I built up a NanoPC-T4 with a 240GB NVMe.   I am using the supplied heat sink and the black metal case that FriendlyArm sells.   While I was copying an image over I got a board failure.  A reset and power cycle didn't work.

The stat light no longer shows activity.   Just power.. that is all.  

Amazon was great and replaced the board.  I am in the process of backing up my NVMe.  

After the backup I will run an OS install and restore the image.

 

Has anyone else ran into this ?  The black metal case isn't hot enough to burn my hands.. it's just very warm to the touch,.

 

Welcome to ARMBIAN 5.67 user-built Ubuntu 18.04.1 LTS 4.4.166-rk3399   
System load:   0.95 2.41 2.69      Up time:       1:00 hour        
Memory usage:  13 % of 3810MB     Zram usage:    20 % of 1023Mb     IP:            192.168.0.135
CPU temp:      80°C               
Usage of /:    18% of 15G        storage/:      4% of 7.3T       

31 minutes ago, jerryn said:

Has anyone else ran into this ?

This would be the first time I heard of such failure. Install rpimonitor (armbianmonior -r) and see what's going on with the temperature while on load.

1 hour ago, jerryn said:

built up a NanoPC-T4 with a 240GB NVMe.   I am using the supplied heat sink and the black metal case that FriendlyArm sells.   While I was copying an image over I got a board failure.  A reset and power cycle didn't work.

The stat light no longer shows activity.   Just power.. that is all.  

Amazon was great and replaced the board.  I am in the process of backing up my NVMe.  

After the backup I will run an OS install and restore the image.

 

First thing that comes to mind is the quality of the power supply... are you using FE's power supply (12v/2a) and their cable? 

 

(FE does supply decent cables and wall warts/psu's for many, if not most, of their boards, and it's a minimal cost)

 

Rockchip SoC's can be very power sensitive as their relatively high performance - anybody that's played with a Tinkerboard, which is an older SoC knows this... and NVME is going to push things even harder...

 

I do find it hard that the board would outright die in your use case - perhaps corrupt the SSD and/or eMMC, but that's usually recoverable...

I am using the power supply that came with the FriendlyArm kit.  

 

I reset and reinstalled Armbian , android8, android7, and friendly desktop multiple times before the failure.

 

Maskedrom and recovery were no longer working. The status led was off.

 

My restore to the replacement board should be done soon.  Hopefully there's some info in the logs.

 

I need more info to prove it but I think it was thermal failure.   How hot can the boards run before the SOC is damaged ? 90C ?

9 hours ago, jerryn said:

I need more info to prove it but I think it was thermal failure.   How hot can the boards run before the SOC is damaged ? 90C ?

SOC should be protected with throttling if we assume it works properly. At least on legacy kernel, while for mainline I won't put my hands in the fire. It's still under development and regressions are possible.

It could also be something else, some dump component, PMIC. If you get time and will ... 

On 12/30/2018 at 6:14 PM, jerryn said:

[...]

I need more info to prove it but I think it was thermal failure.   How hot can the boards run before the SOC is damaged ? 90C ?

RK3399: Junction: 125C; maximum recommended operating temp is 80C for 2.0/1.6GHz (for the K version, but hey). The board has a thermal protection circuit based on the RK3399's over-temp signal (assuming the TSADC interrupt is not set too high or disabled), so it should just die and return to life after reset. That isn't to say that the thing isn't a burning weenie roaster, but surprisingly (to me, at least) the only really hot running component is the SOC itself. The RK808 and various regulators don't seem to get too hot*. Looking at the little enclosure... looks like a nice fit, but a bit tight. In open air (no movement/flow) I get +50C under load with a large (for the board) heat pipe and fin stack, so I'd expect throttling or shutdown with the small heat sink inside the wee box. I'd also expect a reduced component life, but not a hard failure. Probably just bad luck. Never can tell, though - I've seen a few failure modes I didn't expect.

*Given my setup, where the giant (er, relatively speaking) heat sink on the SOC reduces feedback. I can get maximum sustained performance, but at the expense of additional time, money, and space.