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.

Time dependent ray heating simulation

Please login with a confirmed email address before reporting spam

I'm trying to create a light beam which would exist only between specified times. It looks like when the ray is launched it stays alive for ever. I tried to define piecewise time dependent function for the light power. I simulated the thermal effect (time dependent simulation) the ray (Ray Heat Source) creates on a surface it hits (or passes through like in case of a lens). It looks like the temperature on the surface continued to increase after the power (according the funtion) has dropped to zero. Also the Ray Tracing figure indicated the light emission still continues. Is it somehow possible to shut down the light beam at specified time.


7 Replies Last Post 30 gen 2018, 01:36 GMT-5
Nicholas Goldring Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 7 years ago 5 dic 2017, 22:48 GMT-5

Hi Mika,

If you're talking about having a periodic light source, you could release your rays at multiple times and then include a ray termination feature to stop them when a certain condition is met.

The option to include multiple release times is found under Advanced Settings in the Geometrical Optics node. Then check the settings on your release node to choose ray emission times. The ray termination feature is in the dropdown menu when you right click on the Geometrical Optics node.

I hope that helps.

-------------------
Nick
Hi Mika, If you're talking about having a periodic light source, you could release your rays at multiple times and then include a ray termination feature to stop them when a certain condition is met. The option to include multiple release times is found under Advanced Settings in the Geometrical Optics node. Then check the settings on your release node to choose ray emission times. The ray termination feature is in the dropdown menu when you right click on the Geometrical Optics node. I hope that helps.

Please login with a confirmed email address before reporting spam

Posted: 7 years ago 7 dic 2017, 10:56 GMT-5

Hello Nicholas,

thank you for answer. I managed to go one step forward. I defined the release times by giving the range. However, the rays seems to exist also outside the ranges. The heating effect (Heat Transfer) seem to continue outside the range. Also the ray trajectories indicates this.

My simulation case is following. I have defined three boundaries. Each of them is specified for a lighting source by "inlet". Each of them should emit light only during the range specified above. There is a lens which focuses the light from each source to the focal point where the heating effect is observed. So this my effort to model the case where a light source moves over the lens. by defining the stop condition as "no active rays demanding" stops the simulation to the first boundary case And I want the simulation to continue with the next boundary source.

Next I though, maybe I need to define an another ray by another Geometrical Optics physics and define the ray of the next boundary there. I also included another Ray Heat Source in Multiphysics and coupled this to the same Heat Transfer in Solids interface. Some errors followed in time dependent simulation. Maybe this is not a correct approach. An in my application the wavelength of each ray is constant. Should this be done with one Geometrical Optics physics interface.

How about the mesh ? I though by this way I manage with a fixed mesh. Or do I need a deformed mesh ?

Mika

Hello Nicholas, thank you for answer. I managed to go one step forward. I defined the release times by giving the range. However, the rays seems to exist also outside the ranges. The heating effect (Heat Transfer) seem to continue outside the range. Also the ray trajectories indicates this. My simulation case is following. I have defined three boundaries. Each of them is specified for a lighting source by "inlet". Each of them should emit light only during the range specified above. There is a lens which focuses the light from each source to the focal point where the heating effect is observed. So this my effort to model the case where a light source moves over the lens. by defining the stop condition as "no active rays demanding" stops the simulation to the first boundary case And I want the simulation to continue with the next boundary source. Next I though, maybe I need to define an another ray by another Geometrical Optics physics and define the ray of the next boundary there. I also included another Ray Heat Source in Multiphysics and coupled this to the same Heat Transfer in Solids interface. Some errors followed in time dependent simulation. Maybe this is not a correct approach. An in my application the wavelength of each ray is constant. Should this be done with one Geometrical Optics physics interface. How about the mesh ? I though by this way I manage with a fixed mesh. Or do I need a deformed mesh ? Mika

Nicholas Goldring Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 7 years ago 7 dic 2017, 14:49 GMT-5

Since I haven't seen your model, I'm not entirely sure how to configure your Ray Termination feature but it sounds like you put a stop condition on the solver instead of a Ray Termination feature under the Geometrical Optics node (See attached image). Also, I'm still not sure about your release time configuration for each inlet.

As long as you have Ray Heating which includes Thermal Stress(couples Heat Transfer in Solids with Solid Mechanics), you should not need to use a moving mesh.

I would generally avoid using duplicate physics interfaces with some possible exceptions. If you want to upload your model, I'll take a look at it and provide some more feedback.

-------------------
Nick
Since I haven't seen your model, I'm not entirely sure how to configure your Ray Termination feature but it sounds like you put a stop condition on the solver instead of a Ray Termination feature under the Geometrical Optics node (See attached image). Also, I'm still not sure about your release time configuration for each inlet. As long as you have Ray Heating which includes Thermal Stress(couples Heat Transfer in Solids with Solid Mechanics), you should not need to use a moving mesh. I would generally avoid using duplicate physics interfaces with some possible exceptions. If you want to upload your model, I'll take a look at it and provide some more feedback.


Please login with a confirmed email address before reporting spam

Posted: 7 years ago 7 dic 2017, 17:17 GMT-5

Hi,

thank you for you help. I uploaded the model. It's not a big one. It executes in 30 seconds. I must have missunderstod something in the release time configuration. As it looks like, temperatures still increase after 3 seconds, when the rays should not exist anymore.

Mika

Hi, thank you for you help. I uploaded the model. It's not a big one. It executes in 30 seconds. I must have missunderstod something in the release time configuration. As it looks like, temperatures still increase after 3 seconds, when the rays should not exist anymore. Mika


Nicholas Goldring Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 7 years ago 11 dic 2017, 09:22 GMT-5
Updated: 7 years ago 11 dic 2017, 09:25 GMT-5

Hi again Mika,

I haven't forgotten about your model although I'm currently working through some pressing issues within my own simulations. I will try to take a look at your light heating model today or at least by the end of the week. Unfortunately I don't have the heat transfer module but I can still access thermal stress through ray optics which sounds like it will be sufficient.

-------------------
Nick
Hi again Mika, I haven't forgotten about your model although I'm currently working through some pressing issues within my own simulations. I will try to take a look at your light heating model today or at least by the end of the week. Unfortunately I don't have the heat transfer module but I can still access thermal stress through ray optics which sounds like it will be sufficient.

Please login with a confirmed email address before reporting spam

Posted: 7 years ago 3 gen 2018, 04:27 GMT-5

Hi Nicholas,

thank you for the contribution in solving this problem. I contacted to Comsol support and finally the issue was solved. First of all I have misunderstand the presentation of rays. If there are no longer ray emission Multiphysics indicate that by removing the line start (a dot or something else), but still draws the lines indicating the rays already lauched. So that was ok. But the time dependent thermal simulation will always fail because the ray heating interface is not supporting a time dependent case. Although you can run a time dependent simulation as a bidirectional coupled ray tracing case, the solution is a stationary one showing temperatures at time is indefinite.

Mika

Hi Nicholas, thank you for the contribution in solving this problem. I contacted to Comsol support and finally the issue was solved. First of all I have misunderstand the presentation of rays. If there are no longer ray emission Multiphysics indicate that by removing the line start (a dot or something else), but still draws the lines indicating the rays already lauched. So that was ok. But the time dependent thermal simulation will always fail because the ray heating interface is not supporting a time dependent case. Although you can run a time dependent simulation as a bidirectional coupled ray tracing case, the solution is a stationary one showing temperatures at time is indefinite. Mika

Nicholas Goldring Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 7 years ago 30 gen 2018, 01:36 GMT-5

Hey Mika,

Thanks for the update! I'm glad to hear that your issue was resolved - the bidirectionally coupled ray tracing study seems to be an excellent choice. This study should run like so (with the addition of thermal-mechanical stress:

  1. Temperature and elastic stress fields are initially solved without any heat source generated by rays.
  2. Ray trajectories are calculated using the computed temperature from the previous step. Heat sources generated by incident rays upon an absorbing medium are also calculated.
  3. Temperature and elastic stress fields are solved again using the calculated heat source from the previous stfep.
  4. Steps 3 and 4 are repeated for a number of iterations until a self-consistent solution is obtained.

From what you said, COMSOL only stores a solution for the equilibrium teemperature. You may be able to save the temperature solutions at each iteration although it's something that I haven't really investigated myself. Let me know what you think.

-------------------
Nick
Hey Mika, Thanks for the update! I'm glad to hear that your issue was resolved - the bidirectionally coupled ray tracing study seems to be an excellent choice. This study should run like so (with the addition of thermal-mechanical stress: 1. Temperature and elastic stress fields are initially solved without any heat source generated by rays. 2. Ray trajectories are calculated using the computed temperature from the previous step. Heat sources generated by incident rays upon an absorbing medium are also calculated. 3. Temperature and elastic stress fields are solved again using the calculated heat source from the previous stfep. 4. Steps 3 and 4 are repeated for a number of iterations until a self-consistent solution is obtained. From what you said, COMSOL only stores a solution for the equilibrium teemperature. You may be able to save the temperature solutions at each iteration although it's something that I haven't really investigated myself. Let me know what you think.

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.