• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 45 subscribers
  • Views 527 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

Ads1247 Offset Problem

Lelumpolelum
Prodigy 100 points

Other Parts Discussed in Thread: ADS1247

Hi,

  Im trying to measure thermocouple voltage by using ADS1247 A/D converter, and i have measurement error 15764uV-15343uV=421uV ~0.421mV. For me this offset error is unacceptable. This error appears at Gain=128(also 64 ), 22uF capacitor as an reference( 2.048V) . Sample Rate 5. Environment temp 25'C . I've checked voltage at ADS1247 and it is proper value 15343(agilent 344**). This offset exist at all my interesting range(1400uV-15343uV). I dont know what cause this offset error.

 I''ve observed that it could be the ground problem. Im using switching power supply (isolated ground) when i connect other ground ie oscilloscope ground or jtag  Offset disappears  .Of course i have  separated Digital and analog Ground. They are connected in one place by ferrite. . When i disconnect jtag or oscilloscope i have this offset again.

 

  • 0
    TI__Guru** 113765 points

    Lelumpolelum,

    Can you send us a schematic of how your are making your connections and what supply voltages you are using?  Are your supply voltages for both analog and digital being derived from the same source?  How do you have the thermocouple connected?  Are you biasing the thermocouple within the common mode input range of the ADS1247?

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Prodigy 100 points

    I will send schematic tomorrow. I left schematic on my second computer. I will try to describe it. Power supply is DC/DC converter (aimtec) after DC/DC i have PI Filter to suppress switching frequency. 5V from DC/DC go through ferrite and capacitor( 1uF to DVDD ADS1247 from this point  5V goes also through ferrite and capcitor 1uF to Avdd. Other 5V branch supply linear regulator TPS73633 5V/3.3V. 3.3V is for ucontroler. 

    Yes it  is the same source DC/DC Avdd and Dvdd =5V.

    Thermocouple is connected directly from cold junction to ADS1247. Green cable is positive White is negative. I've thermocouple calibrator. Calibrator is ok i've checked voltages.

    Im using biasing from ADS1247 it is in common input range.

    Im think crucial thing in this error is digital ground, when i connect oscilloscope or Jtag ground to this digital ground offset error is acceptable. Sorry for my english.

  • 0 in reply to Lelumpolelum
    Prodigy 100 points

    I think i've solved problem. Problem lies in shielding. When i connect charger to my calibrator offset error increase, when rs485 cable was near compensating cable offset error again increase 100uV. So i decided to use compsating cable with shield i though 0.5m compensating cable without shield its not big deal. Im considering one thing, why agilent 344** measurement was ok. Maybe it has better filtering. Is this filters necessary?. Im asking  because the same line is used to measure thermocouple voltage and PT100 voltage  so im worry about using resistors in this filters. Resistance in paths  can be different, it could be the another reason for error in Pt100 voltage.

  • 0 in reply to Lelumpolelum
    TI__Guru** 113765 points

    Lelumpolelum,

    I have seen RFI/EMI interference cause a jump in readings.  Depending on how severe the noise interference is filtering may be necessary.  Adding the filters will disturb the ratiometric measurement, add noise, etc..  Shielded cable will help to reduce the effects of the interference if it is being generated in the cable, so I do highly recommend it.  These types of situations is what is difficult in matching the theoretical to the actual operating environment.

    Best regards,

    Bob B