stm32f103 clones

[ag123]

this is a plain old topic
https://hackaday.com/2020/10/22/stm32-c ... -the-ugly/
viewtopic.php?t=2038

[ag123]

the actual post for a follow-up is this:

I tested out the temperature sensor for a blue pill, the stm32f103c8 has a poor marking and is possibly clone
'most' things work, I get the RTC freeze issue every once in a wrong sequence of ops.
then I happened to test out the temperature sensor

so checking the specs
https://www.st.com/resource/en/datashee ... f103c8.pdf
5.3.19 Temperature sensor characteristics
V25 min 1.34, typ: 1.43, max: 1.52
slope: 4.3 mV / deg C

Code: Select all

	// specs 5.3.19 temp sensor characteristics
	// V 25 deg ~ 1.43v
	// slope 4.3 mv/C
	float temp = (mv - 1430) * 1.0 / 4.3 + 25.0;

when run based on specs the calc goes

temp sensor analogRead:1597, mvolt:1271, temp(deg C:)-11.98
temp sensor analogRead:1590, mvolt:1266, temp(deg C:)-13.14

room temperature is 30 deg C (around there)

I'm initially thinking that I may have a 'defective' stm32, then I did some tests
I used a value of 1.26 V ~ 28 deg C as the datum and reworked the linear projection.
I tried to warm up the chip, but instead of higher readings, the calculated temperature actually drops and is inconsistent.

this goofy behavior leads to a guess that the chip is after all a clone rather than a real stm32.