Discussion about 4090 and 12VHPWR voltage issues
Discussion about 4090 and 12VHPWR voltage issues
I get it, lower voltage often happens when FPS drops, since GPU coil whine increases with frame rate.
Sure, but performance drops, which affects how the GPU operates no matter how much it's used
The answers to this question are perplexing. It seems I might be misinterpreting the issue, or perhaps others have different views. Could reduced voltage lead to the connector melting? Yes. Electrical power (W) equals voltage (V) multiplied by current (I). When the voltage drops, the current tends to rise. It’s the high current that typically results in overheating of cables. Would a voltage of 11.843V on the 12V rail cause the 12VHPWR connector to melt? Unlikely. That voltage remains within acceptable limits. With a 4090 at 450W, the current would rise from 37.5A at standard voltage (12V) to 38A at 11.843V. An additional 0.5A on the connector shouldn’t pose a significant problem. The 12VHPWR connector is designed for 9.5A per pin, with six pins totaling a maximum current rating of 57A. (Note: The graphics card can draw up to 75W from the PCIe slot, so in reality it would likely only consume 375-400W from the 12VHPWR cable—though some users overclock and push power limits, reaching 450W isn’t entirely surprising.) It’s believed that melting of 12VHPWR cables usually stems from inadequate contact between connectors or poor terminal connections. In summary, software readings aren’t reliable for voltage measurements. The “GPU Rail Power” display you see isn’t a direct voltage reading; it aggregates data from multiple sensors. The lowest and highest readings across these sensors can vary widely. When you view the detailed GPU Rail Voltage in HWinfo64, you’ll notice the minimum is around 11.864V and the maximum is near 12.2V. This wide range explains the apparent fluctuation. Your power supply appears to be functioning properly, and it’s a high-quality unit.
I really needed that answer, I was expecting more coil whine lately, but also playing harder and noticing the numbers helped me connect the dots. I wasn’t sure how to get the input voltages, but as you mentioned they actually reveal a smaller difference between the lowest and highest values—like about 0.2 instead of thinking it was a bigger gap. Even though I know it’s not perfect, it gives me some reassurance.
Also, your explanation made it clearer. This is exactly what I was looking for—understanding the details from the start.
Lastly, this question might depend on individual situations. When should you be concerned? For me, my lowest reading was around 11.77, so I’m just trying to gauge when things might get serious. If I start seeing blue screens or instability, it’s probably time to pay attention.
The price of the part and repair costs are high, so it’s tough to decide whether to wait or fix it. Or maybe just accept it and move forward.
Thanks a lot!
Coil whine can be challenging to manage. I once experienced significant coil whine with a graphics card during Crysis and another game’s main menu, but it disappeared in other titles. It makes sense that changing the game or adjusting resolution triggered the issue. Since coil whine appears only under specific conditions, you might ease it by tweaking graphical settings such as FPS limits or anti-aliasing. You could also experiment with core clock speed or power targets in software like MSI Afterburner. Coil whine itself isn’t harmful to the system—it’s just an annoying sound. If you’re concerned about the 12VHPWR cable, ensure it’s securely connected and the connector isn’t being stressed. If you notice blue screens or crashes, it may not be a GPU problem; those can stem from other causes. If you encounter issues, share your details in the troubleshooting section for further guidance. Overall, it’s not a major concern and you’re unlikely to need much action.