disagree on the ellipsoid height mess bit. You can use the appropriate geoid file to convert the ellipsoidal heights to orthometric heights. I may disagree on the converting to another reference system, depending on whether that coord system is based on WGS84. if it is based on WGS84 then with published formulae you can convert to UTM coordinates. you also have the opportunity to convert to custom coordinates by entering your own central meridian, central scale factor and false E and N to use in the published formulae. I don’t know what your ‘revision mess’ means.
i got there eventually, but this system was a pain for me. there are a few of geometric problems to solve to make this an accurate system for anything larger than a site of 10 acres or so (which includes location of base station). this includes earth curvature, conversion of plane E-W distances to ground distances and grid convergence I think.
For me, having LLh is no problem using a spreadsheet. Although the conversion of LLh to UTM plane coordinates involves a few terms and long formulas, you only have to write them once, then copy to as many cells as you like. Conversion of 1000’s of LLh to ENh in as much time as it takes to write the formulas once. Like I mentioned earlier, write the formulas to accept variables for central scale, CM and false E and N and instant conversion to UTM or custom plane coordinates. I also pull surrounding geoid separations from a lookup table for every single point and do a bilinear interpolation for a unique geoid value to convert every point to orthometric height. from here I simply import into photoscan as local XYZ cartesian coordinates, or import as UTM ENh coordinates if I want to use the products in QGIS later.
i hope i got the gist of your questions. more than happy to expand on anything if you need.