questions for using nPM1300

What is the power consumption to check battery level?

How quickly at boot time is an accurate battery state of charge available?

What is the RDSon in the path from VBUS or VBAT to VSYS?  (Assuming there are either internal diodes or a FET for protection)

Can the VSYS of one nPM1300 be used to power the VBUS of another nPM1300 (chained together)?

Is there a trade off of accuracy of the battery level if the voltage, current and temp are read less frequently?  What is least often the state of charge can be checked to maintain accuracy?

How fast can the state of charge be updated?

How does nPM1300 handle aging batteries?  Is it possible to profile a battery at 5 stages, early life, 1/4 life, half and 3/4 life and end of battery life and generate 5 different profiles and load the right one for the application as the battery ages?  Will this accurately reflect battery aging? 

Parents
  • Hello,

    1. What is the power consumption to check battery level?

    Typical fuel gauge consumption is 8 μA to 10 μA in the active state when using a one-second fuel gauge iteration interval.

    2. How quickly at boot time is an accurate battery state of charge available?

    "The nPM1300 fuel gauge algorithm incorporates internal adjustments to correct any initialization errors resulting from an unrested battery and unexpected reset conditions. These errors typically have a minor impact, and the predictions will converge to the accurate value within a few minutes of normal operation." (docs.nordicsemi.com/.../guidelines.html)

     Power consumption can be found here: https://docs.nordicsemi.com/bundle/nan_045/page/APP/nan_045/power_consumption.html

    3.What is the RDSon in the path from VBUS or VBAT to VSYS?  (Assuming there are either internal diodes or a FET for protection)?

    Rdson (Ron) information can be found in the product spec:

    Here is the VBUS to VSYS resistance:

    4. Can the VSYS of one nPM1300 be used to power the VBUS of another nPM1300 (chained together)?

    It is possible as long as VSYS of the first PMIC complied with the VBUS input constrains: 

    5/6. Is there a trade off of accuracy of the battery level if the voltage, current and temp are read less frequently?  What is least often the state of charge can be checked to maintain accuracy?

    There is a tradeoff, but we haven't quantified this. In general, the fuel gauge iteration rate should follow application activity level. For example, if current consumption is stable (e.g. only maintaining a Bluetooth connection at a slow connection interval) it can be run less often (e.g. every 5 seconds instead of every 1 second). If activity increases, you will get more accurate results if the fuel gauge iteration rate also increases. If current consumption is very low, typically low microamp range when only RTC is running, you can use the nrf_fuel_gauge_idle_set() function to inform the fuel gauge of the expected current consumption in this state (too low to measure accurately by nPM1300), and no further iterations are required until activity resumes.

     We normally use a 1 second iteration interval during battery discharge, and 0.5 second interval when charging. You can go faster as well. This is only limited by the ADC sampling duration and TWI communication.

    7.How does nPM1300 handle aging batteries?  Is it possible to profile a battery at 5 stages, early life, 1/4 life, half and 3/4 life and end of battery life and generate 5 different profiles and load the right one for the application as the battery ages?  Will this accurately reflect battery aging? 

    We are in the process of running characterization tests for this and writing an application note. In general, the fuel gauge handle aging during discharge quite well, as it will correct for errors in the predicted battery voltage.

    As the battery ages and charge capacity decreases, you might see that the charger reports charge complete before the fuel gauge has reached 100%.

    It's possible to account for decreased battery capacity by profiling the (aged) battery again, or you can use the optional neta1 configuration parameter when initializing the fuel gauge. We haven't documented this yet, but we will. neta1 is a scaling value applied to the profiled battery capacity during charging. Setting e.g.  neta1=0.9 will make the fuel gauge treat the battery capacity as 90% of the profiled value when charging.

Reply
  • Hello,

    1. What is the power consumption to check battery level?

    Typical fuel gauge consumption is 8 μA to 10 μA in the active state when using a one-second fuel gauge iteration interval.

    2. How quickly at boot time is an accurate battery state of charge available?

    "The nPM1300 fuel gauge algorithm incorporates internal adjustments to correct any initialization errors resulting from an unrested battery and unexpected reset conditions. These errors typically have a minor impact, and the predictions will converge to the accurate value within a few minutes of normal operation." (docs.nordicsemi.com/.../guidelines.html)

     Power consumption can be found here: https://docs.nordicsemi.com/bundle/nan_045/page/APP/nan_045/power_consumption.html

    3.What is the RDSon in the path from VBUS or VBAT to VSYS?  (Assuming there are either internal diodes or a FET for protection)?

    Rdson (Ron) information can be found in the product spec:

    Here is the VBUS to VSYS resistance:

    4. Can the VSYS of one nPM1300 be used to power the VBUS of another nPM1300 (chained together)?

    It is possible as long as VSYS of the first PMIC complied with the VBUS input constrains: 

    5/6. Is there a trade off of accuracy of the battery level if the voltage, current and temp are read less frequently?  What is least often the state of charge can be checked to maintain accuracy?

    There is a tradeoff, but we haven't quantified this. In general, the fuel gauge iteration rate should follow application activity level. For example, if current consumption is stable (e.g. only maintaining a Bluetooth connection at a slow connection interval) it can be run less often (e.g. every 5 seconds instead of every 1 second). If activity increases, you will get more accurate results if the fuel gauge iteration rate also increases. If current consumption is very low, typically low microamp range when only RTC is running, you can use the nrf_fuel_gauge_idle_set() function to inform the fuel gauge of the expected current consumption in this state (too low to measure accurately by nPM1300), and no further iterations are required until activity resumes.

     We normally use a 1 second iteration interval during battery discharge, and 0.5 second interval when charging. You can go faster as well. This is only limited by the ADC sampling duration and TWI communication.

    7.How does nPM1300 handle aging batteries?  Is it possible to profile a battery at 5 stages, early life, 1/4 life, half and 3/4 life and end of battery life and generate 5 different profiles and load the right one for the application as the battery ages?  Will this accurately reflect battery aging? 

    We are in the process of running characterization tests for this and writing an application note. In general, the fuel gauge handle aging during discharge quite well, as it will correct for errors in the predicted battery voltage.

    As the battery ages and charge capacity decreases, you might see that the charger reports charge complete before the fuel gauge has reached 100%.

    It's possible to account for decreased battery capacity by profiling the (aged) battery again, or you can use the optional neta1 configuration parameter when initializing the fuel gauge. We haven't documented this yet, but we will. neta1 is a scaling value applied to the profiled battery capacity during charging. Setting e.g.  neta1=0.9 will make the fuel gauge treat the battery capacity as 90% of the profiled value when charging.

Children
No Data
Related