Lenovo M82 PSU Upgrade Issue

[MrChompenstein]

Long story short long, I purchased a refurbished Lenovo M82 to use for a budget secondary gaming/backup/htpc PC. This computer will make use of some of the old computer parts I have laying around, and will allow me or someone else to play older titles in another location in the house. Essentially for $150, I got an i5-3470, 8gb ram, and a copy of windows 10. This system was not intended to be up-gradable, thus I saved some money as compared to something like a Ryzen 3 1200, ram, and mobo combo. To this system I will add an SSD, old R9 280, and a 750W power supply (all of which I already had), and in the future add some large hard drives to backup data from other computers.

The issue:

I have concerns about adapting a normal ATX PSU to the motherboard that uses a proprietary 14-pin connector. I have already acquired an adapter for this exact purpose, similar to this one. While the link only references the m92p, according to Lenovo, these systems all use the same mobo. Before attaching the new PSU, I tested the voltages of all of the pins to make sure they were the same . . . and all by 1 were. For reference, post #18 on this thread depicts the layout for this 14-pin connector. In the picture, the white locations represent empty/unused sections of the connector. The problem is, my connector has a pin in the bottom right position. More odd than that, this pin has a tested voltage of 1.180V, which I have never seen on a PSU . . . The 24-pin to 14-pin adapter put 5V at this location. Maybe worth noting, this 1.18V wire is also 22AWG unlike all the others at 18AWG.

 

From what I can tell, this pin has something to do with CPU VCORE, but may only be some sort of reference?
The Testing I have done.

OEM PSU HWMonitor and CPU-Z:
VCORE at 0.848V (Min/Max of 0.848V/1.120V)
VID 1.166V (Min/Max of 0.901V/1.191V)

EVGA PSU (with the 5V wire removed) HWMonitor and CPU-Z:
VCORE at 0.032V (Min/Max of 0.032V/0.032V)
VID 1.166V (Min/Max of 0.891V/1.191V)

EVGA PSU (with the 5V wire) HWMonitor and CPU-Z:
VCORE at 2.608V (Min/Max of 2.608V/2.624V)   ----------> Multi-meter confirmed output of 5V, so I have no idea why this is now 2.608 . . . the 1.180V seemed to stay the same, so some regulation is going on? 
VID 1.166V (Min/Max of 0.901V/1.191V)

My understanding is that VCORE is the actual voltage given to the CPU, whereas VID is the "requested" voltage by the CPU. But I find it hard to believe that the computer would even boot with only 0.032V, or survive 2.608V.

With the 5V pin removed and VCORE at 0.032V, the computer ran Prime95 Small FFT without issue . . . HWMonitor did also report an increase to 44 watts under load but no change in VCORE. The CPU will only turbo to 3.4 Ghz with all 4 cores at 100%, this was the same for both OEM and EVGA PSU. NOTE: I did NOT test Prime95 with the 5v wire attached, the computer was shut down promptly after seeing VCORE . . . although at this point I don't even know what this pin is doing. 

Any insight into what might be going on here would be appreciated. If the CPU loads up fine without this pin, I may just continue without it.

[aezakmi]

you know what they say, if it works don't touch it

 

if it boots and runs p95 without the extra wire I'd say you use it that way, since you're dealing with proprietary motherboard there's no way to know what's that exact pin for except if you find some data sheets of that board

[STRMfrmXMN]

HWMONITOR CANNOT ACCURATELY READ VOLTAGES FROM YOUR PSU

 

UNLESS YOU HAVE DEDICATED HARDWARE FOR MEASURING SUCH THINGS (READ "SEVERAL THOUSAND DOLLAR SUNMOON LOAD TESTER") THEN YOU DO NOT HAVE THE HARDWARE TO DO THIS

 

Sorry for the all caps, but that software has caused more issues than it's solved with how many damn times we get messages from people on our forum about "OH MY GOD MY PSU 12V RAIL IS READING 12.8V IS THAT OK?" when they've used this software that isn't even accurate at reading temperatures...

 

If you want to read CPU Vcore and such, then download HWiNFO64. Otherwise, get something like a Corsair RMi if this specific voltage reading is a feature you want in your PSU.

[MrChompenstein]

Before the new PSU was ever connected to the motherboard, each pin was tested individually with a multimeter (In this case, a Fluke 110 PLUS) to verify the adapter configuration. The voltages seen in HWMonitor and CPU-Z seem to align just fine with the readings I measured beforehand, obviously there was some voltage drop across the motherboard. You do not need JonnyGURU's setup to ensure that the pins are receiving the correct voltage. The advantage to a load tester is that it can ensure the voltages stay within their nominal ranges throughout the range of loads that can be applied to it. With just a multimeter, I am forced to put the probes in series with each individual wire I want to test and also require a computer to generate the load.

 

I did test the wire in question. At idle there was .12A at .848V, under load it increased to .21A at 1.16V. It would seem that the wire does something, but at roughly a 1/4 Watt, it can't be doing much.

I also downloaded HWiNFO64 like you suggested. While certainly a more capable program than HWMonitor, HWiNFO64 showed similar results. Without the extra wire, under Prime95 load, Max VCORE was 0.022V. 

It seems like the computer is fine without this wire, but it just seemed possible that someone out there might recognize this type of behavior. I suppose I could work out how to make a voltage divider for this single wire, but I don't know if that would work.