Weight factors for thermal conductivity are wrong for water and organic

[orting]

It looks like there is a mistake in the computation of weight factors for thermal conductivity for water and organic.
All equations are from
[0] from https://daisy.ku.dk/technical-manual/Ch_5_Heat_Transport_alone.pdf

From equations 5.17 and 5.20 we have

$f_{ij} = 1 / (1 + a_i g_j)$
$f_i = \sum_j^3 f_{ij}$

where
$a_i = K_i/K_0 - 1$

(Note that $f_i$ is missing a factor of $1/3$ to ensure that the weight for the continuous medium is 1. The implementation has that factor but we ignore it here)

From text to Figure 5.2 and the implementation we have that organic particles are modeled as elongated cylinders with an elliptical cross section.
From equations 5.25(c,d,e) we have that the g_j parameters in that case are
$g_1 = 1/(m+1)$
$g_2 = m/(m+1)$
$g_3 = 0$

The parameter m is fixed at 3 in the implementation.

This leads to
$f_o = 1/(1 + a_o \cdot 1/4) + 1/(1 + a_o \cdot 3/4) + 1/(1 + a_o \cdot 0) = 1/(1 + a_o/4) + 1/(1 + 3a_o/4) + 1$

The implementation [horheat.C line 278-280] computes

const double Alfa = -3.0;
k = (1.0 / (1.0 + a / (1.0 - Alfa))
+ 1.0 / (1.0 - a * Alfa / (1.0 - Alfa))) / 3.0;

which becomes
(1 / (1 + a/4) + 1/(1 + 3*a/4)) / 3

where a term equal to 1 is missing.

The same mistake is present for water and possibly ice but I have not looked at ice yet.

From [0] on page 9/10 in the paragraph on "Calculation with air as continuous medium" we have that the g_j factors are
$g_1 = 1$
$g_2 = 0$
$g_3 = 0$

This leads to
$f_w = 1/(1 + a_w \cdot 1) + 1/(1 + a_w \cdot 0) + 1/(1 + a_o \cdot 0) = 1/(1 + a_w) + 2$

The implementation [horheat.C line 261] computes

k = (1.0 / (1.0 + a)) / 3.0;

where a term equal to 2 is missing.