[petr.f.]

Hi all!

Maybe it is a stupid question, but after looking into the code I'm still a bit confused about true meaning of h in the externalWallHeatFluxTemperature. If I assume the following boundary problem

internal fluid -> wall -> environment

there are two heat transfer coefficients coming into play -
alpha_1 between the internal fluid and the wall and
alpha_2 between the wall and the environment.

So which one of these is h?

[alexeym]

Hi,

If you take a look at the sources, here's the way heat flux is calculated:

        case fixedHeatTransferCoeff:
        {
            scalar totalSolidRes = 0.0;
            if (thicknessLayers_.size() > 0)
            {
                forAll (thicknessLayers_, iLayer)
                {
                    const scalar l = thicknessLayers_[iLayer];
                    if (kappaLayers_[iLayer] > 0.0)
                    {
                        totalSolidRes += l/kappaLayers_[iLayer];
                    }
                }
            }
            q = (Ta_ - Tp)*(1.0/h_ + totalSolidRes);
            break;
        }

Ta_ is ambient temperature, Tp is temperature on the patch. Now, assuming you neglected additional thermal resistance, h can be calculated as

1/h = 1/h_{flud->wall} + 1/h_{wall} + 1/h_{wall->external}

[petr.f.]

Hi Alexey,

thanks for the reply. I'm using OF v 2.3.x with a little bit different code

            if (thicknessLayers_.size() > 0)
            {
                forAll (thicknessLayers_, iLayer)
                {
                    const scalar l = thicknessLayers_[iLayer];
                    if (kappaLayers_[iLayer] > 0.0)
                    {
                        totalSolidRes += l/kappaLayers_[iLayer];
                    }
                }
            }
            hp = 1.0/(1.0/h_ + totalSolidRes);

and then

    else if (mode_ == fixedHeatTransferCoeff)
    {
        Qr /= Tp;
        refGrad() =  0.0;
        refValue() =  hp*Ta_/(hp - Qr);
        valueFraction() =
            (hp - Qr)/((hp - Qr) + kappa(Tp)*patch().deltaCoeffs());
    }

where Qr is radiative heat flux. You're right that from theory the overall hp value is

1/hp = 1/h_{fluid->wall} + 1/h_{wall} + 1/h_{wall->external}

I wasn't sure if there is something neglected in the code or not... So if it isn't then apparently

1/h_{wall} = totalSolidRes

so I presume

1/h_ = 1/h_{fluid->wall} + 1/h_{wall->external}

I suppose I have to estimate the 1/h_{fluid->wall} for example from



as mentioned in https://en.wikipedia.org/wiki/Heat_transfer_coefficient

and then do a similar thing for 1/h_{wall->external} to get the h_ for input of the boundary condition...

[petr.f.]

After looking into code of mixedFvPatchScalarField and fluidThermo I think I got it. The h_ is the outer heat transfer coefficient - the 1/h_{wall->external}.

Heat transfer from internal fluid -> wall is solved using the kappa() field, inside the wall with the use of totalSolidRes which leaves the h_{wall->external} to be set by the user.

[zfaraday]

Exactly Petr, h is the external heat transfer coefficient, the value of the solid resistance is calculated if you provide values of the kappa layers and thickness layers, otherwise it's neglected.

Regards,

Alex

[calf.Z]

Originally Posted by zfaraday

Exactly Petr, h is the external heat transfer coefficient, the value of the solid resistance is calculated if you provide values of the kappa layers and thickness layers, otherwise it's neglected.

Regards,

Alex

As usual, should we need the kappalayers and thicknesslayers? What's the meaning of kappalayers (1 2 3 4) and thicknesslayers(0.1 0.2 0.3 0.4)?