My larkbox pro has been acting weird. It just starts shutting off randomly and when I turn it back on will be stuck at bios. And no matter how i get out of bios, it will immediately turn me back to bios instantly. I would have to unplug the device just to boot it normally.

I though i got lucky because my Larkbox was OK for several months of using it. Now im doomed!

It’s sad that this device have very little support from Chuwi. It’s like a scam even.

Anyway, i managed to install a new 1909. But still get the sames symptoms. The device will still randomly shutdown, with led light off. Most of the time, it would just blink and return. When i reboot it from getting shut down, i would boot to bios. And then back to the video. Now im trying to install 20h2. Fingers crossed.

I successfully installed 20H2 and the auto shutoff and screen blinking are still there. I believe all drives are installed properly. All necessary updates also installed. What the hell is wrong with this device.

I believe people already gave up on this device. Hence, the lack of response.

I think the problem is not with the system, maybe your device is overheating or has power problems.
Try writing to

I have a new video trying to confirm that. I installed Open Hardware Monitor to check the temperature. At first boot the device would be at 55C. When running a window of Chrome it would jump to 70C. Several windows up the device will shutdown, or screen will blink.

You can run memtest, this will test the RAM as well as the CPU for overheating.

I tested that and noticed that the device did shutdown. And i also remembered, before all these issues happened, everytime i turn on the device i hear a load fan running. After all the latest windows update from my original OS 1909, as well as the 20h2, the fan no longer run when i turn on the device, it still run but on a very low speed, and stops immediately.

Would you mind attaching a screenshot of your thermal settings on your bios. I would like to try it.

I managed to tinker with mine. I temporarily disabled the thermal settings, and my device reached 80 to 90C with without shutting down… ofcourse i shut it down so not to further ruin the device.

So i just need a GOOD thermal setting. I would really appreciate it.

I have no will to return here for myself.
But this seems to be truly a defective product, I think.
I could not pass this through.

At first, See this (related to LarBox)

Next, I wrote to Christy the problem seems to be serious.

As you see in the previous thread, Christy sent it back to Amazon and got refund immediately.

In a high certainty, LarkBox series has the defect which causes unintended shutdown in Windows10. Defective product.


1 Like

I do not have this device, unfortunately I cannot help with the settings.
But most likely, support will also not be able to provide you with any settings different from the original ones, so you will have to experiment with them, or maybe someone with a working device can help you.
You can also try to replace the device, perhaps you got a faulty one.

1 Like

70 in Celsius degree is NOT overheated condition.
(usually, 5 dgree ~ 105 degree is operational condition on semiconductor junction temperature).

Even if fan starts to rotate, it should be very slow.


1 Like

I have the same error 34. But with no event error 55…

I am new to this event viewer. I have no idea how to diagnose using event viewer.

I dont think the chinese store i got it from will allow me to get a refund. And that will also mean shouldering the expenses.

Currently my solution was to disable disable turbo boost and tinker with thermal settings on bios. In idle im at 55C. And cpu wont reach 70C. But running geekbench compute benchmark makes the screen blink again.

I did that and also disabled turbo boost. But now im stuck at 1.99Ghz.

I would reflash the bios, if that didn’t work, I would replace the thermal interface.
But the best option is to contact the seller and discuss ways to solve the problem.

…device is temp. sensitive due to VLSI and compact packaging … test this theory by directing another fan at the device while running memtest to heat stress unit.