Not sure if the standard you are using allows the Two Box Method used in Europe for simulating such spacer bars, if you can, then it will save a lot of time and work in Therm or CAD. Especially with complex shaped spacers that have very thin
walls.
The Bundesverband-Flachglas (BF) group have tested many manufactures spacer bars, below is link to a Technoform product.
Basically, there are two solid boxes, Box 1 being the secondary seal and Box 2 being everything else, i.e. spacer body, desiccant, primary seal etc. etc. each box has a defined lambda value given in w/m.k.
Alternatively, you could simulate the fully detailed spacer with all its associated materials in Therm (minus the glass, frame) to find is conductivity that you could use as a single material in your glazing and frame. An area weighted average
sort of thing.
Hope this helps.