When I set out my thoughts on the low level structure of a word section (which I believe is a syllable), I included a possible configuration which I now realize is wrong.
I had included the character [l] in with those which come after [i] sequences, such as [n, r, m]. But in truth it does not belong there. Out of nearly 8,800 occurrences of [l] coming after another character, fewer than 50 times is that character [i]. This is compared to about 750 out of 6,750 for [r]; nearly all occurrences of [n]; and 70 out of 1,000 for [m].
The rate for [l] after [i] is just 0.5%. It may well be valid in that it does not break the phonological rules of the language, but it is not usual or general. There is no reason to include it in the generalized structure as though the language often works like that—you could go for pages without seeing an example even though both are common characters. (Admittedly you won’t seem many [im] either, but [m] is much less common though still fourteen times the rate of [l].)
But character [l] cannot sit with [d, s], as they have different contexts. It would seem as though [l] is on its own in how it appears at the end of a syllable. Along with the possibility that [l] forms digraphs, it is beginning to look as though the character is not quite like any other.
I think there are a number of changes that may be welcome to my proposed syllable structure, so it is something to look at for the future.