Extreme Voltage fluxuation fixed by CPU Ratio?...Seriously lost.

[Hardly_Qualified723]

Zero overclocking in use, any reference to OC is just to clarify my part's.

434 Watts according to PCPartPicker

Intel Core i7-7700K 4.2GHz (Quad Core--New)

Hyper 212 EVO 82.9 CFM Sleeve Bearing, Dual Fan (Both stock, one from another EVO--not an actual dual fan model) - one in CPU Fan1, other in ?Pump fan?

MSI Z270 GAMING M7 (new)

Corsair Vengeance LPX 32GB (2 x 16GB) DDR4-2133 (2666 OC)

Samsung 850 EVO 500GB 2.5" SSD (Normal bay, seperate cord)

Toshiba P300 1TB 3.5" 7200RPM HDD (x3 in six bay hot swap, only one inside a bay or in use at any given time, seperate sata cord)

EVGA GeForce GTX 1070 8GB SC2 iCX (EVGA PowerLink using a single cord, but both plugs, even though my card technically only has one power port--no problem before this)

Thermaltake Level Ten GT (fans to floppy power)

EVGA SuperNOVA G3 850W (new)

LG WH16NS40 Blu-Ray/DVD Burner (seperate sata cord)

Win 7 Professional SP1 OEM 64-bit

Running off of a Tripp Lite SU1500XLCD (1350w, IIRC) UPS (non-eco, 120v at all time's)

Everything is from my previous Z170A Gaming M7 / I-6700K build that the Turbo Boost broke on (without use, luckily during warranty), except the EVGA Supernova 850 G3 which replaced a Corsair CX850M, the motherboard and processor.

Problem: All kinds of voltage alarms through MSI Command Center, my old build never had a single issue.

My PC Generally 'had' a VCore of 0.739 - 0.759 in BIOS and that was the alarm I saw the most (low or high), but I've even seen my 12v, 5v, and 3.3v set off an alarm during normal use or gaming (unfortunately too fast to give the values)

MSI was clueless and told me to RMA, EVGA said it was MSI software, but this is what confused me...

I set my CPU Ratio to 42 (Dynamic), Offset for AVX to 0--which puts my Frequency and Ring Ratio (ring still on Auto) at 4200Mhz and so far seem's to solve my problem...oddly.

CPU Base Clock is at 100.00 (default), apply mode Auto.

Never touched any voltage because I'm clueless there...

I plan on swapping back my old power supply as a test, already checked the chip and heatsink placement. There's def. no random screws or metal in my case after how many other thing's i've checked.

BIOS is up to date v1.50 (v15 by filename)

Is this all likely software related and I'm freaking out for absolutely no issue or should I return the board to be safe?

When I changed my setting's, so far I havn't got any alarms.

Only my CPU Core changed voltage wise, which make's sense...

After changes:
CPU Core: 1.200v - 1.209v (WHEN it fluxuates, it does it constantly, sometimes it stays at either value flat)

Before my changes:
CPU Core: 0.739 - 0.759

CPU IO: 0.942v - 0.952v (always changing, but generally in that area regardless of setting's)
CPU SA: 1.056v (stable in BIOS)
System 5v: 5.040v (stable in BIOS)
System 12v: 12.288v (stable in BIOS--seemed a bit high though)
DRAM Voltage: 1.200v (stable, no alarm...ever)

For some reason they don't come up stable while running as far as Command Center is concerned. Low or High. CPU-Z gives me the same VCore as Command Center NOW, I didn't check before as I was frantic, ran a bench and hit 1.210 tops for VCore, temps at 50-60 depending how long it was running before.

So generally, I wanna know if this is likely just software, or I'm justified in freaking out and if I am, what's the most likely suspect out of my three new part's? EVGA say's that all the voltage going off is near impossible on the PSU end.

EDIT: To test a single power supply port as unlikely as the situation may be on a PSU running one rail, it's not impossible--a multimeter was suggested by EVGA, although I know there are less costly, more risky method's.

If it's software, any good recommendations exactly how to rectify it 'properly' in the BIOS or even to test more thoroughly?

Totally at a loss, along with every company I've spoken too...any help would be much appreciated.

Thanks in advance, back to seeing if the alarm goes off again...

EDIT: I just put the ratio up to 44 and 45 (for the first core--stock for Command Center) and the alarm went off immediately during a bench using CPU-Z. It doesn't tell me much, except not to do it lol. But maybe useful info to anyone reading this with a better understanding than me.

EDIT-2: I was about to unplug my power supply and swap with my old one (which is missing the CPU2 4 prong power cord) now that I had time without worrying about losing the PC for a bit...CPU-2 was the first cord I removed since it wasn't going back and seeing how I was told it wasn't needed unless I was overclocking my PC, I stopped mid process and threw my system on real quick, reset all my BIOS options (again), and suddenly i'm no longer getting voltage alarm's from a simple CPU bench on CPU-Z (fingers crossed).

If this does fix it, i'll leave it as an answer myself and mark it solved in case anyone else has this issue.
I'd hate to see anyone else go through as much as I have to find THIS was the problem...talk about a kick in the crotch.

EDIT-3: Waiting on a rushed replacement wire from EVGA, being great like they always are as far as customer service goes, had one ready to be shipped out at 2 in the morning for me, it IS the problem, but i'd rather leave the answer complete with the final result of what was actually wrong, MOBO or PSU.

No responses needed at this point...just updating the post in full for reference. PC is running fine, doesn't mean i'm keeping a bad MOBO if I can avoid it though.

Thank's to anyone who did read this and attempted to figure out what it might have been with such a confusing mess of info, and no way to test, I do appreciate it.

 

[Hardly_Qualified723]

All right...sorry for the late solution, but I wanted to be able to give the best possible resolution to this ungodly phenomenon that I could...and even after all this time I still can't, unfortunately. Either my MSI Z270 Gaming M7 is a fluke, or the board and/or software is a fluke in general.

Simple Version: On my SSD drive, it was the four prong power cord, without a doubt.

As much as I hate to admit my own foolishness, I actually followed the manual and in MY manual at least, it show's the wrong end of the 4 prong CPU-2 power cord (they are literally reverse), no matter how many time's I look...at any angle, it's backwards, no doubt about it.

I matched the new cord (since it destroyed the original) with the MOBO itself this time and remembered why I never read instruction manuals lol. SSD, problem solved.

Now, when I use my hotswap drive's, if I load something too fast--before command center generally (which I have set to auto-load). My voltage alarm goes nut's again and my 5V said it was taking 10V lol...pretty sure that would fry my board or shut off my PC due to fault protection at the very least.

On a side note, fault protection causes my PC to need to wait after shutting down on a hotswap even as the main and only drive (set to hotswap or not) or it will shut down my PC as if there is a power problem. When fault protection is off...no problem and no error's anywhere.

So, basically...if you followed an MSI installation manual or Command Center is the thing complaining...they're both terrible. Check your CPU-2 power cord (I just left mine out since I'm not OC'ing) as a first step.

Longer Version:

If the above is not the problem, first check your BIOS voltage and if that's within decent range it's probably Command Center.

You can always use a Multimeter to check a power supply and MOBO if you know what your doing (I didn't, but I will soon enough), if not you can swap the power supply or just try without the cord like I said, at least it take's the PSU cord out of the equation.

Other idea's I was given was to remove my entire MOBO and run it on the box in the situation your not 100% sure you didn't have a loose screw floating around or a bad metal contact, from there start stripping it down and see if you can't find something causing the problem.

The last option, an this hurt's...remove the heatsink and check the board and the processor for bent pin's--you know, we all have spare jewelry magnifying glasses around (I actually do just wanting a lens with a light attached lol)...or just return the part's if you are lucky enough to be within the time frame...kinda wish I did just because it was so much hassle.

So I'm still dealing with problems, but at least I'm sure it's safe to run, I have too much going on (when it rain's it pour's) to keep up with it and get the info up in a reasonable time frame, so I'm giving what information I do have in hope's it help's someone else who might be running around like a chicken with their head cut off due to a time frame like I was.

I don't believe I can update this after marking it solved, but if I can and anything new come's up...I will. Best of luck...and MSI, fix your damn software and manuals...&%$#s.

In all my years of PC repair/building this has been one of the dumbest, yet most ridiculous to pin down problems I've ever run across. When I do find the exact problem--it'll be a replacement, if I don't just switch every new part out one at a time anyhow...at this point I'm beyond 100% sure it's not a problem with my build itself or any fault of my own beyond the first wire...I've checked everything more time's than I want to admit, even to myself.

At any rate, I have been running it on all the different drives and even though I get a rare occasional siren going off, it only last's a single reboot (BIOS is normal) and it simply disappears...so for once I hope the software is just terrible.

Only time will tell I suppose.