^^ That may be, but brain chemistry has never been one of the three legs of my LD'ing stool (those legs being self-awareness, memory, and expectation/intention, the
Fundamentals), which could explain why you and I are different.
I think -- and have personally found -- that the brain is a bit more elastic than the textbooks and dreaming device manuals let on. You have plenty of ability to "stretch" out REM periods, and perhaps start one up, if your expectations/intentions and self-awareness are powerful enough.
That said, there are averages, and the REM periods you report don't come close to those averages. After six hours of sleep, you should be experiencing REM for a good half-hour, with diminishing lengths of NREM as the hours pass. By eight hours, your REM periods could last 45 minutes or more, with very brief bits of NREM in between. That's per hour, BTW, and not per 90 minutes. Are your machines working properly? ;) How much later on are you sleeping/monitoring that sleep? If your REM is truly that far apart after 6hrs, perhaps you should attempt to sleep for 9 or 10?
That said, I must mention that expectation/intention leg of my lucid dreaming fundamentals stool. If you are monitoring your REM periods in an in-depth manner, and also assuming (from those machines) that your REM periods are always a particular length per 90 minutes, they very likely will eventually wind up
being that particular length. I would recommend setting the machines and their data aside for a little while, and simply attempting some late-morning LD'ing (DILD or WILD, it doesn't matter), and seeing if DEILD's might work for you simply because you are confident that they will. Data is a fine thing, but don't let it redefine your activity.
As you might have noticed by now, I've never held a lot of faith in over-the-counter REM-detection devices, and certainly don't hold their reporting as a concrete foundation for when your dreams may happen. This may have caused some bias in me, I suppose, but it has also allowed me to confidently perform DEILD's whenever the need arises, without concern for maybe missing a REM period.
And that said, keep in mind that even if you are at the beginning of NREM, a DEILD might still yield results. Your conscious push to continue your dream could land you in a DILD later, with no percieved time lapse, or you might even get to enjoy a bit of
NREM dreaming while you await your next REM period. These options only exist, of course, if you choose to allow their existence. Assuming the machines' output defines unbreakable boundaries will indeed give you some unbreakable boundaries. Enjoy the machines, read their data, but don't let them run the show.
Okay, I see I've begun to ramble (and preach) here, so I'll stop.
Tl;dr: REM periods,
statistically, are longer and closer together after several hours of sleep, and very close together after 8 hours of sleep, though yours could be different. Despite that, trust yourself, not the toys.