“I am working with the Jetson Orin Nano SOM (8GB) and having power-up issues. The SOM sometimes boots correctly, and sometimes fails.” The problem is on our custom board.
After comparing the power sequences between successful and failed boots based on Figure 6-4 (Power Up Sequence with Power Button) from the product design guide, I have identified the following:
Observation:
Successful Boot: All signals follow the timing diagram correctly. POWER_EN goes high, followed by SYS_RESET* asserting high after the POWER_EN goes high.
Failed Boot:VDD_IN and POWER_EN are present and stable, but SYS_RESET* remains pulled LOW indefinitely. Since SYS_RESET* stays low, our Custom Board Supplies never enable, and the system fails to start.
I saw there is a note says before SYS_RESET goes high, SOM will check I/O pins’ voltage. What is the I/O pins exactly? Can you give us some directions on this issue? Many thanks…
All of the input/output pins from the module to the carrier on the various interfaces. This is to avoid backfeeding power through the SoC’s pins before the on-module power up sequence completes.
Does your carrier board try to charge the backup battery? This is not supported as noted in the PMIC_BBAT description:
Note:This pin is input only and rechargeable cells or devices such as Super Caps cannot be supported.
Hi Chris,
This is our RTC schematic. R953 and R954 are removed. We’re using rechargeable CMOS battery. The CMOS battery is 2.86V. The +V3.3A is for RTC charging. The +V3.3A_RTC is for test point, it doesn’t connect to others.
Does our schematic make sense to your design?
Do you have any idea on this issue which can help us to find out the root cause? We’ve measured all the test points to see if we can find what the difference between boot successfully and boot failures. However, we didnn’t find anything…
Finding the reason why the low BBAT causes the 3.3V rail to drop by 1.4V would be the first task. Depending on how that rail powers the rest of the carrier, there may be devices or signals at non-functional levels/states.
I have removed the R952. The RTC battery voltage is no longer with +V3.3A。However, the problem is still there. One time can boot up. The other can’t. I think the +V3.3A won’t impact the RTC battery.
The procedure covers two scenarios: with and without an RTC battery.
There is a blue LED indicator on my carrier board controlled by MOD_SLEEP.
During testing with the RTC battery installed (tested around 10 times), the LED occasionally fails to light up, indicating that MOD_SLEEP remains at 0V.
However, when the RTC battery is removed, the LED consistently turns on (tested at least 50 times).
Issue: Orin Nano Boot Failure During Power Cycling (Linked to RTC Battery)
Setup:
Module: Jetson Orin Nano SOM on a custom carrier board.
Peripherals: M.2 SSD, Debug UART, and a Power Key.
Power: Battery-powered system.
Reproduce Steps:
With RTC Battery installed:
Connect the battery and press the Power Key. (System boots normally).
Perform a Power Cycle: Disconnect the battery, reconnect it, and press the Power Key again.
Result: About 5~10 times, the system intermittently fails to boot. The blue LED (controlled by MOD_SLEEP) remains OFF, and there is no output from the debug UART.
Recovery: The Power Key becomes unresponsive. I must physically remove and reconnect the battery to recover.
Without RTC Battery:
Repeat the same power cycle procedure.
Result:No issues found. Tested over 50 times, and the system consistently boots every time.
I followed this article to use the RTC battery on Orin Nano dev kit.
It doesn’t have same problem as our carrier board.
Can I ask other questions…
Can we change 3.3V to 5V instead?
Our carrier board design has two power mode, one is battery mode, the other is DC mode. The DC mode is no problem with boot-up. The issue is coming from using battery mode only. Do you have any suggestions on how to design the battery part?
Are your battery + [battery management and protection] + [pre-regulator for the carrier VDD (and possibly VDDIN to the Jetson)] placed on the equivalent of the DC jack side of the Q25 FET on page 7 of the carrier reference schematics? The FET acts as a soft start, to limit the inrush current from the power supply (DC or battery).
