• State TI Thinks Resolved
  • Replies 5 replies
  • Answers 3 answers
  • Subscribers 46 subscribers
  • Views 120 views
  • Users 0 members are here
Support feedback
Options
Options
Related

INA226: Whether a uA range current value can be sensed by INA226

Wonseok Heo
Prodigy 20 points
Part Number: INA226
Other Parts Discussed in Thread: INA236,

Dear TI teams

I would like to use the INA226 for current sening and I'm wondering if it can actually sense small currents in the uA range.

The currents we wish to sense are as follows:

I = 0uA, 0.0625uA, 0.125uA, 0.1875uA, 0.25uA, 0.3125uA, ..., 15.8125uA, 15.875uA, 15.9375uA, 16uA

(Range of I = 0~16uA, 256 levels (8bit resolution), which means 1 level per 0.0625uA)

According to the datasheet, the shunt voltage input range is -81.9175mV~81.92mV, the LSB step size is 2.5uV, and the maximum shunt offset voltage is 10uV.

Considering the current value and the datasheet, if I use a 1600 Ohm shunt resistor,

  1. The shunt voltage input range would be 16uA * 1600 = 25.6mV (which falls within the 81.92mV range)
  2.  The voltage interval between each level would be 0.0625uA * 1600 = 100uV (which is larger than the 2.5uV LSB step size)
  3.  The offset current is 10uV / 1600 = 0.00625uA (which is expected to result in a 10% error compared to 0.0625uA (the gap between each level))

Based on these results, it appears the desired current value can be sensed and converted to digital value using the INA226.

But I would like to know whether the shunt resistor value is correctly set, whether the current can be actually sensed in the uA range, and whether levels in 0.0625uA increments can be distinguished.

 

Best regards
Wonseok Heo
 
image.png
  • 0
    TI__Mastermind 34870 points

    Hi Wonseok,

    The shunt resistor should work. It is probably a good idea to use the longest conversion time and largest number of averaging to minimize the impact from noise.

    Please note that this device was only recently updated to have low input bias current, making it possible to measure uA current.

    However some of the old rev material might still be lingering in the distribution channel. The old rev has an input bias current on the order of 10uA.

    There is not an easy way to distinguish the two visually, other than reading the dieID register (FFh). The old rev returns 2260h, while the new returns 2261h. There are some mistakes in the description of this register, we’re in the process of updating the datasheet to correct them.

    Regards

    Guang

  • 0 in reply to Guang Zhou
    Prodigy 20 points

    Hello Guang
    Thank you for the reply.
    First of all, I understood that only the new rev can measure uA current.


    I have two additional questions.

    First, is there any way for me to choose between the old rev and the new rev when making a purchase?


    Second, if I purchase the new rev, can the current level be distinguished in units of 0.0625uA (62.5 nA), rather than 1uA? If not, can the current level be distinguished in units of 0.25uA or 0.125uA?

    Best Regards,
    Wonseok

  • 0 in reply to Wonseok Heo
    TI__Mastermind 34870 points

    Hi Wonseok,

    Currently there is no way to specify new rev when you put in an order. Conceivably, after some time when the distribution channel is fully flushed, all that is available is the new rev. What is your project’s timeline? If you need samples, we can supply you with some, and we can make sure the samples come from new rev because the order will be handled manually.

    An alternative device is INA236, which is very similar to INA236 feature-wise. If you choose INA236, there is no rev issue, it is a lot more straightforward.

    For the capability question, yes you’ll be able to achieve 62.5nA resolution. Because you’re using a fairly large shunt, resulting in a sufficiently correspondingly large shunt voltage step size

    Regards

    Guang

  • 0 in reply to Guang Zhou
    Prodigy 20 points

    Hi Guang,
    Thank you for the reply.

    Then if I use INA236 instead, can the INA236 meet the 62.5nA resolution just like the INA226?
    If so, I will use the INA236.

    Regarding the project, we first plan to test using the INA236EVM (or INA226EVM), and if the conditions are met, we will purchase additional INA236 (or INA226) chips.

    If I purchase the INA226EVM, will the same rev issue still exist as INA226?
    Also, I would like to know if the INA236EVM is free of any rev issues.

    Best Regards,
    Wonseok

  • 0 in reply to Wonseok Heo
    TI__Mastermind 34870 points

    Hi Wonseok,

    The has been only one (original) rev for INA236. Your calculations apply to INA236.

    Regards

    Guang