I haven't attempted this exactly, but have tried something similar using toner partitions with different colors inks of the same shade that are used in two of the gray ink positions (shade 1,2,3,5 in carbon as the gray partition, and shade 3 and 5 selenium as a partitioned toner). QuadTone Rip is so capable and there are so many possibilities in approaching profile making that you could spend years testing them all out.
I'm am testing this approach real quick with a set of ink limits and cross overs I have for a 6-ink profile. I just stripped out the ink that aren't being used and moved the gray inks and gray values up to the proper positions. I'll let it dry and linearize it tomorrow and see how it shapes up.
Though from the test gradients I just did with the 1,2,4,6 and the 1,3,5 profiles it seems like the darker inks are dithering too much with the 3 ink profile.
The two main concerns with this approach are: 1) how the different overlap points of each of the three linearized profiles will mesh together to possibly cause banding artifacts in smooth gradient. 2) if the black ink channel with the 4 and 3 ink profile will be too far "to the left" and might block up the shadows and cause problems linearizing.
I'm looking forward to see how this works out.
Thanks,
Richard Boutwell