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.
heat transfer with a solenoid
Posted 8 déc. 2015, 12:42 UTC−5 Low-Frequency Electromagnetics, Heat Transfer & Phase Change Version 5.1 5 Replies
Please login with a confirmed email address before reporting spam
Hi guys,
I'm new into COMSOL so these questions might be quite silly. I was trying to simulate the inductive heating to a stainless steel washer using a solenoid. In my model I have an array of circles to simulate the coil and there is a stainless steel washer in the middle of the model. The results I'm expecting is the temperature raise of the washer with various driving powers. Here are some of my questions:
1. The washer is placed in a petri dish and there's a little water in that dish too to cool the washer, how should I simulate the heat convection from the washer surface to the water? I've tried adding a heat transfer in fluid physics but the results didn't match with our experimental results. Basically I cannot see a cooling happening during the simulation.
2. If I remove the water, say the washer itself is only cooling by the radiation to the surrounding air, should I add a diffusion surface and set the Tamb to be the room temperature? This is how I did it before but still, no significant cooling is observed.
3. This question is more related to the inductive heating itself. We are using 850KHz which should be able to observe the skin effect, I'm not sure whether COMSOL has already taken this into account or there's something else I need to do to calculate the skin effect. If I do, could anyone provide some suggestions on how to accomplish this?
I've attached my simple model without any solutions or so, it's running normally but there might still be some physics mistakes hidden in the model. I'll be really appreciate if anyone can help me check it out and provide some suggestions on my simulation.
Big thanks in advance! :-)
Chenchen
I'm new into COMSOL so these questions might be quite silly. I was trying to simulate the inductive heating to a stainless steel washer using a solenoid. In my model I have an array of circles to simulate the coil and there is a stainless steel washer in the middle of the model. The results I'm expecting is the temperature raise of the washer with various driving powers. Here are some of my questions:
1. The washer is placed in a petri dish and there's a little water in that dish too to cool the washer, how should I simulate the heat convection from the washer surface to the water? I've tried adding a heat transfer in fluid physics but the results didn't match with our experimental results. Basically I cannot see a cooling happening during the simulation.
2. If I remove the water, say the washer itself is only cooling by the radiation to the surrounding air, should I add a diffusion surface and set the Tamb to be the room temperature? This is how I did it before but still, no significant cooling is observed.
3. This question is more related to the inductive heating itself. We are using 850KHz which should be able to observe the skin effect, I'm not sure whether COMSOL has already taken this into account or there's something else I need to do to calculate the skin effect. If I do, could anyone provide some suggestions on how to accomplish this?
I've attached my simple model without any solutions or so, it's running normally but there might still be some physics mistakes hidden in the model. I'll be really appreciate if anyone can help me check it out and provide some suggestions on my simulation.
Big thanks in advance! :-)
Chenchen
Attachments:
5 Replies Last Post 11 déc. 2015, 10:17 UTC−5
Please login with a confirmed email address before reporting spam
Posted:
9 years ago
Hi,
No need to compute heat transfert in the whole domain. Use heat transfert coefficient.
You can add radiations between surfaces, a cooling for the coil is required.
You also have to calculate natural convection in water, there are examples everywhere.
I include a modified file
Best regards
Julien
No need to compute heat transfert in the whole domain. Use heat transfert coefficient.
You can add radiations between surfaces, a cooling for the coil is required.
You also have to calculate natural convection in water, there are examples everywhere.
I include a modified file
Best regards
Julien
Attachments:
Please login with a confirmed email address before reporting spam
Posted:
9 years ago
Hi Julien,
Thanks for your reply and the modified model, that did solve part of my problems!
My current version of COMSOL doesn't support some functions in the Heat Transfer physics you added so I created my own one based on your modification. Basically I cannot do the opaque nodes as you did. However, this version works for me now.
I noticed that you added the Impedance Boundary Condition in the Magnetic Fields physics, I assume this is to simulate the skin effect? When I tried to add this in my original model, I can't select the right boundary (shown as not applicable), do you have any idea why? Same thing happened when I tried to set the Heat Flux for the washer boundary, it was shown as "not applicable", not sure how to fix this problem.
Anyway, thanks for your help! It did provide a lot of useful information:)
Chenchen
Thanks for your reply and the modified model, that did solve part of my problems!
My current version of COMSOL doesn't support some functions in the Heat Transfer physics you added so I created my own one based on your modification. Basically I cannot do the opaque nodes as you did. However, this version works for me now.
I noticed that you added the Impedance Boundary Condition in the Magnetic Fields physics, I assume this is to simulate the skin effect? When I tried to add this in my original model, I can't select the right boundary (shown as not applicable), do you have any idea why? Same thing happened when I tried to set the Heat Flux for the washer boundary, it was shown as "not applicable", not sure how to fix this problem.
Anyway, thanks for your help! It did provide a lot of useful information:)
Chenchen
Please login with a confirmed email address before reporting spam
Posted:
9 years ago
Hi Chenchen,
I think that you doesn't need the surface to surface radiation condition if temperatures are low.
I noticed that you added the Impedance Boundary Condition in the Magnetic Fields physics, I assume this is to simulate the skin effect?
Yes, there is a strong skin effect. You can also make a very fine mesh close to the surface (bounadry layers) and apply a volumic heat source on the conductive domain. It's easier the boundary heat source.
When I tried to add this in my original model, I can't select the right boundary (shown as not applicable), do you have any idea why?
This BC is only applicable on external boundaries so you need to deselect the conductive domain.
Same thing happened when I tried to set the Heat Flux for the washer boundary, it was shown as "not applicable", not sure how to fix this problem.
Same thing for heat flux BC.
Don't forget the natural convection in the small water tank
Best regards
Julien
I think that you doesn't need the surface to surface radiation condition if temperatures are low.
I noticed that you added the Impedance Boundary Condition in the Magnetic Fields physics, I assume this is to simulate the skin effect?
Yes, there is a strong skin effect. You can also make a very fine mesh close to the surface (bounadry layers) and apply a volumic heat source on the conductive domain. It's easier the boundary heat source.
When I tried to add this in my original model, I can't select the right boundary (shown as not applicable), do you have any idea why?
This BC is only applicable on external boundaries so you need to deselect the conductive domain.
Same thing happened when I tried to set the Heat Flux for the washer boundary, it was shown as "not applicable", not sure how to fix this problem.
Same thing for heat flux BC.
Don't forget the natural convection in the small water tank
Best regards
Julien
Please login with a confirmed email address before reporting spam
Posted:
9 years ago
Hi Julien,
Thanks for the reply! That helped a lot!
Speaking of the natural convection in water, I found some examples from COMSOL, seems like there should be a "Heat transfer coefficient" list to select from either natural convection or force convection, but in both our models there's no such a selection... so my question is, in this case, how do you calculate the h to be 5W/m2?
Thanks!! You are such a nice person to help me fix and even improve this model a lot!
Chenchen
Thanks for the reply! That helped a lot!
Speaking of the natural convection in water, I found some examples from COMSOL, seems like there should be a "Heat transfer coefficient" list to select from either natural convection or force convection, but in both our models there's no such a selection... so my question is, in this case, how do you calculate the h to be 5W/m2?
Thanks!! You are such a nice person to help me fix and even improve this model a lot!
Chenchen
Please login with a confirmed email address before reporting spam
Posted:
9 years ago
You're welcome
h=5 W/(m2.K) is a current value. Accurate calculations can be done the Nusselt number in fonction of the geometry of the surfaces. Lot of examples in litterature
Natural convection in the water domain must be calculated with Navier Stockes + heat transfert in fluid.
You can also find examples.
Best regards
Julien
h=5 W/(m2.K) is a current value. Accurate calculations can be done the Nusselt number in fonction of the geometry of the surfaces. Lot of examples in litterature
Natural convection in the water domain must be calculated with Navier Stockes + heat transfert in fluid.
You can also find examples.
Best regards
Julien