ADC reading errors

[con2021]

We read 3.6 V using the ADC, for our mass production devices. The error rate is about 50% of the devices show wrong values between 3,05 and 3,4 Volts.

Even though the ADC input is almost the same for all the devices (about 0.84 V), each modem shows different values after reading this 0.84 V and applying the same conversion formula.

What is happening, is the part-to-part spread that high, that the ADC is different in every device.

[Sigurd Hellesvik]

Hi Maysam,

Our ADC should not have this high error rate.

I have some questions on your test setup, to get a better picture of the situation:

Do you measure batteries? Do you get the same error rates if you measure from bench power supplies?

When you measure using the internal ADC, how are you confirming the voltage of the voltage your measure (Eg. a second voltage meter)?

Regards,
Sigurd Hellesvik