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.

INA293: Linearity around Vsense = 0V

Part Number: INA293
Other Parts Discussed in Thread: INA193, , INA190

Hi,

Could you please provide me the following graph around Vsense = 0V such as "Vout vs Vsense" which is mentioned on page 19 of INA193 data sheet?
If not provided, please provide the measurement data instead.

Best regards,
Kato

  • Sadanori-san,

    As I mentioned on your other post, this behavior is a characteristic of older devices in our portfolio, and the INA293 does not use this architecture. As such there, is no data of this sort to share with this device. The INA293 will operate linearly down to 5mV/20mV, RTO, typical/worst case respectively:

    This means that provided the device is operating within the required common mode range, it should exhibit a linear input/output relationship inside datasheet error epextations until hitting the swing to ground limitation listed above. 

  • Hi Carolus-san,

    Thank you for your prompt reply.

    The error will be very large when Isense is around 0A if using the calculation tool of "Current sense amplifier error analysis INA293" on TI.com
    So, should INA293 be interpreted that it will operate in the linear range around Vsense = 0V?

    Best regards,
    Kato

  • Kato-san,

    Your understanding of larger errors on the bottom portion of the curve is correct, and to an extent, this is true for all amplifiers. The answer yes, the device will operate in the linear range, but there will potentially be significant error based on the system setup.

    While ideal op amps can be driven to zero, practically all real world amplifiers are going to have a swing to ground limitation, as well as a voltage offset that will add error at smaller measurements. The massive rise in error that you are seeing here is due to that offset, and potentially bias currents. How small of a current measurement is your customer trying to make on the lower end? 

    Take for example, the 20mV output we mention above, and say you are using the A1 variant of the device, so your gain is 20V/V. Referring back to the input, this means that our shunt is only creating a 1mV Vsense across the input pins. Meanwhile, the worst case offset for the part is 150uV, which will also be present on the input. This means that in this situation, the total we are seeing across Vsense is 1mV+150uV = 1.15mV in the worst case analysis, or the offset is adding an additional ~15% error to the measurement, because the measurement is so small in regard to the offset voltage. Also, depending on how small the current your customer is trying measure, there could also be significant error being added on due to the bias current that flows into the device being more proportional to the signal.

    Can you share the full scale range your customer is trying to measure here?

  • Hi Carolus-san,

    Thank you for the additional information.

    The full scale current range is 1mA, so the value of Rsense is 124ohm for INA293A1. Our customer is considering detecting 0A as the minimum current.
    Under the above conditions, please tell me the minimum current that can be detected accurately.

    Best regards,
    Kato

  • Kato-san,

    For a current range this small, the bias currents are going to play a huge role in the ability to sense accurately. With a worst case bias current of 30uA, there will be 3% error MINIMUM at the top end of the scale due to the bias current alone. This does not take into account the offset, gain error, etc. As the customer attempts to measure closer to zero, this would continue to increase. 

    I would recommend you look at INA190 for this application. Alternatively, you could also look at our instrumentation amplifiers. These devices have much smaller bias currents that allow for larger shunt resistances to be used and extend the full scale range of the device. 

  • Hi Carolus-san,

    Thank you for your quick response.

    I will discuss this issue with our customer. So, I will contact you if I get additional questions from him.

    Best regards,
    Kato