3.3V power supply for NucleoF103

[marcogrossi89]

Hello,

I need to power the NucleoF103 development board using an external 3.3V power supply. I have read in the NucleoF103 manual that this is possible by removing SB2 (to disconnect the voltage regulator output) and SB12 (to disconnect NRST). I also have read the schematics of the board and it looks that the power supply for the microcontroller (VDD) is provided by 3.3V connector through the JP6 jumper. Is the same thing if I mantain SB2 connected, remove the JP6 jumper and provide the 3.3 power supply to the pin of the JP6 jumper instead of the 3.3V pin of the Arduino connector? Moreover, what is the reason for disconnecting NRST by removing SB12? I have already programmed the board using ST-Link before switching to the external 3.3V power supply. Is there any other way to proceed without removing SB12 (for example connecting T_NRST on CN4 to ground)?

Thanks.

Marco

[SofLit]

Hello,

If you power supply the MCU through JP6, what will be the power supply of the STLINK module? here you will only power supply the MCU. The removal of SB12 will ensure that STLINK will not assert the reset in this state.

You need to look deeply at the schematics to understand the relationship between the power supplies.

[marcogrossi89]

Hello,

I do not need to power supply the ST-Link since I have previously programmed the MCU with the firmware using the USB power supply. When programming is done, I need to switch to the external 3.3V power supply. Regarding the removal of SB12, at what voltage level (GND or VDD) should I tie T_NRST on CN4 to guarantee STLINK will not assert the reset with SB12 present?

Thanks.

Marco

[SofLit]

---

I do not need to power supply the ST-Link since I have previously programmed the MCU with the firmware using the USB power supply. When programming is done, I need to switch to the external 3.3V power supply.

---

In that case you can power supply the MCU with JP6

---

Regarding the removal of SB12, at what voltage level (GND or VDD) should I tie T_NRST on CN4 to guarantee STLINK will not assert the reset with SB12 present?

There is an internal pull-up resistor on NRST pin inside the chip. No need to pull it up/down. CN4 is the connector for an external STLink usage, I don't see the relation here! You need only to disconnect SB12 to disconnect T_NRST output of ST-Link from NRST of the MCU.

[marcogrossi89]

I do not understand very well. The removal of SB12 assures that ST-Link is not able to send reset commands to the MCU. However, when I remove JP6, the ST-Link is not powered. Since, as explained, NRST of MCU has a pull-up resistor inside the chip to prevent reset from noise, how is it possible that ST-Link sends a reset command when it is powered down, even if SB12 is present?

[Andrew Neil]

You don't tie NRST to anything - it has an internal pullup:

[Andrew Neil]

---

@marcogrossi89 wrote:

how is it possible that ST-Link sends a reset command when it is powered down, even if SB12 is present?

---

When the ST-Link is powered down with SB12 present,  it would pull the Target's NRST pin down.

Note that there are also connections from the ST-Link MCO to the Target OSC_IN, and from the ST-Link UART pins to the Target UART pins - you might also want/need to disconnect them ...

[marcogrossi89]

Thanks for the information. The problem is that I must measure some characteristics of the MCU under different power supply voltages (in the allowed range 3V - 3.6V) and I am not very good at soldering small components like SB2 or SB12. I ask if the following setup can cause some problem (in the case the MCU is already programmed):

1) I plug the USB port of the NucleoF103 to the PC with jumper JP6 removed. This way the voltage regulator provides the power to the ST-Link but the generated 3.3V are not provided to the MCU since JP6 jumper is not present.

2) Using a DC power supply (with the GND of the power supply shorted to the GND of the NucleoF103) I provide the target power supply using the pin of the JP6 connector.

Can this configuration work correctly?

[Andrew Neil]

The trouble is, if there are any connections to the ST-Link, you will get leakage effects there - so you won't just be measuring the Target MCU.

This is especially significant if you're planning on measuring sleep currents.

Probably the simplest approach without soldering would be to simply break-off the ST-Link part of the board - it is designed to do that.

[marcogrossi89]

However, if I remove the ST-Link part of the board, how can I reprogram the MCU if I need to do that? I do not need to measure sleep currents but the value of the high and low thresholds of the schmitt trigger circuit integrated in the GPIO interface for pins set as digital input.