VBAT is powering VCC voltage rail

Hi.

I’m using some supercap to power the VBAT input of the Colibri imx6ull module. I’m using a schottky diode to charge the supercap from the +3.3V power supply. The supercap is then connected to the VBAT (pin 40) of the Colibri module.

1662-colibri-vbat.png

The problem is that if I switch of the power supply, the +3.3V power rail is somehow still powered from the backup battery.

  • Is there any hardware problem on the Colibri module?
  • Or is it because of powering the SNVS rail and passively supplying the +3.3V rail from some SNVS IO pins I’m using?

Thanks,
Andreas

The VCC_BATT input of the module (pin 40) powers the SNVS rail of the SoC. Beside the RTC and
power management function, the i.MX 6ULL SoC uses this rail also as IO rail for the pins in the
SNVS block (SNVS pins). Some of these interface pins are available on the SODIMM connector as
regular GPIOs (SODIMM pin 43, 45, 93, 95, 105, 107, 127, 131, 137, and 138). This means the pins remain powered in the “RTC Only” power mode by the VCC_BATT rail. Make sure there is no back feeding on this pin.

Be aware that RTC on the module is not designed for ultra-low power consumption. Typical power consumption is around 25 uA.

We went with an external RTC, and the results seem pretty good so far.

https://www5.epsondevice.com/en/products/rtc/rx6110sab.html

With 1.5F supercap, we guarantee 2 weeks backup over temp and lifespan 10 years. In practice, at room temp, we maintained clock backup for about 6 months.

There is a couple other parts out there similar to this one, but not all of them are available in the distribution chains. Some more accurate than others, with or without built in crystal and varying power consumption.

https://ambiqmicro.com/rtc/

https://www.microcrystal.com/en/products/real-time-clock-rtc/rv-2123-c2/

Thanks for your Input.