Sorry but I will disagree strongly on this and I can actually prove this.
I had my MMM set at a 3.2v/cell lipo cut off, however I was curious as to the packs behaviour so I put on a data logger:
Under load the pack frequently drops below 3.2v/cell before the LVC kicks in, the sampling rate of the LVC is such that it's really based on the packs resting voltage (which stayed over 3.2) not the underload voltage.
Lipos can rebound from amp draw a lot quicker than Nimh, hence a Nimh will trigger a LVC wheras a Lipo may recover before the LVC takes the next sample (sampling rate), My logger was set to take 100 samples a second which was obviously quicker than that of the LVC as it easily showed the packs dropping below what I had the LVC set at. And remember EVERY drop below 2.7-3v a cell does damage to some extent.
Also the drop off is quite noticable when the packs resting voltage drops below 3.5v/cell which is why I have chosen to set my LVC at this point.
Going by your assertions the pack drops frequently below 3v/cell before the LVC kicks in. Obviously this has the potential to cause damage to Lipos over the longer term and as Freezebyte has said he ran to LVC, I think that's a contributing factor, the packs got overly discharged (albeit under load) until one day they gave up.
Happy to share the raw data file, and yes my MMM LVC works fine and has been tested.
But if we followed your advice, on an RC car that pulls some serious amps under load as seen below: (Same Run)
Our lipo batteries are going to amount to nothing more than a
