Zero-Scale System Calibration Mode

Hi, I've been trying to calibrate the ad7730, and I suspect the Zero-Scale System Calibration Mode calibration is wrong (I haven't considered Self-Calibration Mode).
I am using a cell simulator, which at its minimum value reaches the AD7730 the voltage of 0V and at its maximum value it reaches 10mv.
The system calibration process is these 3 steps:
step1: I perform the Zero-Scale System Calibration Mode and obtain offset values.
step2: I proceed with the Full-Scale System Calibration Mode, I get gain values
step 3: finally I continuously read the data and if I can get the data.
Apparently the calibration of the system-full is done correctly, because when reading the maximum value of the simulator the data shows me 65535 (max value of 16 bits), now if the value of the load simulator lowers it to the minimum (0 volts in the ad7730 input) the ad7730 data oscillates between 00E0 (224DEC) when it should be ZERO (0 decimal).
Is it normal that the minimum value does not reach zero?
What am I failing when it comes to Zero-Scale System Calibration Mode?
maybe it could be that the ad7730 is damaged?
When calibrating, is it necessary to do so by deactivating CHOP mode?
When is Self-Calibration Mode necessary?
For this calibration process I am configuring the same filter register value in both calibrations (full and zero Scale System Calibration Mode) - AC activated, chop mode activated, SF = 256, filter register value = 20 00 30 hex

the value of the dac register is 0x20 (it does not subtract or add voltage)

Parents
  • perform the Internal Zero-Scale Calibration and Internal Full-Scale Calibration, but the problem remains, I suspect that the problem is the ad7730 because when reading the test register the value is 010080 hex and according to the data sheet it should be 000000 hex, and the gain register is 0x593CEA and according to the data sheet it should be 0x59AEE7, these reasons may be the reason for the problem ??????

  • 0
    •  Analog Employees 
    on Feb 22, 2021 1:42 AM in reply to Edward-Yamunaque

    Hi, 

    Let me check the details above and get back to you if I see anything wrong. However, can you confirm that the above readings are reading upon power up/resets and without any inputs present in the ADC? meaning you haven't write to any of the registers and haven't perform any measurements?

    Another thing, do you have expected offset value from load cells? 

    Whenever a system calibration mode is used, there are limits on the amount of offset and span which can be accommodated. As you can see on the specification below, the range of input span has a minimum of 0.8xFS and a maximum value of 2.1xFS and the offset can handle with a minimum value of -1.05xFS. However, the span as well as the offset has to take into account of the limitation of the positive full-scale which is 1.05xFS. Let say for example, If the part is used in unipolar mode with a required span of 0.8 × FS, the offset range the system calibration can handle is from –1.05 × FS to +0.25 × FS. Meaning, in determining the limits for system zero-scale and full-scale calibrations, you have to ensure that the offset range plus the span range does exceed 1.05 × FS. As long as your input does not exceed this upper and lower limit you will not end up in out of range signal.

    For load cells application, a self calibration may not be enough since a load cell has an offset or TARE associated with it. Thus, it is recommended to perform system offset calibration to removed the loadcell's offset voltage. The AD7730 has internal DAC to remove TARE values of up tp +/-80mV from the analog input signal. 

    Thanks,

    Jellenie

  • Thanks for answering, I don't have a load cell, I'm using a cell simulator.
    I have a doubt.
    In the final application with the load cell installed, the input of the ad7730 will never have an absolute zero (0v), because with the weight of the tray the load cell will have a torque and this will make the input of the AD7730 never be 0v (absolute) ???
    I think it is better to talk about relative zero.

  • +1
    •  Analog Employees 
    on Feb 23, 2021 5:11 AM in reply to Edward-Yamunaque

    Hi, 

    Okay I think we would never really getting an exactly 0V because the ADC itself has an internal noise. So even after performing system calibration you will still measure the ADC noise. System calibrations will calibrate the ADC offset and gain along with external offset and gain. So the errors remains will be the noise value, that is why you will still see some flicker codes. But that's the benefits of using a high resolution ADCs with lower noise performance. If you want a lowest noise performance, I would recommend using AD7190. 

    Thanks,

    Jellenie

Reply
  • +1
    •  Analog Employees 
    on Feb 23, 2021 5:11 AM in reply to Edward-Yamunaque

    Hi, 

    Okay I think we would never really getting an exactly 0V because the ADC itself has an internal noise. So even after performing system calibration you will still measure the ADC noise. System calibrations will calibrate the ADC offset and gain along with external offset and gain. So the errors remains will be the noise value, that is why you will still see some flicker codes. But that's the benefits of using a high resolution ADCs with lower noise performance. If you want a lowest noise performance, I would recommend using AD7190. 

    Thanks,

    Jellenie

Children