• State Resolved
  • Locked Locked
  • Replies 12 replies
  • Subscribers 46 subscribers
  • Views 3307 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.

ads1232 hangs when input voltage reaches above 3.3 millivolts

sivanantham Annamalai
Prodigy 80 points
Other Parts Discussed in Thread: ADS1232, LM2576

I am working in ads1232 to readout load cell weight data. when i read  all 24 bits of  ADC data the ads stops its conversion at 8388608 counts and if i read 19 bits then 524288, which is maximum value of corresponding bits but it reaches before the maximum value of load cell output(i-e maximum count achieved when load cell output is 3.4 millivolts but maximum output of load cell is 10mV) . I am using single point load cell with the capacity of 20 kg. output range is 2mv/v. supply voltage is 5V both for analog and digital side.ref +ve also connected to 5v. why its conversion reaches maximum counts early. I tried it with all Gain values(1,2,64,128). Please tell me what what can be the problem?.

  • 0
    TI__Guru** 113765 points

    Hi Sivanantham,

    Can you send me your schematic?  8388608 is positive full-scale.  You may want to check your reference connections to make sure that 5V is truly connected.  You may also want to check your load cell connections to make sure that there are no issues there (and that you have the load cell wired properly) and that the output voltage of the load cell is in the correct common mode input range for the ADS1232.

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Prodigy 80 points

    Dear Bob B

    Thanks for your reply.I have attached the schematic section of ADC unit.i measure the reference voltage in refp pin, it has 4.93 as in the power supply given to the ADC.

  • 0 in reply to sivanantham Annamalai
    Prodigy 80 points

    The count measurement per gram variation in 10kg load cell is almost 140 counts.Noise may be higher in my system,please let me know any other information you want from me. If we are able have a telephone conversion, please contact me in this number is +919940018350.

  • 0 in reply to sivanantham Annamalai
    TI__Guru** 113765 points

    Hi Sivanantham,

    REFN on the ADS1232 should connect directly to AGND.  I think this is your main issue.

    AGND and GND should have a low impedance connection near the ADS1232 (is this JP3?).  Also, you may want to replace L26 with a 10 to 20 ohm resistor.  Inductance in the power and/or ground has been known to cause issues.

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Prodigy 80 points

    Thank you Bob,

    I replaced the L26 to 10 ohm resistor,removed the capacitor.Now it works fine but at 80sps counts are drifting like 5 counts per 10 secs.At 10sps +/- 5 counts variation is there.How to reduce this because when we use high capacity load cells this may become a big issue. I am reading only 19 bits of ADC data.Even at 10 sps after 1 hr base oscillation gradually  increased by 30 counts then it started to reduce from there.

  • 0 in reply to sivanantham Annamalai
    TI__Guru** 113765 points

    Hi Sivanantham,

    The ADS1232 will have some offset drift.  You may want to intiate periodic offset calibrations for correction. (You should also issue the self offset calibration at startup.)  This is accomplished by generating extra SCLKs following a data read.

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Prodigy 80 points

    Thank you Bob,

     I am already initiating the offset calibration at start up.I will try with periodic offset calibration too. suggest me the some good Precision Data Converters which doesn't have this kind of offset drifts.

  • 0 in reply to sivanantham Annamalai
    TI__Guru** 113765 points

    Hi Sivanantham,

    Unfortunately if there are semiconductors involved there will be drift.  But there can also be other drift sources as well, or thermocouples created by device and layout.  Nanovolts is a small amount to try to measure.  What we often consider as DC levels really are not when trying to measure nanovolts.  Any disturbance from clocks, inadequate grounding or environmental noise can be problematic. 

    An unbalanced ratiometric measurement can also be an issue.  If you add filtering, you must consider the effect of filtering whether it be on the analog inputs or the reference.  To truly cancel the effects of noise, the noise must be in same phase and relative magnitude.

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Prodigy 80 points

    Thank you Bob,

    I tried with periodic calibration i couldn't find any big difference.I am using LM2576 regulator Ic for 5v supply,there i am using inductor.Is it creates any noise in powerline?

  • 0 in reply to sivanantham Annamalai
    TI__Guru** 113765 points

    Hi Sivanantham,

    You have diodes that I assume are for transient protection.  What is the part number?  Sometimes these have enough leakage to cause drift.  Can you send me your layout?

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Prodigy 80 points

    Thank you Bob,

    I put those diodes in PCB for transient protection. Now i am not using those diodes. i intended to use wurth electronic (Part no.825 500 40 ) in that place.I have attached the layout file.

    with ground fill

    Bottom side

    Bottom side with ground fill

  • 0 in reply to sivanantham Annamalai
    TI__Guru** 113765 points

    Hi Sivanantham,

    I'm sorry for the delayed response.  There are a number of possible noise sources, so it is difficult to pinpoint a single source.  You mentioned the power regulator, and it is possible this is causing some issues as the switching frequency harmonic may alias back into the passband of the ADC.  The ADS1232 has good power supply rejection, but this could still be a part of the noise issue.  We recommend to use a linear, low noise regulator.

    It would also be a benefit to have the differential capacitors closer to the ADS1232 input pins.  It might also make more sense to have the ferrites on the other side of the filter.  Component selection may make some difference as well.  Capacitor grade is especially important.  A high grade of cap is much more stable and has a lower noise dielectric.

    A bigger concern is your basic layout and where return paths are forced into less desirable locations.  Due to the nature of how the ground return flows, it might make more sense to use a common ground.  The grounds are common anyway, so why split them only to bring them together again. Please review the information in this section:

    http://e2e.ti.com/support/data_converters/precision_data_converters/w/design_notes/grounding-techniques.aspx

    Best regards,

    Bob B