F5F Stay Refreshed Power Users Overclocking Second ASUS motherboard CPU power connector failure incident

Second ASUS motherboard CPU power connector failure incident

Second ASUS motherboard CPU power connector failure incident

R
RulwenJr
Posting Freak
Find
786
11-20-2016, 10:36 AM
#1
The initial model was not intended for extreme overclocking, even though it came with ASUS' superior BIOS and loadline tuning along with several temperature monitoring sensors such as CPU, NB, SB, and another. The second unit swapped in after its 4-pin CPU power connector became damaged and fused onto the motherboard. Uncertain whether the issue stemmed from the power supply or the motherboard, I opted for a replacement part and filed an RMA with ASUS. This second model features an 8-pin CPU power connector and is built for overclocking purposes.

Meltdown occurred when the rig was equipped with my FX6300 processor and 8GB DDR3 RAM, paired with a 500W Coolmax PSU and ample cooling solutions. I maintained temperatures below 60°C Celsius. With a water cooler and good airflow, I successfully pushed the CPU to 4.2GHz from its stock 3.5GHz. The machine ran on my FX6300 alongside Steam Home Streaming for games. One incident saw the four-pin connector melt into the motherboard.

The second unit still supports an overclocked FX8350 processor, which reached speeds up to 4.4GHz (reaching 4.55 occasionally). I used Prime95 for extended sessions. It also ran Vengeance LP DDR3 at 1.8GHz with specific voltage and memory settings. The system included a M.2 SSD, SATA III SSD, and two SATA II drives in RAID 0. The CPU cooled well, staying around 12°C, rarely exceeding 55°C. The GPU was an EVGA 1070SC, paired with Precision XOC for Pascal, using the latest drivers. Maximum stable overclock reached 112%, with temperature targets set to 92°C and prioritized values, while voltage and memory offsets were finely tuned. The PSU used was an EVGA 600W unit.

Last night, the eight-pin CPU connector suffered damage when melted around three of its pins. After cleaning, it functioned again, but I reverted to stock settings, enabled Turbo mode, and everything worked. Still puzzled about why these pins keep failing despite maintaining reasonable sensor temperatures. Any suggestions would be greatly appreciated. At this stage, a fan aimed at the connector is in place just in case.

Please note: #2's loadline calibration was set to extreme or ultra levels, or even slightly below the highest preset. I capped CPU temperature limits at 130 or 140°C, anticipating they wouldn’t hit with my cooling setup. The strategy mirrors GPU temperature management—higher load means higher voltage and power demand, which is controlled by the loadline calibration. I doubt anyone considered adding thermal protection to the power connector.

Your feedback, questions, or concerns are highly valued.
Reply
Reply
R
RulwenJr
11-20-2016, 10:36 AM #1

The initial model was not intended for extreme overclocking, even though it came with ASUS' superior BIOS and loadline tuning along with several temperature monitoring sensors such as CPU, NB, SB, and another. The second unit swapped in after its 4-pin CPU power connector became damaged and fused onto the motherboard. Uncertain whether the issue stemmed from the power supply or the motherboard, I opted for a replacement part and filed an RMA with ASUS. This second model features an 8-pin CPU power connector and is built for overclocking purposes.

Meltdown occurred when the rig was equipped with my FX6300 processor and 8GB DDR3 RAM, paired with a 500W Coolmax PSU and ample cooling solutions. I maintained temperatures below 60°C Celsius. With a water cooler and good airflow, I successfully pushed the CPU to 4.2GHz from its stock 3.5GHz. The machine ran on my FX6300 alongside Steam Home Streaming for games. One incident saw the four-pin connector melt into the motherboard.

The second unit still supports an overclocked FX8350 processor, which reached speeds up to 4.4GHz (reaching 4.55 occasionally). I used Prime95 for extended sessions. It also ran Vengeance LP DDR3 at 1.8GHz with specific voltage and memory settings. The system included a M.2 SSD, SATA III SSD, and two SATA II drives in RAID 0. The CPU cooled well, staying around 12°C, rarely exceeding 55°C. The GPU was an EVGA 1070SC, paired with Precision XOC for Pascal, using the latest drivers. Maximum stable overclock reached 112%, with temperature targets set to 92°C and prioritized values, while voltage and memory offsets were finely tuned. The PSU used was an EVGA 600W unit.

Last night, the eight-pin CPU connector suffered damage when melted around three of its pins. After cleaning, it functioned again, but I reverted to stock settings, enabled Turbo mode, and everything worked. Still puzzled about why these pins keep failing despite maintaining reasonable sensor temperatures. Any suggestions would be greatly appreciated. At this stage, a fan aimed at the connector is in place just in case.

Please note: #2's loadline calibration was set to extreme or ultra levels, or even slightly below the highest preset. I capped CPU temperature limits at 130 or 140°C, anticipating they wouldn’t hit with my cooling setup. The strategy mirrors GPU temperature management—higher load means higher voltage and power demand, which is controlled by the loadline calibration. I doubt anyone considered adding thermal protection to the power connector.

Your feedback, questions, or concerns are highly valued.

Reply
Reply
S
ScrubyShawn
Member
Find
68
11-28-2016, 02:01 AM
#2
Considering the current situation, the EVGA 600W 80 Bronze PSU seems insufficient. A higher wattage of 750+ would be more suitable.
Reply
Reply
S
ScrubyShawn
11-28-2016, 02:01 AM #2

Considering the current situation, the EVGA 600W 80 Bronze PSU seems insufficient. A higher wattage of 750+ would be more suitable.

Reply
Reply