Where do you get the form *nata from?

mauya as a modal verb governs a dative in my opinion (mauya tyen)

Well, you can almost always avoid such unpleasant situations if you try to rephrase your sentence, e.g., *cî nidhig mabed darias, boe le ui raithad a tharias (this could be the slogan of a potency pill :D).

Btw, I don't think that bronadui in your sentence should be lenited.

Perhaps we can use ola-, at least for the first half. I see no option nor reason to avoid na- in the last part. Homophones are ever-present in Sindarin, we cannot change that.
Pí anírol vronadui olad, illû bórol chim nad.
"If you want to become enduring, you always need to be enduring."

+Tamas Ferencz I usually draw back on Thorsten Renk's Grammar courses, where he also features this extended infinitive which fits very well where you have an infinitive with "to" to translate. Also found that form in some other courses, could not name a primary source though off my head.

I actually did search for some indications whether mauya- requests a certain case, did not find something for certain, and then just used Renk's elaborations on the passive in Quenya (translating "you are compelled"). Renk basically describes this rendition of the passive as a translation of literal "*it* compels you" where the it is implied and thus the "you" is in accusative and not dative. But I definitely considered a construction similar to known Q. óla tyen.

+Lőrinczi Gábor Wow thats a fancy way to rephrase this sentence :O But to my taste its a bit too far away from the original wording, and not preferrable if there's a more or less safe way to do it otherwise :S

Can ui be used as standalone word? Only knew it as prefix so far. I love +Ekin Gören's illû, a lot better than the bulky and too often used an uireb :P

What's that ola- you're speaking off? Haven't heard of that word as of yet starts researching

+Severin Zahler Using "ui" adv. would do the trick as well but "an uireb", not at all. Since it's "for eternal", rather than "for eternity" (which would be an uir). For ola-, PE22/134.2610

Yeah agree that an uireb was far off optimal, although I think it would have worked as no other safe solution jumped to my head right then.
Could have guessed that ola- is from almighty PE22 :D Did read it twice and am chipping in more and more bits from it into my translations, but am still far off knowing about all of them^^

You can also avoid the using of the gerundium of na- if you change the second part to an imperative clause, e.g., no ui tarch ("be always tough") or just no *uidarch ("be ever-tough", cf. uidafnen).

+Severin Zahler if I am not mistaken the only source for the extended infinitive are the sentences with karita in VT42 (see them listed in the references on http://eldamo.org/content/words/word-2562596025.html) but I personally don't feel that's sufficient to make it into a universal case, especially as we also have sentences in the same essay using the simple infinitive kare in the same position.

I saw somewhere the infinitive from ná to be náve. +Tamas Ferencz ?

Oh, I mean in Quenya, of course.

+Andre Polykanine náve is indeed attested, but it rather looks like an abstract noun to me: http://eldamo.org/content/words/word-3259716995.html

Thanks a ton for all the inputs! Here's what I'll go with:

Q.: Qui merityë na voronwa, mauya tyë na voronwa illumë.
S.: Cí nidhig bronadui nad, boe achen illû bronadui nad.

Regarding the word for "if" in Sindarin: Both the variants cí and ae are neologisms created by David Salo for the LotR and Hobbit movies, going back to Q. cé resp. ai. The qui I am using comes from LVS11 (PE22/158:1605), previously I used quë << qe from QVS (PE22/120:1810). LVS11 is the newer source (1969) over QVS (1948-1950), hence my choice. To choose between ae and cí for Sindarin I do refer to QVS though which has both Q. ai and qe while qe is for "if" and ai is for "maybe, supposing" which definitely fits less well.
Now LVS11 has both qui "if" and cé "may be", and David Salo's NS. cí is said to be derived from cé rather than qui, but given that PE22 was not released when this derivation was created I think I can safely reconnect cí to qui rather than cé...

Again thx for sharing your thoughts!

+Severin Zahler My approach to the matter:
Quenya: KW>qu | quetta / qui
Sindarin: KW>p | peth / pí