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.

ADS1158 - Accuracy issues at higher humidity

Other Parts Discussed in Thread: ADS1158

Hi all,

In one of our designs, we have used ADS1158 to measure the analog inputs. Our design supports both Voltage mode (0-10V) & current mode (0-20mA) on the same analog input. Thus we have used an analog switch, which will connect the analog input to either voltage load or the current load. We have used a 200E fixed resistor as current load & a resistor divider network as a voltage load. Resistor divider will reduce the 10V input range to ADC input range.

Below is the brief block diagram describing the architecture followed for analog input measurement :- 

 

 Figure 1

  We have a accuracy requirement of 0.1% of FSR at room temperature & 0.2% of FSR over the operating temperature range of -20°C to 70°C. We are performing calibration of analog inputs at room temperature to satisfy the above accuracy requirement.

 We are able to achieve the required accuracy over the operating temperature range of 20°C to 70°C when the humidity is set to 0%. But we are facing issue when we test the accuracy at 70°C & 95% humidity. We are observing around 0.3% decrement in accuracy (around 30mV) at 95% humidity. Same behaviour is observed at 95% humidity at 50°C & 60°C.

 We have gone through the ADC data-sheet and it does not specify any effect on the accuracy with respect to humidity. 

 Can you please suggest any reason for this behavior ?

 

 

 

  • Hi all,

    There were some formatting issues in the earlier post. Re-posting the same query.

    In one of our designs, we have used ADS1158 to measure the analog inputs. 

    Our design supports both Voltage mode (0-10V) & current mode (0-20mA) on the same analog input. 

    Thus we have used an analog switch, which will connect the analog input to either voltage load or the current load.

    We have used a 200E fixed resistor as current load & a resistor divider network as a voltage load. 

    Resistor divider will reduce the 10V input range to ADC input range.

    Below is the brief block diagram describing the architecture followed for analog input measurement :-

     

     

    We have a accuracy requirement of 0.1% of FSR at room temperature & 0.2% of FSR over the operating temperature range of -20°C to 70°C. 

    We are performing calibration of analog inputs at room temperature to satisfy the above accuracy requirement.

    We are able to achieve the required accuracy over the operating temperature range of 20°C to 70°C when the humidity is set to 0%. 

    But we are facing issue when we test the accuracy at 70°C & 95% humidity. 

    We are observing around 0.3% decrement in accuracy (around 30mV)  at 95% humidity.Same behaviour is observed at 95% humidity at 50°C & 60°C.

    We have gone through the ADC data-sheet and it does not specify any effect on the accuracy with respect to humidity. 

    Can you please suggest any reason for this behavior ?

    Thanks,

    Harsha Lalwani

  • Harsha,


    I'm not aware of any specifications that include accuracy with humidity. I don't think our competitors have similar specs as well.

    The most likely solution I can think of is to make sure the boards are clean. After soldering parts on the boards, there can often be flux and other residue on the boards. With added moisture, the resistive properties can change and you could be getting some inadvertent leakages in the circuit. This can cause a change in the output impedance of whatever you are measuring. Since the input impedance to the ADS1158 is a very low 65kOhms, it would be sensitive to these changes.

    I would make sure the input path from the terminal to the device pins were clean, perhaps using isopropol alcohol and a brush. I would also clean paths for the MUX OUT and ADC IN for whatever buffer you are using (or even just the pass through for these terminals). Additionally you want to make sure the VREF lines are clean as well.

    Are you looking at multiple boards with this problem? Do they all have a similar change in error? Try cleaning and see if this improves the performance or minimizes the error.

    If this does not help, we can talk about the circuit itself. I may need a schematic and we can discuss options if cannot post it to the forum.


    Joseph Wu