I know that /ɚ/ is longest when it occurs at the end of a word, since it is occurring at the end of an open syllable, as in < rapture> [ræp̚tʃɚː]. What about when it occurs syllable finally in the middle of a word, as in < saturated> [sætʃɚejɾəd]? Since rhotic schwa is at the end of a syllable, would the correct transcription be [sætʃɚːejɾəd] or [sætʃɚejɾəd]?
To be clear, I understand the question to be asking about phonetic, and not phonemic length, since your transcriptions are in brackets. On the phonetic level, "length" is obviously gradient and not categorical, and it can vary substantially between speakers, or between different utterances made by the same speaker. Any IPA transcription of phonetic length will necessarily be based on a lot of subjective choices.
Although you transcribe the sound you are interested in as [ɚ], and refer to it as a vowel, it seems to be common to analyze it instead as a syllabic consonant and to transcribe it as something like [ɹ̩].
To get to your actual question: I think that you ought to take into account stress as well. For instance, to me "rapture" doesn't actually seem to be a very good example of a word with a phonetically long ɚ sound, because the second syllable is unstressed. I think ɚ would be realized with greater duration in words like burr, burred, bird, prefer, preferred.
In a normal pronunciation of the word saturated, it seems very unlikely to me that the nucleus of the second syllable would have enough length to warrant the use of ː in a phonetic transcription. Syncope to a trisyllable [ˈsætʃɹeɪ̯ɾəd] seems possible to me, and I think this implies that the vowel doesn't have to be long in pronunciations where it is present.
In a word like preferring, the nucleus of the stressed syllable would have greater duration than the nucleus of the stressed syllable in a word like converting, so there might be some reason to use a transcription like [pɹəˈfɚːɪŋ].