• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 46 subscribers
  • Views 457 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Original question:

ADC081C027

ADC081C027: ADC081C027 reading error

Apostolos Papastergiou
Prodigy 20 points
Part Number: ADC081C027 
We have installed at the field  ~ 50K devices that use this chip.

We have an issue at the field with some of them ( 1%):

They need about 100 readings in order to give the correct result.

The first reading is about 1V higher than the real value and each next reading is lower than the previous.

After ~ 100 readings ( 1ms intervals) the value is correct and it remains correct as long as we keep reading.

The supply voltage of the chip is correct-stable at 3.3v and the ADC reads the voltage of a li/ion cell having a 20k+20K divider at its input.
  • 0
    TI__Mastermind 24970 points
    Hello,

    Thank you for your question, we are taking a look at it and will respond soon
  • 0 in reply to Cynthia
    TI__Mastermind 24970 points
    Hello,

    Can you measure at the input pin of the ADC to make sure the input is in fact correct, and not reflecting this 1V increase?

    Also mentioned was that 100 samples are needed at 1ms increments to finally reach the expected output.
    How is the output changing? I am assuming that data had to be repeating in the output data.
    The ADC is an 8 bit device, and with an input range of 3.3V, that means that the LSB is 12.9 mV
    If each sample was less than the previous, that means that less than 100 samples would cover the 1V delta.

    I would suggest taking a second look at the timing communications as well, to make sure that nothing has been unexpectedly affected.
    Regards
    Cynthia
  • 0 in reply to Cynthia
    Prodigy 20 points

    Hello Cynthia

    An external multimeter on the Vin pin at the same time reads the voltage correctly.

    As I wrote, on these 1% of the boards if we initiate a measurement some minutes after the previous measurement then we must read many times until we will get the correct result. The first measurement is about 1V higher than the real input and each next measurement is lower than the previous and after some measurements we get the correct value. Maybe some measurements have the same value, the important is tha the chip needs some time to read the correct value...

    What the timing on I2C has to do with this case??

    BR
    A 

  • 0 in reply to Apostolos Papastergiou
    TI__Mastermind 24970 points
    What is the normal use case of the device for the application? For example, when sampling the input, is it usually sampled once, twice, multiple times, and then left be until the next time the input need to be monitored?

    There is no driving circuit being implemented. This device performance best when driven by low impedance, less than 100 ohm, source, currently the input signal comes across a 20K resister divider, this is high impedance input. Thus it can be that the input is not driven properly and the internal sample and hold circuit is not settling. The input is sampled for 8 internal clock cycles, typically 400ns, but this time can be as long as 530ns, it could be that the 1% of the devices you are seeing are on the shorter end. Would you be able to lower the resister divider values? Perhaps placing another 20K in parallel, essentially halving the values and see if this helps.
    You mentioned that multiple devices are at the field; wanted to make sure that this set of devices having issue had not been exposed to anything that could be causing any irregularities.

    Regards
    Cynthia