WDVS old building, unheated basement: Perimeter insulation of basement wall?

Hello,

we are planning to renovate an old building (built in 1973) and insulate it with an ETICS.

The basement is not intended to be used as living space. Therefore, we plan not to insulate it. The basement ceiling will probably be insulated from above. (The screed has to be removed anyway because underfloor heating will be installed; insulation on top reduces heat loss through the concrete terrace.)

Now I am wondering how far down we have to extend the ETICS on the outside or whether perimeter insulation down to the soil is even useful. I have not found anything about this in the forum, only a rather special problem with heated basements is discussed. Otherwise, I have heard different opinions so far, from "insulate down to the soil, it also looks better" to "insulation just below the basement ceiling is sufficient." In a book I have ("Ratgeber energiesparendes Bauen" by T. Königstein) it is advised to go at least 50 cm below the basement ceiling, which in our case would be almost down to the soil. What is your assessment of this? Does anyone have experience with such a construction (insulation not down to the soil)?



In my mind, the basement with perimeter insulation would be somewhat warmer in winter and we would need to heat a little less. However, that will probably never recoup the thousands of euros for the perimeter insulation. But if by too short insulation we create a critical thermal bridge and thus a mold source, it would be worth the money to me.

Happy Easter,

Lukas

P. S.: Are there actually programs/websites that private users can use for free to calculate isotherms for two-dimensional problems like the one I described above? I only know one for 1D problems (wall, ceiling, roof, but no corners). I would rather not pull out the big finite element tool (I have it at work, but I never do thermal calculations there and would accordingly take a long time), even though I would be interested in the result.