I have some experience in modeling the thermoelectric sensor (converting deltaT to V) in COMSOL. Since I was curious about the opposite effect, I modeled it and observed the effect. However, I did not scrutinize it. Franklin Uriel Parás Hernández

Dear: Shayan Davani :

Thanks for your reply, I corrected a couple of things now,

I had some trivial errors, like; I did n't write the minus sign on the Seebeck Coefficient for the n-type material.

and It needed for the electric Ground boundary condition to be set up correctly

But now, when I try to solve the simulation it just shut down, before displaying any error or failure.. and COMSOL just shut down after a few seconds of running the solver ,

Do you think it could due to memory issues ?

Do you think I can send you my .mph file ? I would like to be sure first that there are no issues with the definition of my domains and boundary conditions ...

And , I don't know if you may know a way to set up the solver (or the mesh) to be able to get a rough solution of the model ...

I'll appreciate it very much !

Regards !

I need to check your model. Sure, you are welcome to send it. My email address is [email protected]

Dear Shayan :

Thank you for you reply,

I already solve it with the help of a computer's colleague. The problem was the memory, my computer did n't have enough memory to solve this model, even with a extremely coarse mesh.

But I got the solution.

Now I'm trying to work some calculations , but since It has n't been long since I started using COMSOL, I don't know well how to work with the Results toolbar/module.

I want to ask. Is it possible with the Results toolbar to get the TE's module ZT, Thermal Efficiency, Thermoelectric Output Power?

Or what would be the useful numerical results we can get after solving a simulation for a TE module?

Have you computed ZTs and Thermal Efficiencies of your modules/devices ? Is it easy to do it within COMSOL when you already have the simulation solved?

I appreciate it , Regards !

Great! You may need to define the parameters which are not built-in or provided by comsol. I don't think that Thermal Efficiency is built-in or defined in thermoelectric module. For those which are defined, they are accessible by defining the "Derived Values". For example : Results/Derived Values/Global Evaluation. Then evaluate your parameter of interest e.g. Power in the Expression/electric currents/energy and power/total electric energy.

Dear Shayan Davani :

Sorry for not replying earlier.

Thank you for your help, I haven't been able to compute any of the important results of the TE module from the solved simulation :(

As you said , the thermal efficiency is not an expression which is built up inside the COMSOL TE module as a function to do calculations ...

If I understand well, before computing eta, I need to compute the Thermoelectric Power delivered by the TEG and the Heat Flux at the Hot side of the module.

There is not even a function built up in the module to calculate the Thermoelectric Power or Electric Current, or I haven't found them yet.

The only function I found is the one to pull out the result for the Heat Flux, but I'm not sure if I am interpreting this result correctly,

The function I'am using is:

Component 1 -> Heat Transfer in Solids -> Global -> Net powers -> ht.ntfluxInt - Total net heat rate

(Please see the attached image)

As you may know there are many expression inside the category of "Net powers" and another categories similar to Net Powers, e.g. "Net powers, boundary features, downside", "Net powers, boundary features", "Net power, boundary features, upside" , etc.

Am I using the correct expression ?

If so, with this expression: ht.ntfluxInt - Total net heat rate, What Heat flux I am calculating exactly ?

If I am correct I have to compute the Heat Flux first, because I need the Heat Flux and the Electric Current to compute the thermal efficiency.

Am I right ?

Thank you for any help

Kind regards !

Dear Shayan Davani:

I think I have identified two problems may have been the reason I was n't able to compute the Thermal Efficiency of my TE Module.

1. I think one of the issues is that I'm not saving (or storing) the previous computation in the procedure of computing the Thermal Efficiency. So I believe that if you want to use a Global (or Point) Evaluation to compute some particular parameter, and this parameter is a function of some previous parameters you have already computed, I would suppose you have to store those previous results inside a function or variable.. so you can use these results to compute your final result.

I'm gonna try to explain myself better:

I want to compute eta_th as eta_th = W_gen/Q_hot. To compute eta_th I need to have previously computed W_get and Q_hot, to compute W_gen I need to have previously computed the Electric Current, to compute the I I need to have Pi(the Peltier Coefficient).

So to use each of the previously computed results do, I have to store them inside a variable or a function ? Ho do you manage this problem ?

2. The second issue may be that I'm mixing "Point Evaluation" with "Global Evaluation" functions to compute a result.

i.e. To compute the Electric Current (required for the W_gen) I need The Peltier Coefficient and the Heat Flux across the module ---> ( Pi = Q_dot/I ).

So, I did compute the Peltier Coefficient (Pi) using a "Point Evaluation" function and then I computed the Q_dot using a "Global Evaluation" function. And now I'm trying to compute the Electric Current (I) using a "Global Evaluation" function trying to use the two previous results: the Peltier Coefficient and the Net Heat Flux. It doesn't feel right ..

I feel think the second issue that I'm worrying about may be the main cause ...

Regards !