Nice.
PE22 also gives examples for the use of past imperfect as a consuetudinal (habitual) past, e.g. karalyane "used to do/make", so you might go for *quetalyane

+Tamas Ferencz Thanks. I forgot about that material when rendering this; I must be a bit rusty. Is that past imperfect using the older imperfect participle still valid in the later language or should a construct like quetilane or né quetila be used? I'm also assuming that the past continuous quétane doesn't necessarily have the habitual connotation.

+Robert Reynolds that is indeed a dilemma not easy (for me) to resolve; for if we chuck out karalyane do we also chuck out other contemporary forms of the paradigm, like karúvane etc.? I am unwilling to do that as they bring a welcome expansion of possibilities. I am not saying they are absolutely necessary (languages can function perfectly well with just the three basic tenses after all), but they add diversity, and an "ancient" flavour to the language.
I am now leaning towards the idea that I (i.e. in my idiolect) keep the karalyane forms, but only in their consuetudinal "used to" sense; otherwise the active participle would be in its -ila form everywhere. After all, Tolkien's reason for chucking out the -lya ending was its clash with the possessive pronominal suffix - there is no danger of that in the karalyane forms.

+Tamas Ferencz This does seem like a fertile general discussion topic. I, too, like the plethora of forms based on a few core building blocks for specificity, expressiveness/variety, "completeness"/naturality (lack of excessive "artificial" restrictions within the low-level linguistic framework), and taste. Your ideolectical idea sounds quite sensible in a tricky situation. I think that preserving or at least permitting most such "natural" forms when they don't clash problematically with later structures is generally desirable, as is updating when sensible as with the participle. Hopefully there are ways to retain language consistency and elegance during such processes.