So this is part II that sort of piggy-backs off my other post here: https://community.microcenter.com/discussion/9209/store-policies-motherboard-purchases/p1?new=1. I purchased a Z590 motherboard along with an i9-10900 last night and trying to get it working has been a hassle. The first time I powered up the system I kept it stripped down, just the motherboard, CPU and a stick of ram that I pulled from a working system (kept XMP off), to make sure the system could get to the UEFI screen, and it did. Things got weird after this, I plugged in my NVME drive (port 2_3) and powered the system up again, it went into a boot loop and after about four attempts a B&W screen came up saying the motherboard was in safe-mode and recommended I restore factory settings, which I did. Next, it was a struggle to get Windows 10 to install, I lost count of the attempts, with enough persistence I eventually got it installed with many random crashes along the way. Post Windows install and the machine still randomly crashes, in unrepeatable ways.
Monitoring temperatures and voltages in the UEFI interface and in Windows shows that everything is in check, I’m not seeing any thermal throttling. I still have power limits in place for the 10900, so I wouldn’t expect power to be an issue. The system is being powered with a 750W Antec supply that I have been using for years with a prior build. The RAM and NVME drive also came from a working setup that I’ve been using for the past year. The single stick of ram is in stalled in the recommended slot (don't recall the name, but grey channel closest to the CPU) .
I've tried taking the CPU out, inspecting the socket again (looks good from what I can tell) and re-seating the RAM. The Windows event log isn't showing me anything useful, there are no blue screens so nothing to really log, it's aways a sudden loss of power. I'm going to try installing Windows again on another SATA SSD to see if I still experiencing crashes during the install, other than that I kind of feel like I’m out of options, and not sure what else to try. Does anyone have and ideas on what I could try?
System components at the moment:
RAM (single stick): https://www.microcenter.com/product/618847/crucial-ballistix-gaming-16gb-(2-x-8gb)-ddr4-3200-pc4-25600-cl16-dual-channel-desktop-memory-kit-bl2k8g32c16u4b-black
NVME SSD: https://www.microcenter.com/product/625871/crucial-p5-1tb-m2-nvme-interface-pcie-30-x4-internal-solid-state-drive-with-3d-tlc-nand-up-to-3400mb-s-(ct1000p5ssd8)
PSU (older generation of this): https://www.antec.com/product/power/hcg750-bronze
Greetings. Were you receiving error messages during the Windows install? As you put in your post, I'd try a clean install of Windows again and on another drive is a great idea to see if the issues persist.
I only received one error message during the install, and it seems to be because a random restart corrupted the installation process, a photo of the error is below. A majority of the random reboots happened during the copying phase of the installation, where the machine would reboot back into the installer because a bootloader was never getting installed on the NVME drive. I'm going to work with the machine this evening, and I'll post back if I find anything.
Sure! please let us know what occurs.
So installing Windows on a traditional SATA SSD drive worked well. I was able to get Windows installed as well as a few applications that allowed me to stress test the machine, without experiencing any crashing. I also experimented with swapping out the power supply with a new one while still utilizing the NVME drive, and the machine has been stable after a couple hours of use. I'm wondering if the NVME drive is more sensitive to to power quality compared to the SATA drive. I guess this is something else to look into a little further.
It's possible there's something wrong with the NVMe drive as well. You could install it back in to your PC for storage and test it out with some files that you have backups of and see if it causes any issues / works properly or not.
That's not a bad idea, I'll see what I can get done this weekend.
Regarding the power supply issue, I managed to acquire a 24-pin breakout for the power supply and took a few measurements of the 3V3 line, there definitely seems to be something going on with it.
Output with a 2.8w load:
The output doesn't look that great before a higher load is applied, but gets a bit worse. According to a few review sites, the P5 can draw around 4W under load.
Output with a 4.9w load:
I also noticed that the power supply output will get stuck in the state where the waveform looks like the above image when the load is later reduced.
I'm not an EE, but from the readings about the ATX spec that I've read it looks like this 3V3 line is a bit out of spec in both cases. The 200 mvpp in the first image may of partially been the result of a bad setup on my part (ground spring + scope probe directly to the 3V3 line & GND), but the waveform in the second image I think speaks for itself.
I also connected the EVGA 650GM that seems to keep the system stable to the breakout board and applied the same load conditions, and saw consistent voltage output behavior across a range of loads, though there were some weird periodic spikes of noise at about a 1% duty cycle (@ ~68KHz).
Certainly looks like a bad 3.3V rail based on what you're seeing with your oscilloscope. I'd swap in the EVGA 50GM, do a fresh install and verify stability. If it's stable, 3.3V is the rail powering your M.2 drive. That would confirm the diagnosis.
Yup, looks like it was the 3V3 rail on the old power supply causing the problems. With the new supply installed I ran the PC for several hours at a time over the past few days while downloading a bunch of steam games to keep the drive busy, and it's been stable.
Good to hear that it is running stable, please let us know if you have any other questions!
Submit photos and a description of your PC to our build showcase
See other custom PC builds and get some ideas for what can be done
Services starting at $149.99