TIDA-00548: What is proper initialization to avoid offset and wrap issues in differential mode?

Hi Darryl,

thanks for contacting TI!

When bypassing the PGA there is a higher input current at the AIN signals (~150nA). The 4.99kOhm resistors may cause a drop seen in the result. If the input is open (no loop current). The offset can be automatically compensated by initializing the offset calibration registers.
Regarding 00FFFFFF is seems ADC2 (24bit is read) representing -1d in this case.

Hope this helps!

Best regards,
Lars

The apparent offset is 19 mV on top of the 13mV signal that I can measure with a DVM at the input pins.  The 150nA input current does not account for this.  With 13mV in, the reading is around D1DE23.  I am not using the 24-bit ADC.  After the reading overflows from 00FFFFFF to 0, and I continue to increase the input, the reading from the ADC reaches 043BCE39 at my full scale, which value is larger than 24 bits.

Hi Darryl,

let me check. I'll come back to you!

Best regards,
Lars

Hi Darryl,

Assuming you leave the input unconnected for the offset measurement the absolute voltage should be around 0V. This is ok regarding absolute input voltage range as you bypass the PGA.
D1DE23 equals to 16.012mV which is close to your DMM value (how accurate is you DMM at this low voltage?).
Are you measuring the 04BCE39 at ~3.3mA input current?

Can you explain more on the 00FFFFFF to 0 overflow? The only overflow should occur from FFFFFFFF to 0, i.e. -1 to 0.
The codes for the 24.9 shunt should linearly increase (ideally) from 5197A24 to 197F62B7 for the 4-20mA.

Best regards,
Lars  

I have resolved the initial issue. In order to further characterize the problem I attempted to read back the register contents, and discovered that I could (no longer) read the factory default contents.  I had to change the SPI configuration, and I should point out that for my TMS570LC4357 I had to set Polarity=1 and Phase=0 which is opposite of what the ADS1263 datasheet says. This change removed a mysterious 2X error in the resistance and voltage measurements as well as the odd overflow I saw in the differential measurements.  My current measurements still have a 7uA offset (at zero current), that may be related to the 150nA input current you mentioned.  In differential mode, would I see 150nA on each input or a total of 150nA ?  How can I calculate the expected offset in the current measurement?