• State TI Thinks Resolved
  • Locked Locked
  • Replies 11 replies
  • Answers 1 answer
  • Subscribers 45 subscribers
  • Views 339 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.

ADS1232REF: ADS1232REF

Levi hendrickx
Prodigy 20 points
Part Number: ADS1232REF
Other Parts Discussed in Thread: ADS1232

Can you help me with the ADS1232REF? I bought this to test the ADS1232 to use it in one of our newest products. But unfortunately I can't get the ADS1232ref working.

I start up the ADS1232 with a battery. Went to settings and changes calibration weight to a weight i have here (50g). Also changes units to g. Now i want to read the actual value of my loadcell but i get very strange numbers. The weight is jumping very fast from low to high. can not see anything (Video attached). I also tried multiple loadcells but same result

  • 0
    TI__Guru** 113765 points

    Hi Levi,

    Welcome to the E2E forum!  Unfortunately the video did not attach.  It would appear that you have a very noisy setup and perhaps are calibrating on noise.  Can you tell me the capacity of the load cell (maximum weight that can be applied to get full-scale output) and the sensitivity (mV/V)?

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Prodigy 20 points

    Hello Bob,

    I sent the video on email to your colleage (Jejomar Ildefonso). i dony have it on a link. only in a downlaod version. I tried different loadcell. I mostly use lodcell from zemic with a capacity of 600 gram and a sensitivity of 1mV/V.

     

    The output on the ADS1232 is jumping al around.

    Kind regards

    Levi Hendrickx

  • 0 in reply to Levi hendrickx
    Prodigy 20 points

    Hello Bob, See here the video´s please 

    Kind regards

    Levi Hendrickx

  • 0 in reply to Levi hendrickx
    Prodigy 20 points

    The sound is from the office this has nothing to do with the video

    Thanks

  • 0 in reply to Levi hendrickx
    TI__Guru** 113765 points

    HI Levi,

    I think your biggest issue is you have the shorting jumpers for the input still connected.  Remove jumpers located at J7 and J8.

    Also make sure that the gain is set to 128 and you should use 10sps (low speed) for the data rate.  You may still see some issues with calibration as you are using a calibration weight that is only about 1/12 of the capacity.  This ends up as a low signal level of about 416uV.  You should calibrate closer to full-scale output.

    You may find the information in A Basic Guide to Bridge Measurements useful as well.

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Prodigy 20 points

    Hello Bob,

    Thank you for your reply.

    In the movie i indeed still had the jumpers on but i also tested without the jumpers.

    See here a movie with all the right settings and still the weight is jumping arround.

    I read the whole datasheet and think all the settings are good but maybe i still make some mistakes.

    Waiting for your reply.

    (The sound is from the office and not for this movie)

    Thank you

    Levi Hendrickx

  • 0 in reply to Levi hendrickx
    Prodigy 20 points

    Hello bob,

    It looks like in analys mode i get good raw data but when i switch to scale mode the calculations are going wrong and the weight is jumping very hard.

  • 0 in reply to Levi hendrickx
    TI__Guru** 113765 points

    Hi Levi,

    There will never be a stable code because of noise.  There is both the noise of the conversion due to quantization and external noise due to pick up of the load cell wiring.  To reduce the overall noise in the limited range of measurement you need to make sure that the gain is set to 128 and the data rate (speed) is set to slow or 10sps. Expect to see a lot of code jumping based on the noise tables in the datasheet and displayed in Figure 6-2.  Looking at table 7-1 you can see that the noise-free resolution for a gain of 128 is 18.4 bits which means the number of noise bits from the conversion is 5.5 (24-18.4) or about 48 codes of flicker.

    To determine the amount of noise-free scale counts you take the total number of available noise-free codes and multiply times voltage of full-scale of the load cell divided by the FSR of the ADC.  2^18.4 * (5V * 1mV/V) /  39mV which is equal to (2 * 0.5Vref / PGA gain).  This becomes about 44346 available noise-free scale counts as the full-scale output for the load cell is only about 1/8 of the full-scale of the ADC.  If the full-scale output is represented by a 600g capacity then you should be able to resolve to 0.02 grams stable for the ADC excluding any external noise.

    When calibrating you should use a weight that approximates the full-scale output of the load cell you are using.  So if you show 2kg for the calibrated weight  (CW) and you calibrate with 50g, then you are giving the firmware code the wrong information and cannot expect good results.

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Prodigy 20 points

    Hello Bob,

    I understand that there is noise but the weight is jumping with 10 gram each step. 

    I am callibrating with 200 gram with a 600 gram loadcell. Is this good enough? otherwise i have to buy one of 500 gram.

    The raw data is very precise so the system overal is working.

  • 0 in reply to Levi hendrickx
    TI__Guru** 113765 points

    Hi Levi,

    From your last video you have it set for 2kg and not 200g (0.2kg).  There is a bug in the code when the device changes values around the decimal point.  This same bug will prevent you from using the grams scale as you will be limited to 99 grams as the maximum weight that can be applied.  If you go to the grams setting you are stuck at 99 grams as the largest CW value and if you go the kg setting you have to go to the last available blinking digit and slowly decrease the value below one kg so that the CW setting shows as 00.2KG.  I would be very careful here as you can lock the system up so that the scale mode will no longer work if you move the digit below one.  This will happen if you attempt to save your settings with invalid configuration and you will not be able to reset the device to get it working again.  So I would not save any scale mode settings until you are confident the scale is working as you would expect.

    Keep in mind that the ADS1232REF is just a demonstration of what can be done and will not show great levels of precision.  I would recommend that you manually calculate your slope using the method described in the ADS1232REF user's guide and the calibration section from A Basic Guide to Bridge Measurements.  To evaluate, use the analysis mode and make your own calibration from the codes displayed to determine the zero point, and the calibration point.  The difference in codes from zero to calibration weight will cover the 200 gram spread for any intermediate values.  This information in the end will be useful for building your own system as you will need to develop firmware to accomplish the weighing of items placed on the load cell.

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Prodigy 20 points

    Hello Bob,

    Thank you for this info. This helps.

    I already did some good tests in analys mode.

    Kind regards

    Levi Hendrickx