Working MKR Zero with battery

[herz_tau]

I bought several Arduino Zero boards and they all behave the same.

1. The charge indicator LED (DL2) is constantly lit when the board is powered from the USB, regardless of whether the battery is connected or not. It seems that I am not the only one who has encountered this.
2. What is even stranger is that when the battery is connected, it does not charge, but on the contrary, the current consumed from it (in my case, it is about 120 mA) does, despite the fact that the power from the USB is maintained (the DL1 LED is on). The battery voltage is 4.1 V, and the voltage supplied from the USB is 4.9 V. If the USB is disconnected, the consumption from the battery remains the same. Without the battery, however, the board works and the sketch is successfully executed. The battery polarity has been checked and is correct. Its capacity is more than 2000 mAh. If I connect a supercapacitor instead of a battery, it will only charge to 0.3 V.
   And I didn't find any reports about such a situation on the Internet. Am I the only one who was "lucky" and all my boards are faulty in terms of the charger circuit? Or am I doing something wrong?

[herz_tau]

Addition: the charging chip is definitely sucking current from the battery. The voltage on the Vbat pin is only 0.3V. Almost all the battery voltage is applied to the F1 fuse, heating it up.

[sonofcy]

It will be easier for helpers to understand if you draw a wiring diagram and post it here. Also post all code in code tags.

[herz_tau]

I don't think the code matters in any way. It's a hardware issue. Even the board out of the box, before programming, behaves the same way.
As for the wiring diagram, everyone who has dealt with the MKR Zero has one. Here is the relevant part:
Clipboard_06-23-2025_01.pdf (168,3 КБ)

[jremington]

@herz_tau

The battery polarity has been checked and is correct.

Regardless of that unsubstantiated claim, it sounds like the charger IC was destroyed by connecting a battery with the wrong polarity.

Li-Po battery WARNING!

I dont know if there is a convention for polarity order on JST plug, but most Li-Po batteries you can purchase on-line have it reversed in comparison to the plug on the Arduino board. I came across this post Power an Arduino MKR ZERO by batteries and had the same experience with similar battery I purchased. In-built battery protection circuit stops battery from charging and because of reversed polarity Arduino protection stops it from powering the board. What is misleading is that charging LED …

[herz_tau]

Thanks for the reply. I read that post and it said, among other things:

and because of reversed polarity Arduino protection stops it from powering the board.

Which is absolutely true if you look at the diagram: parallel diode D4 simply will not allow reverse voltage to be applied to the chip.
But I checked the polarity several times, and even if I made a mistake, it would be once, not all four, given the number of boards I have.

[jremington]

@herz_tau

parallel diode D4 simply will not allow reverse voltage to be applied to the chip.

Assuming that the diode is not open or shorted and is installed with the correct polarity. Check the diode using your multimeter (battery disconnected, of course).

Details of the charging circuit:

[herz_tau]

You know, your advice came in very handy. On all boards this diode was soldered in the opposite direction. It never even occurred to me to assume such a possibility.
By the way, all the boards were bought on Mouser.

Thank you!

[jremington]

Glad you found the problem! This comment was an important tip, consistent with a forward biased Schottky diode:

@herz_tau

If I connect a supercapacitor instead of a battery, it will only charge to 0.3 V.