ASUS PRO WS WRX80E SAGE SE WIFI won’t power on after shutdown (random, PSU reset required)

Hardware Hub Build a PC
1

Hi everyone,

I recently built a WRX80 workstation and I’m running into a strange issue that I haven’t been able to isolate.

Problem:

After a normal shutdown, Suspend (S3) or Hibernate (S4)
the system will sometimes refuse to power on:

  • Case power button → no response

  • Motherboard power button → no response

  • BMC power-on command → fails (no response after 3 tries)

This happens randomly (not Always).

The only way to recover is:

  1. Turn off the PSU

  2. Wait for capacitors to discharge

  3. Turn PSU back on

After that, the system boots normally.

Behavior once powered on:

  • System is completely stable

  • No crashes or instability

  • Passes long stress tests (AIDA64, OCCT)

  • No issues under load

However:

  • On the first boot after PSU reset, the system sometimes power-cycles once or tow during POST but It always eventually completes POST successfully whit AA Q-Code

Diagnostics / current state:

  • BIOS and BMC firmware updated to latest available versions (as of 16/03/2026)

  • Q-Code: AA (normal)

  • No POST issues

  • All LEDs off except for the blinking green BMC LED

  • AC power loss behavior: Restore to last state

What I’ve already tested:

  • Replaced PSU → no change

  • Verified all cabling

  • Tested RAM → no errors

Question:

At this point I’m trying to understand where to focus:

  • Could this be CPU-related at all?

  • Or is this more likely:

    • motherboard power logic / standby rail (5VSB?)

    • BIOS/firmware edge case

    • something related to WRX80 platform quirks

Also:

  • Any reliable way to test the CPU?
    (Motherboard and CPU are out of warranty so no RMA for now)

———————————————————————————————————————————
Build Component;
Motherboard = ASUS Pro WS WRX80E-SAGE SE WIFI
CPU = AMD Ryzen Threadripper PRO 5975WX
GPU = RTX 3080Ti + RTX 5000
RAM = G.SKILL Trident Z RGB - 256 GB (Kit 8 x 32 GB) - DDR4 3200 UDIMM CL14
Cooler = Silverstone XE360PD
Power Supply = Silverstone SST-DA1650-G, 80Plus Gold, modular - 1650W
Storage = 8 x SAMSUNG 990 PRO M.2 2000 GB PCI Express 4.0 V-NAND MLC NVMe

NVMe HBA =

4 x ICY DOCK EXLink MB308A - NVMe PCIe 4.0 x16 to 2 port SlimSAS 8i (SFF-8654) + redriver

PCIe Data Cable =

8 x ICY DOCK MB406L-B - data cable SlimSAS SFF-8654 8i to 1 SlimSAS SFF-8654 8i, 50 cm

SAS/SATA Data Cable =

2 x SilverStone SST-CPS04 - data cable Mini SAS HD SFF8643 to SFF8643, 0.5M

M.2 Heatsink = M.2 SSD Aluminum Heatsink with Silicone Thermal Pad

Backplane =

  • E1.S Backplane for optical drive bay, Hot Swap, x8 Bay E1.S + x8 E1.S → M.2 adapter
  • 8 Bay NVMe/SAS/SATA 2.5 inch, optical drive bay backplane, hot-swappable

Fans =

PHANTEKS D30 PWM Regular Airflow D-RGB, 3 pcs, 120mm - Black (push)

PHANTEKS D30 PWM Reverse Airflow D-RGB, 3 pcs, 120mm - Black (Pull)

Case

Anidees AI Raider XL
Full Tower Tempered Glass XL-ATX/E-ATX/ATX ,
Supports 12 x 5.25” Drive Bays & 480/360 Radiator

2

Welcome!

This sound exactly like what would happen if you didn’t have enough power capacity on the 5VSB rail and the PSU went into protection mode after using too much current. Do you have another power supply that can do more than 3A on the rail? or alternative do you have a powerful fan you can use to force feed air into the PSU while it’s “off” to prevent the tiny 5VSB circuit from overheating?

3

ok, sorry if I’ve been inactive for a while but since it’s an intermittent problem it took a long time to see if each test changed anything.

@twin_savage
unfortunately I don’t have another power supply that can do more than 3A on the 5VSB rail . . . I’m not even sure they exist.

but thanks from one of the tests I did, I excluded the power supply;
very trivially once the problem arose as usual, without turning off the power supply, I disconnected the 24 pin from the mobo and shorted the power pins and the power supply started.

This should’t have happened if it were in protection mode.

Among other things, I also tried a minimal configuration and it didn’t solve the problem.

I know it could be a problem related to Windows not shutting down the PC properly and sending it into a sort of shutdown/hibernation loop? in which the components shut down but the BIOS gets stuck during the shutdown/hibernation phase and obviously since it didn’t turn off it doesn’t start up again either.
The symptoms are extremely similar to mine, in fact, but I don’t think that’s actually my case since when I assembled everything, I did a clean install of Windows twice for FW updates and component configuration needs, and it gave problems both times.

I’ve come to the conclusion that the problem could be either the MOBO or the CPU…
but I was just thinking that if the problem was HW it should always happen and not randomly…
if it were software it would make sense but the BIOS has been updated as well as the BMC and Windows has been clean installed twice…

what am I missing?
Why does it keep happening?
And above all, WHY IS IT RANDOM !?!?!

4

I wanted to follow up since I’m also having this issue. I initially thought it was my EVGA power supply since it had a concurrent issue of randomly browning out after a couple minutes but even after swapping out to a different power supply, this issue has persisted.

My specs:

Threadripper 3995WX

Asus Pro WS WRX80E SAGE SE Wifi

128GB ECC DDR4

5

Hi everyone,

I have been experiencing the same issue.

I purchased my WRX80-SAGE SE WIFI in January 2026. From the beginning, I have been trying to use this motherboard as a desktop/workstation rather than leaving it powered on 24/7 at idle. Unfortunately, after some combinations of rebooting, shutting down, or using sleep/hibernate, the system can enter a strange power state where it completely refuses to power on again. The only way to recover it is to physically switch the PSU off, wait a few seconds, and switch it back on. After that, the system works perfectly again.

After about six months of troubleshooting, testing different hardware, BIOS configurations, and communicating with ASUS support, I have started to suspect that the problem is related to the motherboard/BMC/ACPI power-state handling.

The problem seems to be reproducible mainly around sleep/hibernate and certain power-state transitions. I have also been able to trigger it when using the IPMI web interface to perform power off, power cycle, or hard reset operations.

I have tested and changed almost everything I could think of: HDDs/SSDs, GPU, RAM, BIOS settings, etc. Eventually, I tried disabling the BMC using the physical BMC switch on the motherboard.

The disadvantage of disabling the BMC is that the motherboard no longer controls the fan speeds, so all the fans run at 100%. I solved that by installing an Aqua Computer QUADRO fan controller.

With the BMC disabled, I have been using the system normally for the past few months: powering it on when I need it and shutting it down through Windows, while avoiding sleep/hibernate. So far, I have not experienced the power-state problem during this period.

This makes me suspect that the BMC may be involved somehow. With the BMC enabled, repeatedly rebooting, powering on/off, or using different power-state transitions seems to eventually leave the board stuck in this strange state. With the BMC disabled, the problem has not occurred in my normal desktop/workstation usage.

I also tested another PSU to rule out a possible PSU/motherboard communication issue. My current PSU is a be quiet! Straight Power 13 850W (ATX 3.1), and I tested an older be quiet! Dark Power Pro 12 1200W (ATX 2.5), which is from roughly the same generation as the platform was originally designed/validated around.

I tested both PSUs using their different rail configurations, including the combined/overclocking configuration, and the result was essentially the same: the motherboard could still eventually enter the problematic state.

Interestingly, with the older ATX 2.5 PSU, the IPMI interface did give me a communication-related error when I requested a power-on. With the newer PSU, I previously only saw the BMC attempt the operation several times without successfully powering the system on. I’m not sure how significant this difference is, but I thought it was worth mentioning.

At this point, I have returned to my most stable configuration:

BMC disabled + Aqua Computer QUADRO for fan control.

This works, but obviously I would much rather be able to use all the motherboard’s functionality, including the BMC, without having to leave the workstation running 24/7.

I really hope ASUS can investigate this further and, if the problem is indeed related to BMC/ACPI power-state handling, release a BMC firmware update that fixes it.

I also noticed another situation that can trigger the same behavior. After a power outage/grid interruption, particularly when power is restored and there appears to be a current/power transient, the motherboard has sometimes ended up in the same “stuck” state and refuses to power on again until I completely cycle the PSU’s AC power.

My impression is that the board may be entering some kind of protection or latched power state, although I cannot say for certain what is happening internally.

This has happened even when using a normal UPS. Because of this, I have since upgraded to a sine-wave line-interactive UPS in the hope of providing cleaner and more stable power during outages and power restoration. I have not yet been able to determine whether this completely eliminates the problem, but I thought this detail was worth mentioning because it may provide another clue about what is triggering the issue.

Hopefully this information helps others who are experiencing the same problem, and perhaps ASUS can identify what is happening with the BMC and power-state management and give us a firmware update that fix this.

System specs:

Motherboard: ASUS WRX80-SAGE SE WIFI

CPU: AMD Threadripper Pro 3945WX

RAM: 64 GB ECC RDIMM (2 × 32 GB SK Hynix)

PSU: be quiet! Straight Power 13 850W (ATX 3.1)

Test PSU: be quiet! Dark Power Pro 12 1200W (ATX 2.5)

CPU cooling: be quiet! Silent Loop 3 360mm AIO

Case fans: 3 × 120mm be quiet! Silent Wings 4 HS, front intake/pull configuration

Main drive: 4 TB NVMe, ~7000 MB/s

Backup drive: 1 TB 2.5" HDD

Fan controller: Aqua Computer QUADRO

Additional fan: be quiet! Light Wings HS, rear

GPU: NVIDIA Quadro K600

1 Like
6

Did you already connect the additional two power cables to the extra inputs PCIE_PWR_1 and PCIE_PWR_2 that this mainboard has at the bottom? Does this help you?

I also have a WRX80E and recall the same “need to power cycle before I can boot again” problem. However, I am not sure if these additional power cables really helped here, because I also abandoned S3/suspend-to-RAM completely at the same time. Now I am only using S4 hibernation, almost daily. I no longer need to power cycle my mainboard and did not have this problem since several months (BMC/IPMI is enabled and working).

Good luck! (And if you try it, please report back whether it helped in your case or not.)

1 Like