I have not definitively figured out my issue just yet, but at first glance it appears that one of my AC Adapters for one of my NUC6i7KYK NUCs may not be liking the battery output of my APC Back-UPS XS 1500 (Model BX1500G).
I have two identical NUC6i7KYK systems, both on the same model of APC UPS. One is mine, the other is my wife's, in the same room. I also have a third UPS which keeps my NAS and home network alive, in the case of a power outage.
- Yesterday morning at exactly 8:00am, all three UPS units attempted a self-test (I think). I believe it happens every 2 weeks.
- The one holding up the NAS and network does fine.
- The one powering my NUC does fine, and I get the notification that the unit passed the self-test through the APC PowerChute app. I was just doing some light browsing in Chrome at the time.
- My wife's NUC rebooted as if you had yanked the power plug and reinserted it. She was in the middle of playing Warcraft at the time.
I immediately began checking her UPS. Was the battery bad? The UPS unit did not think so. Was there a real power glitch and did I have the plug misinserted in a "Surge Only" socket? Nope - it was properly plugged into a socket for battery backup. When the system came back up, I manually ran the self-test and it passed (but we were not playing Warcraft, or running any heavy loads).
I'm beginning to wonder if the AC Adapter that shipped with these NUCs is sensitive to the "stepped approximated sine wave" output of these UPS units. Is it possible that under light loads that they work OK, but under heavier workloads they fail and reset themselves? Do these AC Adapters have active power factor correction? This is sometimes a problem with the stepped sine wave output of some UPS designs.
I guess for a definitive test, I should run some form of heavy workload and then manually try the UPS self-test procedure again. Warcraft is running full screen and that's not convenient or consistent. Is there some form of Windows-based PC stress test I should try?
OK - So far it all seems inconclusive. I can't figure out why the one computer failed to run on the UPS the other day.
On both computers I ran Prime95. The UPS's reported that this took the original 34 watt loads to 86 watts. Both computers performed very consistently, fans ramped up as expected. I then ran the APC PowerChute Self-Test after the computer was running Prime95 and the consumption was 86 watts. Both UPS's passed. As I was a little frustrated that neither failed, I then did the "cord yank" test on my wife's UPS and the system still held up without a glitch.
It would seem at this point that the Skull Canyon-APC UPS failure I found is intermittent.
I'd like to get Intel's response on the Skull Canyon NUC's AC Adapter's compatibility or sensitivity to the APC UPS's stepped approximated sine wave output.