The Senyecan language has 30 phonemes: 24 consonants & 6 vowels. The
consonants are arranged on a grid of 3 columns & 4 rows. The 3
columns are plosives, fricatives, and sonorants. The 4 rows are
bilabials, dentals, alveolars, and palatals. This gives 12 pairs
when voicing is considered. The pairs are (and this is the Senyecan
alphabetical order): p/b, f/v, mh/m; t/d, th/dh, lh/l; ç/ß,
s/z, r/n;
and c/g, ch/gh, yh/y. These are not the graphemes I had originally
decided would best serve (I have an aversion to digraphs), but I had
to modify them for the conlang group. I chose estset since the
Germans aren't using it any more :-)

In X-SAMPA these would be: p/b, p\/B, m_0,m; t_d/d_d, T_d/D_d,
l_d_0/l_d; t-\s/d-\z, s/z, 4-0/n; and c/J\, C/j\, j_0/j. The only
allphone in the original phonetic system was that 4_0 often becomes
r_0
in the "interrogatory r" which is always sentence final.

The 6 vowels are ordered from close front to open back rounded: i, e,
a, o, oe, u, in X-SAMPA i, e, a, o, 8, u. In the alphabetical order
the 6 vowels follow the 24 consonants

There is also an epenthetic schwa (@) used to avoid unacceptable
consonant clusters. It is never written in the original orthography,
but I always include it when transliterating to the Latin alphabet.
In the conlang group I will use ü.

I know that some of these don't correspond to modern phonetics, but
who am I to argue with a native speaker!