Orin NX 16GB overcurrent warning and running slowly

Autonomous Machines Jetson & Embedded Systems Jetson Orin NX
,
1

Hello,

We bought several Orin NX 16GB SoMs, and 3 of them exibit a weird behaviour: They run very slowly, both to boot and after boot. Opening a terminal takes >5s with nothing running. We’ve tested multiple combinations:

  • JP 5.1.1 and 5.1.2
  • carrier board from Orin NX Dev Kit, Orin Nano Dev Kit and a custom one
  • Multiple SSDs, which were confirmed to work as expected with other SoMs

dmesg and UART logs don’t show anything that sticks out, but we can provide both of them if they help

The only thing that we’ve identified to be the same across the problematic SoMs and doesn’t happen on the others is that the power GUI shows Current Reading for VDD_IN rail at 30100mA, which definitely isn’t true. Computed Power consumption also shows a nice 500 000 mW.

Is there a way to fix this issue?

Thanks!

3

Do you have any application running?

4

Nothing is running. Flash the device, it boots automatically and no app is started

6

Is this a known issue? Can we do something to fix it?

7

No, it is not a known issue.

Could you share me the result of sudo tegrastats after you boot up?

8

Here: tegrastats.log (10.9 KB)

Excerpt: 11-21-2023 13:46:56 RAM 1543/15523MB (lfb 3283x4MB) SWAP 0/7761MB (cached 0MB) CPU [2%@91,1%@91,1%@89,0%@91,9%@91,5%@91,3%@89,9%@90] EMC_FREQ 0%@2133 GR3D_FREQ 0%@[903,0] VIC_FREQ 729 APE 174 CV0@48.218C CPU@50.656C iwlwifi@43C SOC2@49.093C SOC0@49.5C CV1@49.281C GPU@48.593C tj@51.625C SOC1@51.625C CV2@48.406C VDD_IN 508684mW/508684mW VDD_CPU_GPU_CV 921mW/921mW VDD_SOC 2238mW/2238mW

Power GUI SS:

Screenshot from 2023-11-21 13-47-15
Screenshot from 2023-11-21 13-47-15894×943 73.7 KB

Device was only booted, nothing running

9

Do you have other module to cross check? Looks like hardware problem.

10

We have at least one other module (possibly two) from the same order (of 20 pcs) that exibit the exact same behavior

11

Please also check the power adapter and also the base board.

12

Again, tested on multiple carrier boards, multiple setups, and this happens on these modules regardless of the setup - where other modules work

So I get this is a hardware issue that can’t be solved, so the modules aren’t usable

13

Hi,

Could you share me the dmesg from these problematic situation?

14

Sure,

dmesg.log (76.0 KB)

15

Hi,

Could you share the dmesg from two devkits (different modules + carrier boards) which both have such problem?

And also the serial number of these modules.

Thanks.

16

Shared one of the modules, will do the other one in a bit, as we took the cooling off of it to use on a working module.

By serial number you mean what’s physically written on the SoM, or the ECID/SKU/batch number, etc.?

17

You could also dump the eeprom info directly.

sudo i2cdump -y 0 0x50 for the module
sudo i2cdump -y 0 0x56 for the carrier board

As this is something new, I may need to gather some more info later. Sorry in advance.

1 Like
18

module with the dmesg attached above

i2cdump -y 0 0x50
No size specified (using byte-data access)
     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f    0123456789abcdef
10: 00 01 00 01 36 39 39 2d 31 33 37 36 37 2d 30 30    .?.?699-13767-00
00: 02 00 fe 00 00 00 00 00 00 00 00 ff 00 00 00 00    ?.?.............
20: 30 30 2d 45 53 31 20 41 2e 33 00 00 00 00 00 00    00-ES1 A.3......
30: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
40: b0 48 00 00 05 54 b3 2d b0 48 31 34 31 34 32 32    ?H..?T?-?H141422
50: 32 30 30 30 31 37 33 00 00 00 00 00 00 00 00 00    2000173.........
60: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
70: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
90: 00 00 00 00 00 00 4e 56 43 42 00 ff 4d 31 00 00    ......NVCB..M1..
a0: 00 00 00 00 00 00 00 00 00 00 00 00 05 54 b3 2d    ............?T?-
b0: b0 48 01 00 00 00 00 00 00 00 00 00 00 00 00 00    ?H?.............
c0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
d0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
e0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
f0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 38    ...............8
19

May I also ask your base board eeprom as you said you are testing on devkit?

20

We also need syslog with error case too.

21

Let me ask one critical question. Is your test based on NV devkit or your custom board?
Some logs inside your text file looks not a NV devkit behavior.

22

Again, tested on 3 carrier boards and the behavior is the same. If a certain setup helps you in debugging the issue, we can set it up