Discussion Forum

Hi

From my understanding, when you fetch a library material data set it's loaded when you do the operation in the BC GUI, you see he values entered in BOLD, but then you can always change or adapt some of them, or complement others not defined. set up a funciton for your variable and enter this after you have loaded the material, or what I often do, in my "private" material library, I make a copy that I adapt to my needs.

Good luck
Ivar

Hi - I apologize for describing the problem so imprecise.

I added a new material using Library->Materials/Coefficient Library and entered values for e.g. electrical resistivity and conductivity, but I did not find a field to enter the Temperature coefficient of the resistivity (a or alpha).

In the materials library I could define a function, but how do I find which name I need to give that variable that represents 'a'? I guess it should be something like emdc_alpha but I did not find a description of the variable names in the documentation.

Thanks a lot,
Kurt

Hi

Well the way I understand you, is that you want to do the same as most of the basic materials have been defined, with " material constants" defined as functions depending on the temperature "T". Take a look at ie.e water or air in the basic material library and chekc the function and plot tab.

In the Material Library, "New" selection define a "function" (lower right above the "help" button) as an "analytical", or more probably as a Table. Then, once the new function has been defined, enter the name of the function for the library variable list for the one you want to define. the Func(T[1/K]) unit is to ensure that the variable T you use is translated to Kelvin and made unitless for the function entry.

Hope this helps
Good luck
Ivar

Hi - nice trick with Func(T[K]).

Here is an answer to my question in case somebody else is looking for it some day:

The variable name, respectively names, I was looking for were
<res0>1.06e-7[ohm*m]</res0>
<alphares>0.0082[1/K]</alphares>
<T0>295[K]</T0>

These parameters were not present in the Materials/Coefficient Library after creation of a new Library. Therefore, if the above parameters are manually added to the corresponding .dat file, the parameters become accessible in the Materials/Coefficient Library.
There is probably a nicer way to do that, but at least it works. An example of such definitions can be found, e.g., when instpecting copper in the Basic Material Properties.

Hello,

my materials are set.. it is Water(liquid) which needs to be heated by rf.

When having the bolded variables functions of the library, my model would not compute.

If I set fixed scalars instead (Cp,Rho and k = 3 or else) then the model computes..?

what is wrong?
the plots of the Water Cp functions are there, but it does seem to apply to my model?

what I am doing wrong?

Thanks

Lukas

Hi

are you sure that the bolded ones do not refer to the variables "T" and /or "p" the temperature and pressure, and that you have defined these ? At least as "Constants", probably you would like to have T defined in your equations elsewhere.

This is easy to forget when setting up models with non-linear materials

have fun Comsoling
Ivar

Thanks for your answer,

but defining a T in my options/constants at 300°K does not help.
It says then: duplicate variable name..

for p it does matter since it is not involved.

it seems to me that the material functions are there but not linked to my model..

did I miss some weak or else option somewhere? but it works with scalar values?

Lukas

Hi

not what I can understand like that, duplicate name means that you have a T already from i.e. a thermal model, and p might not be used for your materials. Check the bold values, when they show formulas, they might mention variables names, and these MUST be defined, but are not always in the model. For pressure there is also the discussion absoolute or gauge value ...

Have you checked the solver properties page ? that you have all physics correctly linked in, and that they are not i.e. all in static, while you are doing a parametric sweep ?

often the easiest is to resume back to a 1 object simple test and to apply the physics and material properties one by one and solve independetly, then start to couple physics, this can hel pto identify the culprit

have fun Comsoling ;)
Ivar

Thanks again,

well, now it is mixed up completely.. i need to recreate it.

I do not know what all these settings means.. weak, or transient or how to know if the physics are linked or not..
they should apply to frames..

I will go back to some tutorial. the microwave oven..

cu later.

Hi again

yes really play the examples first, as COMSOl is a unique programme but it assumes that you know well your physics, and that you understands COMSOL's notation (of the physics-maths)

But once you have grbbed the essentials it's realy amazing ;)

Have fun Comsoling
Ivar

Error: 6297
Polynomial piecewise functions cannot be evaluated for complex arguments.
- Function: mat2_k_liquid_2
Failed to evaluate variable.
- Variable: kxx_htgh
- Defined as: ((mat2_k_liquid_2((T/unit_K_cf))*unit_W_cf)/(unit_m_cf*unit_K_cf))
Failed to evaluate variable.

I managed to have my previous running model where the water was heating with microwaves, but with the scalar values of Cp, K and rho.

When getting the bold Functions out of the Library (plots are there) I get those errors.

Yes, those mat2_k needs to be set somewhere.

So the question is: how do I import those functions/values out of the material library?
selecting and apply in the subdomain settings is not enough here.

how do I transfer the functions into my model?

is it enough to set a single value of each variable, afterwards the complete functions will be taken over?

I try..

Lukas

mat2_k_liquid_2 --- mat2_k_liquid_2(T)
mat2_rho_liquid_2 --- mat2_rho_liquid_2(T)
mat2_C_liquid_2 --- mat2_C_liquid_2(T)

I put this into options/expressions/scalar expressions

but this did not help.

I am running 2D, not like the tutorial.. may this make troubles?

I have for my water rf property eps: 80-i*10

Does this influence the error message? complex arguments?

honto ni mondai desu..

lukas

Hi

I have som trouble to follow you with your explanations. In RF normally a complex "epsilon" means absorption, but if I remember right (I'm not in my office and I have no Comsol at home ;) function calls should be only real, so you need one function for the real part and one for the complex part (to be checked)

Using the library materials means that you must understand their functions, and check that you are within their range, before you call them, several of these material data are only for a certain temperature range, and several have different states etc

Again, I would advise you to start first with only reasonnable constants for your materials, then later make them depend on temperature or whatever, one by one. This you can also do by making you own functions, there are very many sides in Comsol to master

have fun Comsoling
Ivar

Hi again

On thing stroke me, if you have these names mat2 ... appearing in normal typeface it could be that they are pointing to "old" material data.

By this I mean that if you load in a material, ad then you replace it by another material that does not have defined all variables, the new ones are in bold and the old ones in normal typefont, but then depending on how you clean up (or not) your model material db the values might point to nowhere thereffore an error, or what is more tricky, your material is a combination of the two (first and second choice) and your results are wrong

good luck
Ivar