This sentence is somewhat eratic! What did You calculate? A new COP ?

For the "COP 5.6" I reduced it to very close to COP 5.0

There is a T4 multiplication error, and a poor heat transfer coefficient approximation. (The one used for Lugano seem OK).

There isn't enough information to assess the input power quality, and probably using an ε 0.8 (or whatever the later test found), the COP could go back up again. I posted all this earlier, in another who-knows-where-it-is-now comment.

Back to the earlier discussion: Since using the calculator I linked to, I can duplicate the Lugano results with their provided figures, I deem the calculator to be effective, or at least the math is compatible or equivalent with what was done in the report. Slightly increasing the radius of the main Lugano tube in the calculator effectively deals with the ridges for a decent approximation without a lot of fuss. (I think by 2 mm, maybe 3 mm ; I forget. Try it out). You can add the lengths of the caps, and rod sections together from each side to simplify things also (8 cm long by 4 cm diameter for the caps, for example). Don't forget to do the 2/3 area adjustment for the tubes after calculating their heat output, since they are bundled. I have no problem with that assumption, since the real calculation would be a nightmare, and probably make almost no difference.