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.

Prestress Analysis, Frequency Domain

Please login with a confirmed email address before reporting spam

HI,
I am asking about if we can use variable load (function of time or frequency) instead the constant load in frequency domain (Perturbation), and how to do that. if we know that the constant load will be multiplied by cos(wt) to convert it to harmonic, what about if we have a load like 10*sin(w*t)?
Thanks in advance.

5 Replies Last Post Dec 23, 2014, 6:45 p.m. EST
Henrik Sönnerlind COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Feb 13, 2014, 2:59 a.m. EST
Hi,

If you want to apply a phase shift to a load, you can do it in two ways:

1. Right-click on the load and add a Phase subfeature

or

2. Write the load in complex form, e.g. as 10*i or 10/sqrt(2)*(1+i).

The expression for the load (both amplitude and phase) can be written as function of the frequency, e.g. 10*freq^2.

Regards,
Henrik
Hi, If you want to apply a phase shift to a load, you can do it in two ways: 1. Right-click on the load and add a Phase subfeature or 2. Write the load in complex form, e.g. as 10*i or 10/sqrt(2)*(1+i). The expression for the load (both amplitude and phase) can be written as function of the frequency, e.g. 10*freq^2. Regards, Henrik

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Feb 17, 2014, 7:23 a.m. EST
Hi Mr. Henrik,
I am so thankful for your reply,
1-Your answer helped me to solve a part of my problem. I want to ask you: I converted the function of the wave I used to frequency domain- using Fourier transform- and used it in the "prestress-frequency domain" study, Now how to isolate or disable the effect of "Perturbation" of the study from the function I'm using? I mean the effect of cos(wt)?
2- Why we cant see "prestress-time dependent Analysis" in comsol but we see prestress-frequency Domain?.
thank you Sir.
Hi Mr. Henrik, I am so thankful for your reply, 1-Your answer helped me to solve a part of my problem. I want to ask you: I converted the function of the wave I used to frequency domain- using Fourier transform- and used it in the "prestress-frequency domain" study, Now how to isolate or disable the effect of "Perturbation" of the study from the function I'm using? I mean the effect of cos(wt)? 2- Why we cant see "prestress-time dependent Analysis" in comsol but we see prestress-frequency Domain?. thank you Sir.

Henrik Sönnerlind COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Feb 18, 2014, 9:42 a.m. EST
Hi,

I am not sure that I fully understand your first question, so I will start with the second:

2. In time domain, there is no need for a special prestressed study since there is no assumption about linearity. You can either run an initial Stationary study and use the results as initial values for the Time dependent study, or simply apply the pre-stress load during the first time steps of the Time dependent study.

1. As for your first question, do you mean "How do I in postprocessing differentiate between prestress results and perturbation results"?

The answer to this question is: There is a control named "Expression evaluated for" in the settings for all results, such as graphs. The default for a perturbation study is "Harmonic perturbation', that is only the harmonic component is shown. There are six different possible choices, which you will find described in the documentation.

This control is only shown if the data set contains a perturbation solution.

Regards,
Henrik
Hi, I am not sure that I fully understand your first question, so I will start with the second: 2. In time domain, there is no need for a special prestressed study since there is no assumption about linearity. You can either run an initial Stationary study and use the results as initial values for the Time dependent study, or simply apply the pre-stress load during the first time steps of the Time dependent study. 1. As for your first question, do you mean "How do I in postprocessing differentiate between prestress results and perturbation results"? The answer to this question is: There is a control named "Expression evaluated for" in the settings for all results, such as graphs. The default for a perturbation study is "Harmonic perturbation', that is only the harmonic component is shown. There are six different possible choices, which you will find described in the documentation. This control is only shown if the data set contains a perturbation solution. Regards, Henrik

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Feb 19, 2014, 3:13 a.m. EST
Hi,
I really don't know how to express my deep thanks for your guidance, your answer really helpful,
Sir, while you told me that I can get pre-stress with time dependent, so let me focus on it and give up the frequency domain for time being.
I actually did both method you recommend me to get pre-stress: by making initial stress (of the stationary step) to put it in the time dependent step and the second way by adding the static load as first time step, but both of these ways give me what called " stress wave" which is undesirable for my simulation because I am simulating an ultrasonic wave in the Time-Dependent study. For this reason I tried last time to get use of Pre-stress-Frequency Domain. I am so grateful for your answer which were important for me .
Best Regards
Nadom
Hi, I really don't know how to express my deep thanks for your guidance, your answer really helpful, Sir, while you told me that I can get pre-stress with time dependent, so let me focus on it and give up the frequency domain for time being. I actually did both method you recommend me to get pre-stress: by making initial stress (of the stationary step) to put it in the time dependent step and the second way by adding the static load as first time step, but both of these ways give me what called " stress wave" which is undesirable for my simulation because I am simulating an ultrasonic wave in the Time-Dependent study. For this reason I tried last time to get use of Pre-stress-Frequency Domain. I am so grateful for your answer which were important for me . Best Regards Nadom

Please login with a confirmed email address before reporting spam

Posted: 10 years ago Dec 23, 2014, 6:45 p.m. EST
Hi Henrik

If you could, please , spare your time to answer this query, I shall be grateful

I want to simulate the behaviour of the piezoelectric beam which has residual stress. I input the residual stress as initial stress in the Solid Mechanics.

The resonant frequency does change when the analysis is done in Eigen Frequency and Pre-stressed Eigen Frequency.

The same does not reflect in pre-stressed Frequency domain analysis.

I am giving a constant load at the free end and frequency sweep for prestressed frequency study has the resonant frequency obtained from prestressed eigen frequency

Thanks and Regards
Anurekha

Hi,

I am not sure that I fully understand your first question, so I will start with the second:

2. In time domain, there is no need for a special prestressed study since there is no assumption about linearity. You can either run an initial Stationary study and use the results as initial values for the Time dependent study, or simply apply the pre-stress load during the first time steps of the Time dependent study.

1. As for your first question, do you mean "How do I in postprocessing differentiate between prestress results and perturbation results"?

The answer to this question is: There is a control named "Expression evaluated for" in the settings for all results, such as graphs. The default for a perturbation study is "Harmonic perturbation', that is only the harmonic component is shown. There are six different possible choices, which you will find described in the documentation.

This control is only shown if the data set contains a perturbation solution.

Regards,
Henrik


Hi Henrik If you could, please , spare your time to answer this query, I shall be grateful I want to simulate the behaviour of the piezoelectric beam which has residual stress. I input the residual stress as initial stress in the Solid Mechanics. The resonant frequency does change when the analysis is done in Eigen Frequency and Pre-stressed Eigen Frequency. The same does not reflect in pre-stressed Frequency domain analysis. I am giving a constant load at the free end and frequency sweep for prestressed frequency study has the resonant frequency obtained from prestressed eigen frequency Thanks and Regards Anurekha [QUOTE] Hi, I am not sure that I fully understand your first question, so I will start with the second: 2. In time domain, there is no need for a special prestressed study since there is no assumption about linearity. You can either run an initial Stationary study and use the results as initial values for the Time dependent study, or simply apply the pre-stress load during the first time steps of the Time dependent study. 1. As for your first question, do you mean "How do I in postprocessing differentiate between prestress results and perturbation results"? The answer to this question is: There is a control named "Expression evaluated for" in the settings for all results, such as graphs. The default for a perturbation study is "Harmonic perturbation', that is only the harmonic component is shown. There are six different possible choices, which you will find described in the documentation. This control is only shown if the data set contains a perturbation solution. Regards, Henrik [/QUOTE]

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.