WHEA Logger 18

HR2022

Member
Hi there,
because my PC keeps suddenly shutting down (without a bluescreen), I have installed HWinfo. The WHEA logger 18 error is displayed in the event viewer.
Is it possible to isolate the problem with HWInfo and what information is needed?

Thank you very much.
 
HWiNFO should tell you some more details about the error when it appears and also the Event Viewer.
 
Hi Martin, yes, I know. In the Event viewer appears the WHEA 18 error. But how can HWiNFo help me to fix the problem?
 
It will only tell you some more details where it occurred (CPU, PCIe, etc.) but won't fix it.
 
Oh, I know that HWiNFO can`t fix the problem. But the WHEA Logger error can be the power supply, the RAM, the CPU or the mainboard. Can I narrow down the problem with HWiNFO?
Can someone help me to do that and which informations do they need?
 
Something I noticed: the power reporting deviation shows in the minimal a red 88,1% (in idle)
Can you tell me the reason for that?
 
Hi there,
because my PC keeps suddenly shutting down (without a bluescreen), I have installed HWinfo. The WHEA logger 18 error is displayed in the event viewer.
Is it possible to isolate the problem with HWInfo and what information is needed?

Thank you very much.
If you click it on event viewer it will show you the error type
Like:
Untitled_15.png

Usually these types of errors comes from OCing system memory sub-components or using too much (Vcore) under-volt on the CPU (like negative values on AMD's Curve Optimizer)

Whats the specs of your system and what have you (if) done with it so far, settings-wise?
 
Hi zach,

be quiet! PURE BASE 600
AMD Ryzen 7 3800X
ROG STRIX B450-F Gaming
32 GBytes RAM DDR 4 (2x16 GB)
Nvidia GeForce RTX 2070 Super
Samsung SSD EVO Plus 1TB M2
Seagate 4TB SATA
Seasonic FOCUS PX 650 W Netzteil
WIN 10

What I do:
actual Bios Version 4901
Virus scan
Disable fast start
Current graphics drivers
perfmon /report
sfc /scannow
dism /Online /Cleanup-Image /ScanHealth
Mem-Test

The event viewer shows the same like your picture except Prozessor-APIC-ID: 0

PS: In this Moment I see some errors in HWiNFO: CPU-Cache-L3 Error
 
Last edited:
Is this a new build?
How long do you have it running on same specs/settings?
Have you ever overclock the system (CPU or RAM) ?

Can we see a screenshot of HWiNFO sensors window, like this below?

CBR23_Single_04.png

But it has to be like this, so all these are visible...
 
Hi, here is the info:
I see, this shouldn't create errors or a shut down, unless some incompatibility is in play between CPU/RAM or Board/Ram, or the CPU has gone wrong.
You haven't answer a few questions, Like did you have the system long? Did it always crash and restart?

And some new ones:
Have you ever enable XMP/DOCP profile on RAM?
It seems its off right now and memory (MCLK) is out of sync with the CPU bus interconnect (FCLK)

Whats the rated speed of your RAM? Right now is running on 2133MHz (1067MHz real) as the memory controller as well and the interconnect between the CPU and memory controller at 1200MHz

See the red boxes in your screenshot.

Untitled_17.png

You can also see the same with ZenTimings memory info app

Untitled_18.png
 
Hi, I have the system for 2,5 years. After 1,5 years the crashes begin. I can show my bios and CPU-Z infos, too.
Before the new bios update I have the DOCP profile loaded, but also with crashes.
 

Attachments

  • CPUZ_4.jpg
    CPUZ_4.jpg
    88.3 KB · Views: 8
  • CPUZ_3.jpg
    CPUZ_3.jpg
    68.9 KB · Views: 6
  • CPUZ_2.jpg
    CPUZ_2.jpg
    82 KB · Views: 6
  • CPUZ_1.jpg
    CPUZ_1.jpg
    106.5 KB · Views: 6
  • BIOS2.jpg
    BIOS2.jpg
    504.8 KB · Views: 9
  • BIOS1.jpg
    BIOS1.jpg
    596.9 KB · Views: 8
Im not very familiar with ASUS BIOS settings on AM4 platform, so Im just trying to remember what I can from others with issues like yours.

Can you try to disable that "PBO Fmax Enhancer" And try to find FCLK setting and set it the same with RAM and memory controller? It should be 1067MHz

The last 3 years I saw many cases like yours. I'm also member of TPU forums

You can search there, you'll find similar cases like yours
For all I know the last thing to do is to try to set memory settings on manual and not let it on auto.
After this you may want to try to change a different memory kit. But before you spend any money you can try to ask there (make a new thread) for opinions and experience of others while you display your issue and your system in details.

Is your RAM Corsair and what is the rated speed on DOCP?
Also, is it on the board's QVL? As... is it officially compatible the RAM/Board combo?
 
Thanks for all the hints. One more question: The CPU L3-Cache error displayed in HWiNFO (more than one error), is it a defect from the CPU, so that I have to change the CPU?
 
It could be, but also these errors may appear if there is some incompatibility between the board and RAM/CPU. It’s the combo of all 3 that some times can cause this.
Or could disappear if you change some settings.

These require some trouble shooting for maybe a few days.
That’s why I was suggesting to disable Fmax Enhancer and set MEMCLK/UCLK/FCLK on the same (synced) speed. And there is a lot more (settings in BIOS) to be tried on to try to fix this.

And in the end if nothing can make it go away then next steps is to try different RAM on the same board/CPU or different CPU on the same board/RAM.
This is the way to pinpoint the culprit.
 
Hi zach, I disabled the Fmax Enhancer and set MEMCLK/UCLK/FCLK on same speed. Alway shut downs. Now the bios is reseted. Of course the crashes are there.
I bought new RAM Crucial Ballistix BL2K16G26C16U4B 2666 MHz, DDR4*. It should be compatible with my system. But when I installed the new RAM, the PC won´t start and the white Q-LED is on. I still despair.
 
Since the bios update, there are more crashes. Now I see, that the CPU Core Voltage is always over 1.4 Volt. Before the bios update it was by 1,14. How can I decrease that?
 
What is your windows power plan? Its best to use the "balanced power plan" after you got the latest chipset drivers for your board from AMD directly.


Voltage on a Ryzen3000 can be anywhere between 0.9V and 1.5V depending the CPU load
Use the "CPU Core Voltage SVI2 TFN" on HWiNFO sensors to monitor CPU core voltage

This one:
Untitled_19.png
As you can see on your screenshot you also got before up to 1.48V

Most other readings and software are more inaccurate about core voltage.

And its best to also enable the "Snapshot CPU Polling" option on HWiNFO main settings (right click system tray icon)

Untitled_20.png

Did you also try DOCP with the new RAM? You install them on slots 2 and 4, right? (A2 and B2 slots)
Did you pay almost 150€ for that RAM? Is it on board's QVL?
Its so low on speed (2666) and only 2x8GB. You could've bought at least a 3200 kit
 
Hi zach, so much work. Thank you. Because I have try so many things and there are almost the crashes, I ordered a new CPU. Lets see what happens.
 
Whea errors on Ryzen are usually memory controller / IF errors. I have 3800x myself and I noticed whea errors returning to me during hot summer days. Ryzen is very temperature sensitive as well. I was pushing memory controller quite a bit so I was not too surprised to see errors return and easy fix was to drop clock a little bit. Could have also been caused by my move to AMD GPU from Nvidia's GPU as time frame matched so well.

That aside the RAM could have higher clock and that would force your CPU to run 1:1:1 ratio, forcing memory controller / IF to run higher than it is capable of running which causes said errors. Doesn't seem to be the issue in this case though.
 
Back
Top