So I have an SR-2(love it!) overclocked modestly.  My RAM voltage is set for 1.55v.  When I built the machine I put in 6 sticks of 4GB gskill 1600 RAM.  I didn't do anything crazy with the RAM.  Have it set to less than the spec if I remember correctly since I'm not a RAM overclocking type of person.  I have RAM voltage set to 1.55v only for stability and had no problems at 1.5v.  Anyway, I've had 4 sticks of RAM go bad in my rig since building it in Nov 2010.  I may have had one bad stick to star with, I don't remember.  At one point I had installed some of my spare RAM, but it oddly wasn't detected completely either.  I ignored the anomaly as part of my overclocking settings and since the spare RAM was temporary I was okay with not having all RAM working.
 
So a few months back I noticed that only 20GB of RAM was showing up during BIOS initialization.  I upped the volage from 1.5v to 1.55v but that didn't help.   I ignored the problem after reviewing tons of forum posts talking about how some RAM doesn't get detected and whatnot.  But now about 2 months ago it went down again to 16GB.  I continued to ignore the problem until 2 nights ago.  I had a need for more than 16GB of RAM.  I rechecked everything and I didn't see anything wrong.
 
I've only seen 2 sticks of RAM go bad despite working in IT at the largest Navy base on the east cost in over 15 years of IT work.  So.. in my mind.. pretty rare.  More rare than newegg reviews would have you believe. :P
 
So I started troubleshooting my system.  Set everything back to defaults, got rid of overclock, updated BIOS, etc.  I was convinced that I had the same RAM problem other people have had in the forum and that I would not be able to fix this problem.  Started swapping RAM sticks and slots and finally after alot of on/off cycles I decided I'd go extreme.  Disabled CPU1 via jumper and tested one stick at a time in slot 1 of CPU0.  Sure enough, I had 2 bad sticks in the RAM I used in this machine.  The machine wouldn't POST with those 2 sticks installed.  I tested my spare RAM and sure enough 2 of those sticks were bad too.
 
I have a corsair 1200w power supply and a GTX570.  My wattmeter only shows 300w when gaming so I'm sure its not a power problem.  I did plug in every single one of those "optional" power plugs for those that are extreme overclockers.  Since I don't consider myself extreme I only did it to be safe as I prefer system stability ;).
 
So is it possible that I have a setting wrong that isn't the RAM voltage that is causing this failing RAM?  I'm inclined to say no, but having 4 bad sticks to RMA really peaks my interest that I'm doing something wrong or something is wrong.  Per e-leet my RAM voltage is set to 1.55v.  I haven't used a voltmeter or anything to check the voltage.  Should I get a voltmeter just to check the voltage?
 
I haven't tried to use those RAM sticks in any other machine.  I will be in 2 days when I have a spare system to test them.  It would really add to the confusion if they test perfectly in another machine but not the SR-2.
 
Thanks for any insight!

My RAM says 1.5-1.65v.  I actually had no stability issues at 1.5 and only upped it to 1.55 hoping to fix the issue.  It didn't obviously.
 
It could be I just ended up with a bad batch.  That might be all it is.  I'm just surprised at how many bad sticks I found and how sudden it was.  The last bad stick of RAM I have seen was 2005.