Sorry for late response, had been away on a vacation
Have a look at doco:
The second link states it is the height from wall with unit of lengthl. Hope this helps.
well thats all clear but I'm rather trying to correlate the "average rougness height" used in CFD to more common definitions like here:
or the roughness in the first link I posted.
Example: the customer has a pipe with Ra = 3.2. Would entering 0.0032 mm as wall roughness value be correct?
Well, here is the catch! Ra value just represents the texture abnormalities of the surface but the value "r" in the SimCFD documentation (in length unit) is the sandgrain roughness and both are not always comparable. Nikuradse has done extensive experiments to correlate the two:
Nikuradse, J. 1933 Laws of flow in rough pipes. VDI Forschungsheft 361. In translation, NACA TM 1292, 1950.
You can find additional information here, and more if you decide to do a bit of search:
I am not aware of any formula that will straightaway give the sandgrain roughness from Ra value and there exists a lot of literature around this. From one of the references above, the RzJIS value given in the Wiki link you posted comes closest in approximating the sandgrain roughness and Ra is the worst.
Will be interesting if Autodesk personnel have anymore to add on this...