nRFBell

Created by [Eduardo] • Started on May 26, 2025

nRF52840 development board with the nPM1100 battery charger, inspired by the nice!nano & Supermini nRF52840

May 26, 2025

My only pcb projects have been from Hackpad and Pixeldust. With Hackpad, I completely ruled out the use of passive components, however after using decoupling capacitors during Pixeldust, I had found new motivation to make a development board. Not any normal development board, but an nRF52840 dev board that includes its own battery charger like the nice!nano. I didn't want to copy the exact specifications and schematic, so I chose to use the nPM1100 battery charger from Nordic.

How I want to configure the nRF52840 so far:

  • Normal Voltage supply mode (same power to VDD & VDDH)

How I want to configure the nPM1100 so far:

  • Supplying the nRF52840 with buck regulator (VOUTB to VDD & VDDH)
  • 1.8v from buck regulator (VOUTBSET0/1 to GND)
  • Termination voltage at 4.2v (VTERMSET to GND)
  • USB port detection for VBUS limit (D+ & D- to USB host, ISET to a GPIO)
  • Buck regulator automatic mode (MODE to GND)
  • Ship mode available (SHPACT in pinout or onboard button)
  • Indication LEDs

I have this so far:

May 26 schematic Im unsure, however, of all the capacitors and resistors needed, as well what specific crystals and their capacitors to use for the nRF52840 chip.

Total time spent: 3h

June 1, 2025

The nPM1100 was a terrible choice for this board. 1. Too complicated 2. The voltage of the GPIOs on the nRF52840, which I want at 3.3v, are tied to the voltage of VDD. 3. This makes the nPM1100's bulk regulator that outputs 1.8v-3.0v useless on this board (to my knowledge) 4. I don't want a large inductor on this board which I believe is required for the nPM1100

After looking at this post, I decided to continue with a Texas Instruments battery charger, landing on the BQ25185. I chose this over a BQ2407x because it supports LiFePO4 batteries. Now for my next journal I have to finish wiring this and figure out an LDO.

June 1 schematic

Total time spent: 3h

June 2, 2025

I finally finished with the battery management (I think) but I am unsure about how I want to have the charging current (ISET) and input voltage & VBATREG (ILIM/VSET) configurable on the board as they may take up too much space.

Circling back to my uncertainty on decoupling capacitors and such, Nordic recommends using the exact decoupling capacitors from example schematics they give on the datasheet.

I also made a pinout inspired by the nice!nano and the Liatris controller. From the nice!nano, I took the extra 2 pins but made 1 VBUS so that its possible to charge the battery without the USB-C port.

Onto the LDO, I am now realizing that I need a boost converter or a charge pump if I want to use a LiFePO battery?? I may be approaching this the wrong way.

June 2 schematic

Total time spent: 7h😭

June 10, 2025

Changed out the battery charger for 2nd and 3rd time 🏴‍☠️ I wired the CN3063 battery charger with its beautiful datasheets only to realize it didn't have power-path which is something I wanted. Then I had to face the BQ24232 and its datasheet. Texas Instruments has to have the worst datasheets ever made. A literal hour was spent on solely the ITERM pin (probably my inexperience but still).

Anyways, I finished placing decoupling capacitors for the nRF52840 and also editted the pinout.

June 10 schematic

Total time spent: 5h

June 14, 2025

The next thing I wanted to add and learn about were diodes, specifically a Schottky diode on the VBUS of the USB port. Because I put the VBUS or now 5V on the pinout, I needed one to prevent whatever happens when VBUS and 5V are powered diodeless. I am still unsure on what Schottky diode to choose as I didn't want much voltage from the USB to be lost.

I moved onto the LDO where I chose the MP20051, an 3.3v LDO with a very low dropout. I realized that I should be using VDDH on the nRF52840 with the output from the battery charger so that it would last longer. I was skeptical when I read In High Voltage mode, the configured output voltage for REG0 must not be greater than REG0 input voltage minus the voltage drop in REG0 without any clarification but then I saw it would just lower VDD/REG0 output so that would work in making the battery get used up more.

June 14 schematic

Total time spent: 4h