• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 78 subscribers
  • Views 352 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.

MSP430F6723A: MSP430F6723A ADC performance

Nikolay Isakov
Prodigy 460 points
Part Number: MSP430F6723A

Dear colleagues,

I need to measure  0...1V signal using SD24B.

Internal reference allows this: 

But the question is the stability of this 1V point - if we calibrate our device using only one 1V point, do we comply with Performance Data in 5-38 Table of the datasheet? Maximum differential input voltage is +-920mV at G=1

  • 0
    TI__Genius 17960 points
    Hello Nikolay,
    this is a question, which cannot be answered based on your inputs perfectly.
    On one hand the ADC performance is heavily dependent on the HW setup you're using, and on the other hand the single point calibration is not influencing the ADC performance as such in terms of linearity or other architecture and silicon performance errors, as these are defined by the SD24B ADC design.
    The given performance levels of the ADC in table 5-38 are based on the ADC design performance. They cannot be influenced as such by calibration. The resulting errors in contrary e.g. due to offset or gain errors can be reduced by additional calibration and corrections. So one could improve the resulting performance in measurement accuracy.
    A single point calibration is a compromise, and will certainly not allow you meeting the best possible performance, which could be above the table parameters, as it allows you not to distinguish the errors in your measurement results which are resulting from offset errors and the errors coming from gain errors. Thus its value is limited. A two point calibration allows the separation of gain and offset errors, and thus the possible quality of corrections of the measured values can be better.

    Best regards
    Peter
  • 0 in reply to Peter Spevak
    TI__Prodigy 460 points
    Peter,

    OK, thank you. I had nearly same thoughts.
    But I wanted to make one thing more clear: if we use internal reference and G=1, are we ever able to work with 0...1V signal? While we have 1.15V int ref, but only +-920mV input range at this gain.
  • 0 in reply to Nikolay Isakov
    TI__Genius 17960 points
    Hello Nikolay,
    I assume you're referring to table 5-35 in the device datasheet. If your reference voltage is 1.15V, according to the datasheet, table 5-35 specifies min. +/-910mV input voltage range. That's ensuring the specified performance of the ADC. Exceeding this input voltage range, going to full scale input range degrades the performance.
    Now this would in your case, if connecting the negative reference input with GND, which usually is the case, then close to 1V you would face a degraded performance. So if you'd like to have full performance, you need to either lower the input voltage, e.g. by a divider, or you need to shift the voltage reference values, to be able to utilize a portion of the negative input range for your 0-1V signal range. But this would be probably more difficult to achieve than the pre-divider solution.

    Best regards
    Peter

**Attention** This is a public forum