General Mechanical

General Mechanical

How to model wall contact on a exterior surface?

    • suraj9735
      Subscriber

      Hello!


      In an typical injection mold base we know there are top clamping plate, cavity plate, core plate, Side bar, Ejector plate, and bottom clamping


      etc.



      For my analysis I choose just two plate, core and cavity plate. To heat the both plate I put heater and give the "heat flow=250 W" in each plate in Thermal transient.


       




      Moving to more real condition I want to put contact resistance on outer surface of the core and cavity plate because these surfaces are in contact with another top clamping and side bar plate but these plates are not included in my model. Can anyone please guide me how to put "contact resistance between two wall on a single surface"?


      Note: I just have single exterior surface that is attached with another clamping plate but these are not included in my model.

    • Karthik R
      Administrator

      Hello,


      Here is a video tutorial which explains 2 ways of establishing contact between various surfaces. I hope this helps.



      Is there a reason why you do not want to include the rest of your system in your analysis?


      Thank you.


      Best Regards,


      Karthik

    • suraj9735
      Subscriber

      In the tutorial they have two different surface so they can easily couple them or put manual thermal conductance value. But in my case I just have one surface and want to set some wall contact resistance. It seems me I can only put convective heat transfer coefficient (h) with single surface. For that I need to predict some correct value of h.


       


      The rest plate of mould is not the area of interest. So I just excluded to make my model simple. If that part is must for correct modelling then I can include. But I also face same problem because top and bottom clamping plate will touch to the machine base. So what boundary condition should I put there?


      Thanks!

    • Karthik R
      Administrator

      Hello,


      You are correct, there is a way to model the heat transfer using an effective heat transfer coefficient. Using simple one-dimensional thermal resistance analysis, you can compute the effective heat transfer coefficient. The analysis will resemble something like this:


      1 / (h_cav-top, eff * A_cav-top) = ( l_top-clamp / ( k_top-clamp * A_top-clamp ) + ( 1 / ( h_top * A_top ) )


      Here (knowns):


      l_top-clamp = length of the top clamp


      k_top-clamp = thermal conductivity of top clamp


      A_top_clamp = cross sectional area of top clamp


      h_top = convective heat transfer coefficient at the top surface of top-clamp


      A_top = surface area at the top surface of top-clamp


      A_cav-top = surface area on the cavity plate on which you want to implement this heat transfer coefficient


      You can use the above equation to estimate the effective heat transfer coefficient. However, there are some big and important assumptions involved in this procedure.



      • The heat transfer is only one-dimension. That is, there is no heat transfer along the top clamp surface. Heat transfer is only across it and therefore, we end up using l/(kA) type of resistance analysis.

      • I am also assuming that the top clamp is not touching the machine base and am modeling it using convective heat transfer coefficient. I did this only to illustrate the method. If, in your case, the top clamp is touching the machine base, you might have to include a conductive thermal resistance (l/(kA)) type of analysis.


      During numerical modeling, it is extremely important to make your problem well-posed. What this means is you want to include all the necessary aspects of your problem to make sure that the boundary conditions you plan to apply to solve the problem are physical. It is sometimes not feasible to include everything in your analysis. In such cases, you make a note of the assumptions you are using in your model.


      If I were solving this problem, I would include the top clamp and use a heat transfer coefficient to model the heat loss at the top surface of this clamp. By including the top clamp in my analysis, I am ensuring a 3-dimensional heat transfer. It is likely that my solution might indicate that heat transfer is primarily across the plate, as opposed to along. This, however, becomes a conclusion rather than an assumption in my model.


      To estimate the heat transfer coefficient at the top surface of the top-clamp, you have to understand the physics at this surface. Is this going to be natural or forced convection would be the first question you need to ask yourself?


      I hope this helps and provides some perspective.


      Best Regards,


      Karthik

    • suraj9735
      Subscriber

      What an explanation!


      This gives me a very clear idea so Thank you very much for your splendid explanation.


      Now the heat transfer coefficient on the top clamping plate (contact resistance modeling) is an important aspect for me. If you have modeling reference to such problem you can share with me.


      On your serious note. I included the rest of the plate in my model. I believe that it will give more accurate result and also we can avoid one-dimensional heat transfer assumption in the clamping plate.



       


      now my next modeling is to control the temperature of the mold. I looking for PID thermal control modeling. Do the ANSYS have PID thermal control model?


      I want to find heat flux rate graph that can result in the desired output of temperature plot. Because I want the mold temperature to be constant at 180 deg but it exceeds or reduced unnecessary? Hope you understand my question!


      Thanks!

    • suraj9735
      Subscriber

      I make a tabular heat flux data to control the temperature



      But my maximum temperature exceeds 180 deg. So I want to model a PID control. But don't have an idea about "Can ANSYS do that?"

    • Karthik R
      Administrator

      Hello,


      Quick remarks about posting your question: If you have a new question, please create a new post. This way, your question will get the attention and visibility it deserves.


      About PID controller on ANSYS:


      I am sure PID controller can be implemented on ANSYS Fluent. You will need to use user defined functions to implement the control. As far as PID control implementation on transient thermal goes, I will let other experts on the community provide an answer.


      I hope this helps.


      Best Regards,


      Karthik

Viewing 6 reply threads
  • You must be logged in to reply to this topic.