Using wwtrim function from WRS package

I am trying for quite a while now to run a robust analysis on my within-by-within repeated measures design data.

IV: The first factor has 10 levels, the second factor only 2 levels. Each of the five participants did obtained data in all conditions.
DV: The dependent variable is numeric continous.

As the residuals of the data are not normally distributed, I found the wwtrim function from the WRS package (explained in Wilcox (2012) Introduction to Robust Estimation and Hypothesis Testing) to be optimal.

Does someone has experience with this function?

I frequently get the error message:
"Error in solve.default(test) :
system is computationally singular: reciprocal condition number = 4.37511e-18"

Thank you!