Consonance circle scales

After pondering the commatic scale I used for the Prokoviev retuning,
I was led to consider circles of intervals centered only very roughly
around a fifth. If we take 7/5, 10/7, 3/2, 8/5 and 5/3, the deviation
from a fifth involves 15/14, 21/20, 16/15 and 10/9. If we take
products of the form (15/14)^a (21/20)^b (16/15)^c (10/9)^d which come
to 2^19/3^12, the reciprocal of the Pythagorean comma, it turns out
there is only one such, (15/14)^(-3) (21/20)^(-3) (16/15)^2 (10/9)^2.
This leads to a cirlce of twelve intervals composed out of
(7/5)^3 * (10/7)^3 * (3/2)^2 * (8/5)^2 * (5/3)^2 = 128. Each of the
fifth-like intervals in this circle is a 7-limit consonance, there are
no wolves. Of course, the 12-et mappings range from 6 to 9, so if we
want to retain the circle structure in a retuning we would need to
order the scale out of linear sequence.

However we order it, such a scale might be called a consonance circle
scale. There are still too many of them to try to assess all of them
by brute force, though assumptions of one kind or another can cut that
down. For instance, we might assume the two fifths are contiguous,
leading to a 1-4/3-3/2 subscale, and two tetrachords. We probably
mostly should alternate small with large "fifths"; in particular a 7/5
and 10/7 together is an octave, which we probably do not want. A 7/5
paired with a 5/3 leads to 7/6 intervals, and a 10/7 paired with 8/5
leads to 8/7 intervals, and these are obvious ways to proceed. That
leaves us with on 7/5 paired with a 3/2, giving 21/20, and one 10/7
with a 3/2, giving 15/14. We are from one point of view looking at the
union of two six-note scales of
(21/20)*(15/14)*(8/7)^2*(7/6)^2 type, and that should be manageable.