[Avid-L2] Do Heat Sinks and Thermal Pads have a break in period?

I had to replace my NVME M.2 SSDs after heat related failure.  Fortunately they were still returnable.  I got replacements and a slew of thermal pads and Heat Sinks.  The M2s that failed had reached temps over the 70 degree C limit.  I noted some hitting 73 degree C.  Now with all the heat sinks and pads I'm testing the unit and so far on a long 8 hour copy to the new M2s the most I've seen is 58 degree C with an idle temp of 48 degree C.  As this copy progressed the 56 to 58 degree C temps that I saw in the first hour of copy have gone down to 48 to 52 degree C.  Given the transfer rate hasn't changed and the ambient temp has gone up a couple degrees I'm wondering if as the M2s heat up do the thermal pads that are stuck to the M2 to transfer heat to the bottom of the drive sled and also to the full length heat sink on the top of the M2 warm up and somehow adhere or conform more to the M2 surface?  Would that account for the overall drop in Temp after the first hour? The full length Heat sink is on a full length thermal pad sticking to the Heat Sink and the Top M2 Label side and it's held with three rubber bands that came with the Heat Sink.  I can't think of any other reason the over all temps of the 5 M2s dropped from 56 to 58 down to 48 to 52 degree C.  I haven't found anything googling yet but I'm looking.

Another interesting behavior is the second of my to NAS units that holds 5 M2s would run up in the 63 degree C range without any heat sinks or thermal pads.  Then I noticed when a copy ended the temp would soar to 70 to 73 degree C sending the units fans into full throttle mode.  It would happen repeatedly just as the finder copy progress bar would disappear.  I only notice this behavior on my second unit which is where 3 of the 5 M2s died an early death from what must be overheating.  I'm waiting to hear from the manufacturer on this but I'm curious if there is something that happens at the end of a big copy that would cause the M2s to suddenly work super hard and generate extreme heat.  Is there some sort of indexing that happens at the end of a copy that might account for this behavior?  I haven't ruled out a hardware problem on the second unit but I will no more once I get the new M2s in it with the Heat Sinks and Thermal Pads and see how it performs.  the only other difference is my first unit is in an iso box quiet box and has air forced from the bottom of the rack and exhausted out the rear top.  Perhaps that creates a better cooling air flow but the exhaust temps are are around 83 degree F where as the second unit is in the room open air which is more like 70 degree F.  I would have thought the unit in the warmer quiet box would run hotter than the free standing unit but what do I know?

John Moore Barking Trout Productions Studio City, CA bigfish@pacbell.net