta = Real was not entered correctly. The difference with damped/undamped can be seen in the peaks of the response
Attachments
reults.png
(106.74 KiB) Not downloaded yet
Sayan
Posts: 112
Joined: 06 Dec 2020, 12:44
Antispam: Yes

Re: Damping problem

Post by Sayan »

Hiiii,
Thank you for pointing that out...But even after modifying same result I am getting as I got after eliminating the line "Element=p:2" and posted here as jpg file... Initially damping is there upto 2s approx but after that there is no effect..it's dying very fast though damping is very low in magnitude...i have attached modified sif file also...To check non-linearity displacement vs velocity plot is required..this plot will not be a smooth ellipse if it's non-linear, for linear case it will be ellipse..which I checked after 3s it's becoming either smooth ellipse or circle..
Attachments
case.sif
(2.31 KiB) Downloaded 154 times
kevinarden
Posts: 2312
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Damping problem

Post by kevinarden »

The no P2 is not a solution, the mesh is not fine enough to solve the problem without using higher order elements. use of
Element = "P:2" is a must for accuracy, or a much more refined mesh.
Attachments
results.png
(112.77 KiB) Not downloaded yet
Sayan
Posts: 112
Joined: 06 Dec 2020, 12:44
Antispam: Yes

Re: Damping problem

Post by Sayan »

In my case the problem.stilll exists, it's not showing damping effect..Was there any additional changes except element p:2 and Rayleigh Damping Beta line?It's matching completely when I compared both with and without damping...
Attachments
case1.sif
(2.26 KiB) Downloaded 150 times
case.sif
(2.26 KiB) Downloaded 145 times
kevinarden
Posts: 2312
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Damping problem

Post by kevinarden »

Neither sif has any damping defined in the material.
Post Reply