A problem in using Volumetric heat loss coefficient

[hexuanze]

Dear Gpyro users,

I encountered a potential problem when using Volumetric heat loss coefficient (hvl), caused by two ambient temperatures (one in BC, another in general) in Gpyro setting.

I confirm that regarding Gpyro 0.8 version, the ambient temperature value in the equation hvl(T-T∞) is from the general setting TAMB, where T∞ is a fixed value. If we have a boundary condition, in which the ambient temperature is transient, there will be an issue to use Volumetric heat loss coefficient (hvl). 

What we want to simulate in the heat generation term of hvl(T-T∞) is using a transient T∞, but Gpyro cannot achieve this because T∞ value in the equation hvl(T-T∞) is not from BC but the general.

This problem could bring trouble in simulating a transient ambient temperature boundary condition. Does anyone have experience on this? Or even better does anyone know the solution?

Thank you very much,

Best regards,

Xuanze

[frarich]

Dear Xuanze,

I have not encountered this problem before. I would try three approaches. One is that you simulate the geometry in more than one dimension and then explicitly define the boundary. One is to use an average heat loss over the assumed temperature range by using a higher h or Tambient. Another one would be try to use a kinetic reaction to get a volumetric heat loss.

The simplest would probably be to simulate the problem with the lowest Tambient and highest Tambient. That will give you a range of how big the variation actually is and if it is important to further investigate.

If you want it as a feature you could open a ticket on the github Gpyro page

Franz

[hexuanze]

Dear Franz, 

Thank you for your reply.

I believe this Volumetric heat loss coefficient (hvl) setting is not a problem for version 0.7, which is your preference. Because in version 0.7, the ambient temperature in General is relevant only if USE_BC_PATCHES = .FALSE. Therefore, it is not conflicted to the ambient temperature in BC. 

However, this function of "USE_BC_PATCHES = .FALSE." can not be used in version 0.8.

Thanks to your suggested solutions. Currently, I can simulate my experiments by adjusting the h in BC. But ignoring the Volumetric heat loss coefficient (hvl) in the governing equation seems like not very convincing. In my experiments, Volumetric heat loss coefficient (hvl) through side walls have a very small effect, but I would like to find a way to properly deal with it. 

Regarding your proposed solutions:

1. One is that you simulate the geometry in more than one dimension and then explicitly define the boundary. 

Reply: It is OK to use 2D, but there is another dimension still needs the Volumetric heat loss coefficient (hvl)

2. One is to use an average heat loss over the assumed temperature range by using a higher h or Tambient. 

Reply: adjusting h can solve the problem. But the heat transfer through other two coordinates are not considered. 

3. Another one would be try to use a kinetic reaction to get a volumetric heat loss.

Reply: I think it does not work. The heat generation due to Volumetric heat loss coefficient (hvl) is h(T-Ta). While the heat generation caused by a kinetic reaction is exponentially related to T. 

Thanks for discussion. 

Can I also ask if other users know how to solve this 0.8 version problem related to Volumetric heat loss coefficient?

Many thanks,

Regards,

Xuanze He