Burned Asus P8Z68 Pro Gen 3 Power Connector

[heiii3]

Hi everyone,
My computer shut down suddenly and it turned back on with no screen yesterday. Strangely I could not even shut it down by holding the power button for 4 secs. Upon inspecting I noticed that the 24pin power connector was literally stuck on the motherboard with burn mark at the motherboard power socket. Using excessive force I finally removed the connector and noticed the plastic part of two of the pins were completely melted. What should I do at this point? Whose fault was that?

System Spec:
Intel i5 3570k (no oc)
ASUS P8Z68 Pro Gen 3 Bios 3402
16gb Ripjaw PC12800 (2x8g)
Crucial m4 256GB
2x PNY GTX 750ti OC (was mining at that time, no overvolt, slightly overclock, stock bios)
Antec HCG-620w
zalman z7 plus

I have some pictures

Any help would be appreciated. Thanks!

 

[denis280]

is this a new build.the motherboard could be shorted with the case.

 

[rgallant]

that would happen [burn the 12v yellow wires ]on some mb's with power sucking gpu's that took more than 75 watts each out of the pci-e sockets and did not have a 12v molex for the pci-e vid. cards.

-but your system does not seem to have a sata or molex port for the 12v boost and 750ti's are not power sucking so not sure .

- but did you have a 8pin for the cpu or just 4 ?

 

[heiii3]

That's not a new build. The motherboard (P8Z68 Pro Gen3) is designed to support SLI or crossfire, and it has like 12 or 16 phase power for CPU so it has an 8-pin for the CPU. I just wondered what I should do now (I think this took my motherboard and probably the power supply).

 

[RU482]

wtf is 12/16 phase power when your talking about DC voltage?

Unfortunately, I'd be inclined to think that you are at fault, due to the overclock and heavy loading of the video cards and possibly CPU. At stock, those cards are potentially approaching the limits of power that the PCIe slots are capable of providing under heay load. Add in the fact that the motherboard is a couple years old (well, it's a couple year old model anyway) and there is some aging of the components, and you've got some power hungry RAM and SSD.

Pulling this out of my ass, but I'd say you were running the system at the worst case scenario for the motherboard. Should have gone with video cards that had the aux power connectors on them.

 

[coffeejunkee]

Power hungry ssd?

I have no idea what happened, although the cards exceeding the 75W limit sounds somewhat plausible, but it's a limit right? so how can a card pull more than the limit?

 

[heiii3]

wtf is 12/16 phase power when your talking about DC voltage?

Unfortunately, I'd be inclined to think that you are at fault, due to the overclock and heavy loading of the video cards and possibly CPU. At stock, those cards are potentially approaching the limits of power that the PCIe slots are capable of providing under heay load. Add in the fact that the motherboard is a couple years old (well, it's a couple year old model anyway) and there is some aging of the components, and you've got some power hungry RAM and SSD.

Pulling this out of my ass, but I'd say you were running the system at the worst case scenario for the motherboard. Should have gone with video cards that had the aux power connectors on them.

You are saying it is my fault? I guess there are several things that I need to clarify. 16 phase power is from Asus, not from me. Go check the website for this motherboard model and you will know what I am talking about. The CPU is 3570k (77w at most at full load, no overclock) and is definitely not power hungry. I used 2x8gb 1.5v ram at stock and an SSD . How can these be power hungry? The motherboard was designed to handle SLI or crossfire, and the graphics cards were GTX 750 ti at stock bios, meaning that NVidia actually puts a limit in the bios for these cards. The overclock is just slightly and there is a power limit for new gen NVidia graphics cards. When I overclock the graphics card actually reduces the core voltage at higher core speed so as to stay within limit. Check any review for this graphics card and you will find out that even overclocking to the max the card still only uses a little bit more than 60w. So in the end the system was running well within the limit.
By the way, ASUS did not restrict the board to be used only with graphics cards with auxiliary power connectors when using multiple cards. So how can this be my fault?

The power supply used was the Antec HCG-620w (made by Seasonic, >85% efficiency) with a single 48A +12v rail. The whole system load wattage (according to kill-a-watt meter) when mining was under 200W. Unless there is a power surge (which I don't know how it happens) the motherboard probably is at fault.

 

[rgallant]

pull the board out of the case
I'd pull/smell the vrm heat sinks and check for anything burnt but as you say your system is/should not be drawing many watts.
-could come down to loose pins in the socket just heating up over time.

 

[heiii3]

Just an update, I tried to test the power supply with a power supply tester, and surprisingly the power supply is still working. But with two melted pins I doubt that I will use it for another machine. What should I do at this point? Should I complain to ASUS about my issue?

 

[bononos]

Just an update, I tried to test the power supply with a power supply tester, and surprisingly the power supply is still working. But with two melted pins I doubt that I will use it for another machine. What should I do at this point? Should I complain to ASUS about my issue?
....

I think your psu might still be good because the pins are still there, just the plastic bits around the copper leads have melted. You could keep it as a backup.
I don't know why the motherboard/psu blew, it seems more likely to have been a surge.

 

[Z15CAM]

I had a very similar experience when a PCB screw in a NEW PSU was not properly seated and came loose. Every time I would turn the tower on it's side in attempt to diagnose the issues it created the screw rolled around in the PSU shorting out various components.

I discovered a couple of singed terminals in the ATX Connector before looking at the PSU which other words worked perfect at times.

I was fortunate to find the problem before that catastrophic burnout you have.

From that experience I always take the cover off NEW PSU's looking for foreign crap and the PCB screws are tight.

----------------------------------------

i7 2700k/ASUS P8Z68-V Pro Gen3/Corsair H110 AIO running @ 1.032v/1600Mhz to 1.384v at 4800Mhz 24/7 between 32 to 67C, 16GB of Samsung MV-3V4G3D-US DDR3 at 1.34v/1866Mhz 9-9-9-24 1T with 4GB's assigned to a RAMDisk drive to handle Win7 64 slough to negate writes to the SSD, Samsung 840 Pro 256, 2 x's WD5001AALS HDD's in Raid-0, 1 x's WD1002FAEX 1TB, ASUS DRW-24B1ST DVDRW, Hauppauge WinTV HVR-1250 PCIe, 2nd Intel NIC for Dual Networking, XFX 850W Pro Black Black Edition, XSPC RAZOR GA-R9 290X at 1250/1500, Koolance 401x2 RP with 280x60x140 Rad, Nactua 140 Fans, Fractal Design ARC Midi R2, QX2510 Samsung PLS 2560x1440 display at 120Hz.

TOTAL Investment on this Platform including 13% HST Tax Plus Shipping is approx $3,200 spent over a period of a year.

 

[mindless1]

Power hungry ssd?

I have no idea what happened, although the cards exceeding the 75W limit sounds somewhat plausible, but it's a limit right? so how can a card pull more than the limit?

That's a limit as a rule for a system designer not to exceed, while an overclocked video card can easily exceed it. However this limit is more about how the board and PCIe slot are designed, while the limit for the 20+4 pin connector is higher.

That's the thing about system design, it's more than just will a card fit in a slot. If two stock speed 750 OC have 60W+60W TDP then some o'c could push them beyond 144W (two 12V leads at 6A each * 12V = 144W). That is, assuming the video cards do not have a power socket.

It is likely the system, both motherboard and PSU can continue to be used as-is if a couple of changes are made. 1) Power fans from a PSU lead rather than motherboard headers. 2) Use a video card with its own 12V socket.

You may be able to clean the contacts on the board and PSU and use the 12V rail from it again as an alternative to the above, but don't keep mining with a pair of 750 Ti o'c.