• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 45 subscribers
  • Views 1607 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

ADS1220: vs ADS124s08 regarding power consumption

Scott Miller1
Intellectual 515 points
Part Number: ADS1220
Other Parts Discussed in Thread: ADS124S08

I am comparing ADS1220 and ADS124s08.  Obviously the latter has a more extensive feature set, but notwithstanding, I want to understand the power usage of these two devices.  

In my application, I plan to take an ADC reading (and maybe a temperature reading from an external diode, which requires changing the PGA setting....), then falling asleep for x seconds (from 1 to ??? seconds).

Can anyone on this forum give a rough estimate of how much current is consumed during the waking portion, and sleeping portion of this cycle, so I can make some meaningful calculation on battery life?  Of course I will need to add the microprocessor in the mix (MSP430, or Bluetooth micro in a module), but I'm just talking about the AtoD in this case.  

I plan to power a 5k bridge with the internal reference, current mode (at the lowest uA that I can get a good reading), PGA at 16 to 64 (depending on bridge sensitivity).  

The ADS1220 does not give very much information about current consumption (unlike the ADS124s08 datasheet). It advertises 120uA in duty cycle mode (is this average current through a duty cycle??), where the duty cycle sample rate is 5SPS minimum.  I don't need 5SPS in my particular application.  So I will be probably manually putting the 1220 asleep, in the same fashion as the 124s08.  The 124s08 advertises "Low Power Consumption: As Low as 280 µA", but I can't find in the data sheet which mode this is in or the conditions surrounding it.

I am really looking for two answers, with the first being somewhat vague:

1) relative current, as in "the 124S08 will consume half the power of the ADS1220" in the mode described above.  

2) a ballpark estimate of consumption (for each ADC) during sleep and wake functions, as well as how long it will be in each mode (i.e. it takes x ms to wake up, take a reading from a differential pin pair, switch the PGA to a different value, take readings from a second pair of pins, switch back to the original pin pair (prep for next cycle), then fall asleep for x ms) within a one second cycle.  

Thanks in advance!!

  • 0
    TI__Guru** 113765 points

    Hi Scott,

    If you look at page 8 of the ADS1220 datasheet there is a table showing current consumption for the various modes in the Power Supply section.  Duty-cycle will not be practical in your situation.  This mode is the same as normal mode except that one conversion is taken then sits idle for 3 conversion periods essentially being 1/4 the normal mode data rate in continuous conversion mode.  The current given for duty-cycle is for the on time and startup overhead. In general, the faster the operation the higher the current.  The larger the gain the higher the current.  So in the end it does make a difference for the actual settings you use such as gain of 16 versus gain of 64.  Not a substantial difference for the ADS124S08 (35uA), but a large difference for the ADS1220 (170uA).

    As to the 280uA current operation for the ADS124S08, this would be analog current running with PGA bypassed (85uA) and digital running off of external clock (195uA).  This information is provided on page 12 of the datasheet.

    At a gain of 32, internal reference on, IDAC on, and external reference input you will see roughly 1.5 times greater analog current for the ADS1220, but nearly 3 times increase for the ADS124S08 over the ADS1220 for digital current.  So in this particular scenario they are roughly equal.  However, there are ways with both devices where power can be minimized by various means. 

    Power-down current is given in each of the datasheets within the tables. There will be some differences between devices on startup from the power-down state, but this time will be very small compared to the communication.  One other consideration no matter which device is chosen relates to analog settling when powering up the bridge and reference. 

    Getting a rough idea of current consumption would be taking your device on time currents for analog and digital for the period the device is on, and the off time current for when the device is off.  To know a more realistic consumption requires more specific details, such as SCLK speed, specific commands issued, data rates and other register configuration as well as what else is happening during the on time. So all of the things mentioned in (2) will require adding up all the operations you intend to conduct, plus any analog settling (which depends on your circuit implementation) for your on time.

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Intellectual 515 points
    Two follow-up questions:
    #1 What affects the Analog Power Down mode (variance of 0.1 to 1.5uA) of ADS124S08?

    #2 As a general rule with either of these two devices....and I think I know the answer to this.... assuming a 1 second wake/sleep interval, is it generally better to:

    A) put the device in sleep mode after taking the reading, or
    B) removing power to the device altogether?

    What about a 2 or 10 second interval?

    I.e. does the power consumption required for re-configuring the AtoD and needed settling time in scenario B greater or less than the sleep mode power consumption?

    Here are my settings: 5k bridge connected to PGA between 16 to 64, powered by IDAC set to some constant current (TBD). 20SPS. Diode (to measure temperature) connected Analog inputs, IDAC sourcing current @ 10uA. User internal clock of AtoD.

    Normal operations after Power up (or wake up). After settling time and configuring registers, take 5 samples using 10MHz SPI bus @20SPS. Change Current source uA value and output pin, change input pins and PGA setting, take reading from Diode and transmit to SPI. Go to sleep (or be externally powered down).

    Here's my analysis: Since settling time of the internal reference with 1uF cap is 6ms, turn that on first. Then perform all the configuration of the AtoD during that time (write INPMUX, PGA, DATARATE (or just keep defaults), etc. Near then end of the internal reference settling time, enable the PGA (for its 190us "startup" time). After the 6ms internal reference settling time, enable the IDAC (write to REF, IDACMAG, IDACMUX, registers) and wait for its 22us settling time.

    (My) conclusion: The current consumption during the active state is effectively the same (whether coming from POWERDOWN ("sleep") mode or from physically being powered down, because the settling of the internal reference, PGA, and IDAC is the same in either case, and the configuration needed from a hard power off can be done during that time. So the difference in current draw becomes the difference between the powered off state (0uA) and the current draw during the POWERDOWN mode (0.1 to 1.5uA IAVDD plus 0.1uA IDVDD).

    Do you agree with the conclusion?
  • 0 in reply to Scott Miller1
    TI__Guru** 113765 points

    Hi Scott,

    See my comments below.

    Best regards,

    Bob B

    Scott Miller1 said:
    Two follow-up questions:
    #1 What affects the Analog Power Down mode (variance of 0.1 to 1.5uA) of ADS124S08? [Bob] As stated at the top of the Electrical Characteristics table, the typical is at 25 deg C.  The min/max values related to the full operating temperature of the device.
     
    #2 As a general rule with either of these two devices....and I think I know the answer to this.... assuming a 1 second wake/sleep interval, is it generally better to:

    A) put the device in sleep mode after taking the reading, or
    B) removing power to the device altogether? [Bob] When the device is completely powered down, you must wait for device startup and account for power used when writing the configuration. For the ADS124S08, you will also need to wait for the internal reference to settle if the internal reference is used.  Also, completely powering down the device will require any bypass caps in the circuit to also charge, which may also consume additional current.  Usually for this short of off time, using the powerdown feature of the device is actually better.
     
    What about a 2 or 10 second interval? [Bob] For the ADS124S08 it may make more sense as it will burn more power, but I can't tell you where the sweet spot is regarding what time interval will be making it worthwhile.  For the ADS1220, I know of a customer that initially was powering off and on the device but later switched to just powerdown as the overall consumption was lower due to the factors mentioned above.
     
    I.e. does the power consumption required for re-configuring the AtoD and needed settling time in scenario B greater or less than the sleep mode power consumption? [Bob] See my response above.
     
    Here are my settings: 5k bridge connected to PGA between 16 to 64, powered by IDAC set to some constant current (TBD). 20SPS. Diode (to measure temperature) connected Analog inputs, IDAC sourcing current @ 10uA. User internal clock of AtoD.

    Normal operations after Power up (or wake up). After settling time and configuring registers, take 5 samples using 10MHz SPI bus @20SPS. Change Current source uA value and output pin, change input pins and PGA setting, take reading from Diode and transmit to SPI. Go to sleep (or be externally powered down).

    Here's my analysis: Since settling time of the internal reference with 1uF cap is 6ms, turn that on first. Then perform all the configuration of the AtoD during that time (write INPMUX, PGA, DATARATE (or just keep defaults), etc. Near then end of the internal reference settling time, enable the PGA (for its 190us "startup" time). After the 6ms internal reference settling time, enable the IDAC (write to REF, IDACMAG, IDACMUX, registers) and wait for its 22us settling time.

    (My) conclusion: The current consumption during the active state is effectively the same (whether coming from POWERDOWN ("sleep") mode or from physically being powered down, because the settling of the internal reference, PGA, and IDAC is the same in either case, and the configuration needed from a hard power off can be done during that time. So the difference in current draw becomes the difference between the powered off state (0uA) and the current draw during the POWERDOWN mode (0.1 to 1.5uA IAVDD plus 0.1uA IDVDD).

    Do you agree with the conclusion? [Bob]  Mostly, however, configuration settings will hold in powerdown and will not require rewriting the registers.