Measuring Ground current does not seem to match datasheet on STM32L476RG (Nucleo-L476RG)?

[SWenn.1]

Hello...

I am using a Nucleo-L476RG dev board along with https://github.com/STMicroelectronics/STM32CubeL4 .  I have built the code for PWR_STOP1_RTC and am trying to measure ground current.  I have a 10 ohm resistor on the ground side of 3.2V battery.  I have removed JP6 and JP5 on the board.  I am using a USB PicoScope.  At the moment I have two issues:

1.  I load the FW onto the target but I don't understand why this FW does NOT seem to run if I don't keep the usb connected to the nucleo?  Why is this?  I would like to disconnect this as it is just a source of noise and I am getting no where near the approx.   4 uA - 20 uA the datasheet specifies.

2.  For some reason the ground current (when 1. above is connected) measures negative?  I can plug into a traditional scope and it is not negative.  I did calibrate probe and when connected to picoscope I did the autozero...Any thoughts here?  

Thanks

 

[Sarra.S]

Hello @SWenn.1, 

Try probing the NRST pin, this could happen if the NRST line is being held low by the ST-Link when the USB is disconnected

2nd issue is probably related to probe calibration since it's not negative using an oscilloscope

[SWenn.1]

@Sarra.S ...Thank you ...that worked with using a pullup to VDD.

[Andrew Neil]

don't forget to mark the solution:

https://community.st.com/t5/community-guidelines/help-others-to-solve-their-issues/ta-p/575256

 

[Andrew Neil]

---

@SWenn.1 wrote:

this FW does NOT seem to run if I don't keep the usb connected to the nucleo?  Why is this?

 

---

In addition to the unpowered ST-Link holding NRST low, as @Sarra.S mentioned, note that the MCU clock source on most* Nucleo boards defaults to the ST-Link - have you taken that into account?

https://community.st.com/t5/stm32-mcus-boards-and-hardware/nucleo-f103rb-run-on-startup/m-p/687751/highlight/true#M19679

 

On the NRST thing:

https://community.st.com/t5/stm32-mcus-boards-and-hardware/external-powering-of-nucleo-l031k6-hsi-able-to-run/m-p/663084/highlight/true#M18423

 

EDIT:

* It seems that the Nucleo-L476RG is an exception:

 

[SWenn.1]

Just a quick follow up for documentation purposes and others....

To get a good clean measurement on ground current using a scope and a ground resistor I had to:

connect a resistor between NRST and VDD as a pullup

I also had to remove the 0 ohm resistor labeled SB12 NRST.  This was critical as somehow the 3V3_ST_LINK net had 2.5V on it even without USB connection and JP6 not connected.  It was feeding back thru somewhere and causing the STLINK chip to have power.

[BarryWhit]

---

@SWenn.1 wrote:

To get a good clean measurement on ground current using a scope and a ground resistor I had to:

connect a resistor between NRST and VDD as a pullup 

---

That's odd, given that it has an internal pull-up. 

If something was indeed holding the NRST low, it would make more sense to stop that from happening, rather then add a pullup to fight it. Otherwise, it's like turning on a space-heater instead of turning off the A/C. I'm not sure adding a pullup is enough to skew your measurements, but it definitely won't help.

[Andrew Neil]

---

@SWenn.1 wrote:

I have removed JP6  

---

JP6 is the Idd measuring link - when you remove that, you disconnect power from the Target MCU.

The idea is that you insert a current meter here to measure the Target MCU's current consumption: