• State Resolved
  • Locked Locked
  • Replies 10 replies
  • Answers 1 answer
  • Subscribers 49 subscribers
  • Views 1810 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.

Suggest current sensing IC part number with integrated ADC

Sathyasai Prasad
Prodigy 180 points
Other Parts Discussed in Thread: ADS1118, ADS1220, INA226, INA190, INA219

Hi,

I want to measure the current across voltage rails which are mentioned below.
Rail 1: 1V @ Max:2.5A
Rail 2: 1.2V @ Max:1A
Rail 3: 1.8V @ Max:0.9A
Rail 4: 3.3V @ Max:0.05A

So I am using 1-4milli ohm resistor(1206 package) for measuring current because of less drop.
So please suggest current sensing IC part number which can sense minimum of 1mA or less and maximum current ratings as mentioned. It will be better if the IC should have integrated ADC with high resolution and more number of channel if possible.

Thanks in Advance.

Regards,

Sathyasai

  • 0
    TI__Guru* 94415 points
    Sathyasai,


    Measuring this level of current requires low noise and a lot of precision. If you have a 4milli-ohm resistor and are measuring a 1mA current, that's an input voltage of 4uV to resolve.

    At first, the device I was thinking of was the ADS1118. However, at the highest gain setting, the input range is ±0.256. With 16-bits, the LSB is 7.8uV. However, the LSB size seems like it would be too large for your measurement.

    Instead, it would be better to use the ADS1220 with the gain of 4, but with the PGA bypassed (gain is achieved through the capacitive scaling when the PGA is disabled). The PGA needs to be bypassed so that the input can be in the input range of the ADC. If you were measuring a high-side or low-side shunt resistor, this wouldn't be in the input range of the PGA. With a gain of 4, bypassing the PGA, the LSB size using a 2.048V reference would be 0.06uV. This would give you enough resolution, as long as the noise isn't too high.

    For the ADC, what data rate are you looking for? What resolution do you need? Do you need any other features from the ADC? There may be other alternatives (like including a current sense amplifier for gain and level shifting the input to be in the PGA input range). Whatever your system requires, it would give more information to help in a recommendation.


    Joseph Wu
  • 0 in reply to Joseph Wu
    Prodigy 180 points
    Hi Joseph,

    Thanks for suggesting ADC part numbers but I am looking for current sensor IC with integrated ADC.

    To measure 1mA current with 2-4mohm resistor, ADC LSB size should be 2-4uV.
    ADS1118 is nearly matching to our requirements but still we require accuracy which will come to 2-4uV. It will be better if ADC resolution will be >16bits and remaining parameters are not a problem.
  • 0 in reply to Sathyasai Prasad
    TI__Guru* 94415 points
    Sathyasai,


    Sorry I didn't get back to you earlier, but we've been discussing your application behind the scenes.

    Have you looked at the INA226? I haven't really looked at the noise performance, but it was one of the devices that came up in our discussions.


    Joseph Wu
  • 0 in reply to Joseph Wu
    TI__Genius 16200 points
    Sathyasai,

    The issue is trying to measure such low currents with such a low shunt value is just not really possible. As Joseph calculated, at 1mA and 4mohm, the VSENSE is only 4uV. Our lowest offset current monitor (current sense amp with integrated ADC) is the INA226 , but it's worst case room temp offset is 10uV which means a 250% error at your minimum current, ignoring the effect of error due to bias current. Normally when trying to measure currents that low, we would recommend looking at our INA190 or an instrumentation amp to address the bias current errors. But with your small shunt, none of these will work as the VSENSE will be less than VOFFSET. Raising the RSHUNT value to get an acceptable VSENSE/VOFFSET ratio is my recommendation.
  • 0 in reply to DHarmon
    Prodigy 180 points
    Hi Dan,

    Thanks for the suggestion. INA226 is looks fine for me except offset voltage but in INA190 does not have integrated ADC which is not possible to use in our board.
    I have looked into INA219 which is having zero drift meeting all our requirements but having 12bits resolution only. So please suggest IC which are having zero drift like INA219.
  • 0 in reply to Sathyasai Prasad
    TI__Genius 16200 points

    Actually all of the devices listed are part of our zero-drift architecture portfolio - the INA226 is actually a much more accurate solution than the older INA219:

  • 0 in reply to DHarmon
    Prodigy 180 points
    Hi Dan,

    We planned to use INA226 only because of more resolution. Thanks for suggesting.
  • 0 in reply to DHarmon
    Prodigy 180 points

    Hi Dan,

    We have used INA226 in our design as you suggested. I am trying to measure the current for 1volt with 2mohm resistor.

    I have taken care pull up resistors for I2C lines.

    So please review my schematic and let me know if any anything needs to be taken care. Please observe schematic which is shown below.

    1.Whether VBUS should be connected to the supply side or load side?

  • 0 in reply to Sathyasai Prasad
    TI__Genius 17100 points
    Hi Sathyasai,

    The VBUS pin should be connected to the supply side where Vin+ is connected as you have it in this schematic.

    Overall the system looks good, but the resistor divider on A0 pin may not be necessary. How will it be used? Ensure to use pull up resistors (1kΩ - 10kΩ) to the supply for SCL and SCA lines in series with the 0Ω ones.


    My apologies for the delay on this response! Hopefully this is helpful; let us know if we can help with your project.


    Thank you for your time,
    Manuel Chavez
  • 0 in reply to Manuel Chavez
    Prodigy 180 points
    Hi Manuel,
    Thanks for the suggestion.
    We have provided pull up(not mounted) and pull down resistor option for A0 pin because in future we can change the I2C address if required. This is not resistor divider network.
    We have provided pull up resistors for two I2C lines.

    Thanks & Regards,
    Sathyasai