• State Resolved
  • Locked Locked
  • Replies 11 replies
  • Answers 3 answers
  • Subscribers 47 subscribers
  • Views 1474 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.

INA226: INA226AIDGST

Bhadresh Patel
Prodigy 90 points
Part Number: INA226
Other Parts Discussed in Thread: INA228, SYSCONFIG

Hi

We are using the INA226AIDGST  power monitor IC in our design and we are reading power and current register value which are not satisfactory so find below attached image as well as our max current and voltage table.

suggest how to slove this problem.

R3507 Current sense value is 0.001E and INA226 communicate through I2C with FPGA Board.

Voltage and current table information.

Below are the result which we have getting during testing on our board.

All result in hexadecimal.

Power LSB Value  as per below one

Below are the result which we have getting during testing on our board.

All result in hexadecimal.

  • 0
    TI__Mastermind 26395 points

    Hello Bhadresh,

    Thank you for using the TI forum. If I do the register conversions for the data you sent me, I get:

    Shunt voltage register: 9 * 2.5 µV = 22.5 µV 

    Bus voltage register: 0x3BE in decimal is 958: 958 * 1.25mV = 1.1975 V

    Current register: 0x44 to decimal is 68: 68 * 0.0003299865723 = 22.44 mA

    If we look at the expected current, we get:

    I = V/R = 22.5 µV / 0.001 = 22.5 mA.

    So, from this, it looks like everything is working fine, according to the measured voltage at the input pins. However, that being said, with your specific setup, the expected error with <1A measurement will be very high.  For example, here are your settings put into the error calculator tool on the INA226 webpage:

    This error can be reduced by better utilizing the full scale range of the INA.  For example, right now with a 1mΩ shunt, your maximum current is measured at 10.813mV, but the sensor has a full scale range up to 81.92mV, so if you used a shunt closer to 7.5mΩ then you would be using 81.0975mV at full scale, which is most of the available range.  This would change the error to:

    This is much better, but still not great at lower currents. If you need higher accuracy, you could switch to the INA228 (which is pin compatible with the INA226). With your same shunt value, you would have this as your error (using the calculator on the INA228 product page):

    But, if you changed your shunt to 15mΩ to utilize 162.195mV of the 163.84 mV range, then your error would be: 

    Which is significantly better. 

    Besides those comments, I also would like to note that you have used input filter resistors in your design (R3506 and R3508), but not an input filter capacitor (C_FILTER on the left, from the INA226 datasheet):

    Without the capacitor, those resistors won't add much benefit, so I would recommend either adding the capacitor, or removing the resistors (or replacing with 0Ω).

  • 0 in reply to Mitch M
    TI__Mastermind 26395 points

    One more side note, if you switch to the INA228, you can use SysConfig to help with your software update:

    https://dev.ti.com/sysconfig/index.html?product=ascstudio&module=/ti/sensors/currentsensor/INA228

  • 0 in reply to Mitch M
    Prodigy 90 points

    Hi 

    We are use this IC multiple places but shunt register value reading is different but we have used 0.001E in all IC so why its reading different hexadecimal value ??

    check below one for more refrence

    All shunt register value different

  • 0 in reply to Bhadresh Patel
    TI__Mastermind 26395 points

    Hello Bhadresh,

    This also has to do with the accuracy issue I mentioned above.  With your specific settings and a low current, you can have 20-70% error, so the values will all be rather different.  It looks like they are sitting close to 0A with most just positive and one slightly negative. 

  • 0 in reply to Mitch M
    Prodigy 90 points

    Thanks, Mitch,

    I want to know one more thing regarding design connection like we have connected IN+ and IN- pin of INA226 connected with shunt resistor as per attached image so can you tell me is it okay or we need to reverse as per the application circuit.

    In design IN+ PIN connected with 2nd number pin of the resistor and IN+ connected with 1st pin of shunt resistor so find the attached image and let us know if any correction requires.

  • 0 in reply to Bhadresh Patel
    TI__Mastermind 26395 points

    Hello Bhadresh,

    I'm having a hard time finding a schematic/pinout for that resistor, but in general the IN+ and IN- should connect to the smaller pins of a 4 pin resistor. You can look at our shunt resistor layout video, which discusses 4 pin resistors near the end of the video: https://training.ti.com/ti-precision-labs-current-sense-amplifiers-shunt-resistor-layout

  • 0 in reply to Mitch M
    Prodigy 90 points
    Mitch M said:
    2.44 mA

    Hi Mitch,

    Our max current is 10.813A but it's shown only in 22.44mA so we are confused why it's showing as this can you elaborate on this ??

    If any shunt resistor related things it's not showing this much error please suggest on thisis

  • 0 in reply to Bhadresh Patel
    TI__Mastermind 26395 points

    Hello Bhadresh,

    The 22.44mA was the current for the specific register set you gave, not the max current.  If you would like to see an example that uses the maximum current, then feel free to send all register values for the use case when 10.813A (or similar) is being used, and I will walk you through the calculations. 

  • 0 in reply to Mitch M
    Prodigy 90 points

    Thanks Mitch,

    Voltage = 1.2V 

    Max Current = 10.813Amp

    Shunt Resistor = 0.001E

    According to the above table, the current LSB are calculated as per the given formula:  Current_LSB = Maximum expected current/2^15 

    For VCCHBM_1V2   => current_LSB = 10.813 /2^15 => 0.000329986572265625 (A)

    The calibration values are calculated as per the given formula:  

    CAL = 0.00512 / (Current_LSB * Rshunt)

    For VCCHBM_1V2   => CAL  = 0.00512 / (0.000329986572265625 *0.001)  =>15,515.782853972070655692222324979 = 15516 = 0x3c9c

    Above is our calibration setting but we are not getting it correct result.

    Below one we have got it in our design.

    Schematic

    Also, we are getting the wrong shunt voltage value as of now it's showing 9*2.5uV = 22.5uV but its not match our current requirement.

    can you suggest which is the best register setting for our case??

  • 0 in reply to Bhadresh Patel
    TI__Mastermind 26395 points

    Hello Bhadresh,

    What current value were you trying to send over the shunt for that use case? Those numbers are for the same example that I went over above.  From those values, the current register is the correct value for the measured shunt voltage. 

    Also, as I stated before:

    Mitch M said:
    With your specific settings and a low current, you can have 20-70% error

    The INA226 is not going to be able to get a very accurate measurement for your specific setup at low current.  You are really better off switching to the INA228. 

  • 0 in reply to Mitch M
    TI__Mastermind 26395 points

    Also, for your schematic, I see that you have input resistors (R3706 and R3705), but no input filter capacitor.  I recommend either adding a cap, per Figure 21 of the datasheet, or removing the resistors (ie, change to 0Ω).