Sudden shut down, and yellow light ASUs b650e-I, 7800x3d

[zik9200]

Hi everyone, I’ve run into a problem.

PC SPECS:

• Case: Asus ROG Z11 Mini-ITX
• Motherboard: Asus ROG STRIX B650E-I
• CPU: Ryzen 7 7800X3D
• Cooling System: Asus ROG Ryuo 240
• GPU: Asus ROG Radeon RX 6900 XT STRIX LC
• RAM: Corsair 96 GB (2x48GB) DDR5 6000 MHz Vengeance RGB (The RAM is compatible with the motherboard)
• PSU: Asus ROG STRIX 1000G
• SSD 1: Samsung 970 EVO PLUS 2 TB
• SSD 2: Samsung 990 Pro 1 TB
• BIOS: 3287
• Video 1 - https://www.dropbox.com/scl/fi/ow5wa1e9n7l81djerffom/B650-I.mp4?rlkey=dkdc6w2444twjqaybwv1kay1q&st=d...
  Video 2 - https://www.dropbox.com/scl/fi/4dbf1y3f9t6f9yhlqnojc/B650-I-2.mp4?rlkey=9exn0qblsfy2b8308pdn2km7z&st...

Description:
The PC is approximately one year old.

Over the past month and a half (maybe two), unexpected reboots have started occurring without any Kernel-Power 41 (63) errors.

The reboots are completely random - the issue is floating and unpredictable. The system can restart while idle on a cold boot, under heavy load after 4-6 hours, or even in the BIOS, which points more toward a hardware issue, though I don’t rule out AGESA problems either.

What I’ve already diagnosed:

1. Excluded Surge Protectors - connected the PSU directly to the wall outlet, also tested with ECOFLOW DELTA 2. No result.
2. Flashed different BIOS versions, starting from 2214 up to the latest 3287 (hoping one AGESA version might be faulty). No result.
3. Reinstalled Windows. No result.
4. Fully discharged the motherboard/capacitors, removed the CMOS battery, cleared CMOS. No result.
5. Completely disassembled the PC, inspected all connectors, reconnected every component.
6. Removed both SSDs and stayed in BIOS to see if I could catch a reboot - no luck, reboots still happened. SSDs excluded.
7. Disconnected everything possible: USB, Wi-Fi, HDMI, RGB lighting (to eliminate the 5V line). No result.
8. Tested with one RAM stick at a time in both slots. Both modules were tested individually in both slots. No result. I ruled RAM out (though I don’t have another kit to verify).
9. When reboots didn’t occur, I ran OCCT on all possible test settings - no errors, no voltage drops detected.
10. Tried logging voltages via HWiNFO with very short logging intervals hoping to capture any drop - logs were clean. No result.
11. Excluded temperatures as a cause. The build runs extremely cool: CPU and GPU idle at 27–30°C, under load CPU stays under 65–75°C, GPU is even cooler.
12. The GPU has two BIOS options. Asus software doesn’t allow flashing them; only through Linux. I switched between both BIOS modes - no result.
13. Tried various BIOS settings: enabled/disabled MRC&PDE, Secure Boot, BAR Support, adjusted SOC voltage (~1.20V), even disabled Armoury Crate in BIOS to prevent NVRAM interference - too many to list. No result.

The PC behaves inconsistently: sometimes it runs fine for half a day, other times it reboots every 5–10 minutes. Occasionally it goes into a reboot loop, restarting right after POST, or even during POST.

Originally, the system was relatively stable. I didn’t enable EXPO, even though the manufacturer states my RAM is 6000 MHz capable - I kept it at 4800 MHz since enabling EXPO sometimes caused reboots.

P.S.
Memory training can take a long time - it can hang on yellow LED for 5+ minutes! This puts a heavy load on the IMC. It can be disabled, but I keep it on for testing.

Power sequence:
+5VSB → EC/SIO → Power → PS_ON# → PSU ON → 12V/5V/3.3V → VRM → PWR_OK → CPU → BIOS (SPI Flash) → RAM training → GPU/chipset → Boot → OS

Personally, I suspect the motherboard first, and PSU second.

In my opinion, what’s left to check (basically everything):

1. Power Supply Unit (PSU) - since Kernel-Power 41 (63) often points to 12V line issues. Could be the cables too (modular PSU, tight bends in Mini-ITX case).
2. Motherboard.
3. CPU - unlikely, but possible.
4. Memory - probably fine, but still worth checking.
5. GPU - don’t believe it’s the cause.

For me, the two main suspects are the motherboard or the power supply.

That’s all I could recall about the main diagnostics.
At this point, only the method of elimination remains — unfortunately, I don’t have spare components for cross-testing.

There are no proper service centers for computer diagnostics in my city — well, technically there are, but they’re all at the level of just upgrading a laptop :-)))

So I have to figure things out on my own.

I’ve gone through a lot of Reddit posts and saw tons - an unreal number - of problems with this motherboard & am5 platform, and almost no one had a solution.

I found a thread where someone with the exact same motherboard and processor as me is experiencing the identical issue.
https://www.reddit.com/r/AMDHelp/comments/1kdtcr8/sudden_shut_down_and_yellow_light_asus_b650ei/

But no solution is described there yet. I messaged the person to ask what resolved it and am waiting for a reply.

Does anyone have suggestions what I can try?

I’m thinking about buying some cheap RAM and PSU to rule those out. Maybe someone had a similar issue - what did you do?

Or just lose my patience and switch to Intel...

[Nate152]

Hello zik9200

Have you tried updating the motherboard bios with the bios flashback feature?

You'll need a usb flash drive and access to the ASUS Support page, the latest bios for your motherboard is 3287 - ROG STRIX B650E-I GAMING WIFI | Motherboards | ROG United States

If you'd like help updating the bios with the bios flashback feature, just say so and I'll list steps how to do it.

[zik9200]

Thanks for the reply - that’s exactly how I’ve been flashing the BIOS, since I keep getting reboots… and that’s the safest method.

I also checked the DisplayPort cable - I thought maybe it was carrying some charge, but even without it, the reboots are the same.

I forgot to mention that I also ran MemTest86, and there were no errors - which confirms the RAM is fine and indirectly suggests the CPU is fine too.

I’m increasingly leaning toward the motherboard being the culprit.

I haven’t received a reply from the person who had the same issue, but I found his earlier messages where he mentioned that he replaced both the motherboard and the processor right away.

[Nate152]

Software, such as corrupted system files, drivers or malware can cause rebooting.

Try running the system file checker.

Open Command Prompt as an administrator.

Type in sfc /scannow  (put a space between c and forward slash).

Hit Enter, it will do a scan, when finished it will tell you if there were any corrupted files and if they were repaired. 

I had some corrupted files and it repaired them. 

[zik9200]

Sorry, but did you even read what I wrote? I reinstalled Windows, and I’m getting reboots even without the SSD (without the system), while in the BIOS - did you watch the video I attached?

I can run a scan, but that’s just a standard check that won’t really help… I’m getting the Kernel-Power 41 (63) error with no description, which means the power is cutting off. From what I’ve read, drivers can cause reboots — especially network card drivers - and updating them has actually helped some people. But reboots at the POST level can only be caused by hardware issues or a problem with AGESA/BIOS. 

[Silent_Scone]

Hi @zik9200 

Inspect the PSU rails in HWiNFO for sag. Spontaneous reboots at idle are often attributed to power. 

[zik9200]

Alright, I understand. I’ll write back once I find a solution - maybe it’ll help someone else in the future. Thanks.

[Raknor]

Hello

I think I have the same issue.

Setup:

• Motherboard; Asus ROG B650e-i
• CPU: AMD 7800x3d
• Cooling: Corsair iCUE H115i Elite Capellix
• GPU: Gigabyte 3070Ti
• RAM: Corsair 32 GB (2x16GB) DDR5 6000 MHz Vengeance
• PSU; Seasonic Focus SPX 80 650W
• 1 SSD M2 (samsung, forgot the model), 2 SSD SATA (mx500)

The AIO and the SATAs are powered on the PSU.

At install (last year), I had:

• Weird issue with usb dongles (it was like, when I plug various dongles for razer devices, at some points, it looses power and keyboard/mouse were erratic) => had to put the dongle at front, not working at back
• Asus armoury crate making the system unstable => no install of armoury crate, manual install of all drivers

Since a week, I have random sudden hard reboots. No relevant logs in Event viewer so I also suspect hardward issue.

Currently:

• I disassembled the computer, to make sure all cables are well connected.
• I am running without GPU.
• Put a spare PSU (still Seasonic, ATX, 650W)

Last night, I plugged a smartphone on the MB usb back port (computer on). It was fully charged this morning.

I connected another smartphone to be charged as well. A few second, it looked like the phone warned the charging device was faulty. And a mere second, another hard reboot (didnt have it during the night).

So now I am thinking there's a defect in the power management of the B650e-i. The defect seems random, like a faulty capacitor somewhere.

Already contacter the seller for the PSU, but now, thinking it's the MB (luckyly, same seller)

[zik9200]

Wow, that’s yet another confirmation. I’m now 90% sure the issue is with the motherboard. I’ll tinker around for a few more days, but I’m thinking of getting the newer B850-I model. It has a more powerful VRM phase design, and there are tons of Reddit posts about problems with the B650e-i... while there’s complete silence about the B850 — probably because it’s still new. The B650 worked fine for a year, so maybe it has some kind of power delivery defect, poor soldering, or something that will start failing over time for many users.

[Silent_Scone]

Honestly, check this HWiNFO too, rather than making assumptions. Although there will be a deviation from the real time voltage, there may be enough sag on the line that it will indicate an issue.