High current draw Vbat Bluepill

[NLHL]

Hi,

This is the first time I am starting a post, sorry in advance if it's not in the correct place.

I have a Blue-Pill with original STM32F103C8T6. I have Vbat hooked up to a 0.33 F super cap via 100 ohm. Vdd is connected to V bat with a BAT54C Schottky diode. Vbat is decoupled with a 100 nF MLCC capacitor of good quality.

I am using the RTC clock and the backup SRAM. It all works well. However, when powering down it appears that Vbat drops rapidly while drawing somewhere between 2 and 3 mA (depending on the charge level) of the super cap.

At first I thought that I had broken the chip by accident and therefore I ordered another Blue pill, but the effect stayed the same: high current consumption.

I checked the decoupling capacitor. I checked the diode, even took it out during Vbat operation. I also let the program run in an empty while loop. I disconnected the programmer. Nothing seems to help. The AC level on the 32.768 kHz crystal in V bat and normal mode is about 300 mV, measured with a 1:10 scope probe. I have put all unconnected pins in a known output state. Holding the reset button pressed makes no difference, as expected.

According to the schematic diagram of the Bluepill, Vbat is only connected to pin 1 of the board. PC13 (alarm) is connected to the status LED but there is no current flowing through that pin during Vbat operation.

I use CubeIDE 1.13.2.

I made designs with the F411 and the same supercaps, which gives me no problems at all. There is probably some silly reason why I measure this high current, I probably overlooked something. But what?

Regards,

Harold

[Peter BENSCH]

Welcome @NLHL, to the community!

How can you be sure that an original STM32F103 was used on your Blue Pill?

The Blue Pills have been fitted with counterfeits for years, and of course only the manufacturer can guarantee or help with its functionality.

Regards
/Peter

[NLHL]

Hi Peter,

Thanks for the quick response.

That question crossed my mind but how can I tell? To me it looks like an ST device, it's not a clone. I deliberately bought it from a trusted Dutch supplier (four times the price of a clone) and avoided the Ali clones which seems to have a CS32 chip.

Attached I have a picture of the board/chip.

Regards,

Harold

[Peter BENSCH]

Unfortunately, marking is no longer a clear criterion; instead, there are many users with websites dedicated to this topic, e.g.

• GreaseWeazle or
• mecrisp-stellaris-folkdoc.sourceforge.io with a test programme or
• images taken by Richi's Lab of fakes/counterfeits and genuine devices

You can give it a try and share your findings here.

Good luck!
/Peter

[TDK · Accepted solution]

The chances of getting a genuine STM32 on a bluepill board even if you wanted one are vanishingly small. It is not an official board. Probably the only way to ensure this is to buy a board, then replace the chip with a genuine STM32F103 from a reputable supplier.

[AScha.3]

@TDK ,

My experience is different : i have some (10+) "blue pill" boards, most have STM chip, just a few have CKS or GD "F103" , but never i got a  fake ("ST" logo, but other chip inside). When reading the description, most time the supplier (on Ali) even write something like : "original F103 from China" - then you know, this will be no STM chip.

But most people dont read the "details"...and then complain about a "fake".

Maybe its also because i always read the comments before i buy such a suspect thing, with one comment about "not working" or similar, i look for other supplier - there are enough, who send what they promise.