Last chance before RMA (B650M-HDV/M.2)

Not seeing any way to edit. But I figured I should summarize my pages of test notes and things I've tried already to resolve the hanging/crashing/instability:

- Lowering RAM speed to default JEDEC 4800 (initially I had noticed eventual failure to clear POST memory test on anything higher than 5200 MHz, and resolved to stick with base speeds until I resolved other instability (which I could not.) My DDR5 kit is not cleared for OC stability with this mobo, but I think it should at least work at 4800 MHz?)
- Going to 1x16G ram stick or the other, either slot
- "Clean boot" disabling unnecessary startup programs
- Updated BIOS to latest as of mid January 2025
- setting CPU to PBO 85C with those defaults
- memtest86 (6+ hours) with zero errors
- successfully ran cinebench 2024, heaven benchmark, OCCT power test 15+ minutes all with normal results and no issues (sometimes the system would be stable for a while, enough for this kind of testing)
- Switching SSD to other M.2 slot
- Removing SSD entirely and running only off of 6TB WD HDD
- Win10 / Win11 multiple installs each
- removing nvidia and AMD drivers in safe mode with DDU and manually installing
- removing my 4070 Ti Super GPU and running just with the R5 7600 iGPU
- installing different chipset drivers
- sfc/scannow (often it would find corruption at least it did with the SSD OS installs - but I don't know if it was SSD or from the crashing or the motherboard)
- running crucial executive SSD utility - it would always clear 1 test but then gave I/O error if I attempted a 2nd test. This would persist unless I rebooted.
- manually used powershell to check the hash checksum of large files on the SSD, no errors found
- Windows memory diagnostic, no errors

- Examined Windows Event Logs. Here are some of the noted errors for several days of testing/crashing & restarting:
(yellow!) 10016 DistributedCOM (80 instances)
(red!) 1796 TPM-WMI Secure Boot (8 instances)
(yellow!) 129 storahci reset (3 instances)
(yellow!) IO operation retry (2 instances)
(red!) 3503 Device Association Service discovery failure (2 instances)
(red X) 41 Kernel-Power "rebooted without shutting down cleanly" (2 instances)
(red!) 6008 previous shutdown unexpected (2 instances)
(red!) 29 Win failed fast startup (2 instances)
(red!) 7043 Windows Update did not shut down properly

If only 1 thing is wrong, it has to be either mobo, PSU or CPU. But if multiple things are wrong, I guess it could be a faulty SSD AND driver conflicts I never resolved... I don't have any other PSU or CPU I can swap in to check those. I'm so exhausted with all the testing I just want to RMA the mobo unless someone has other ideas to try.

As I said, I do not have any other parts to try.

As for PSU, it passed 15+ minutes of OCCT power stress test & ran until I was satisfied and shut it down, and also was up in safe mode over 8 hours and never had any issues in regular safe mode. That also seems to rule out other hardware things. But I noticed once when I booted to safe mode w/ networking, when shutting down it had an error and crashed. That was with a TP link wifi adapter. However, I had the same kinds of crashing problems on wired connection.

I'm currently not worried about XMP/EXPO. I just want to get it working even if it's at base JEDEC 4800 MHz. It never found a single error on memtest86 or on windows memory test. It only fails the POST ram training at higher frequencies. What can I do to pin down if it's still, somehow, the ram?

What seemed like might be crashing out from driver conflicts, can't just be that, since I get errors even on installation of windows. So then I'm thinking what if it's a faulty SSD and ALSO drivers? If it's multiple things wrong it's just so hard to troubleshoot.