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.

INA219: Looking for advice on a more accurate, lower current, lower voltage variant similar to INA219

Part Number: INA219
Other Parts Discussed in Thread: INA226, INA233, INA228, INA238

Hi -

On our previous board design, we used 4 INA219 devices to measure voltage and current from the 4 voltages that supply our chip.  These were satisfactory in how they performed, but the limits in bandwidth and accuracy leaves us trying to find something more targeted and a step better for our next board design.

The requirements we have for a voltage/current measurement device are:

  1. Our SoC is a very low power part where we want accurate real time measurements in the milliwatt range.
  2. We want to accurately measure supply voltages between 0.5V and 3.6V that supply our SoC on our board.
  3. We want to accurately measure the currents from these supplies in the range of 10's of microamps to 100's of milliamps.
  4. We prefer digital output device where we can read measurements (I2C or SPI).
  5. We need a High-Side measurement device.
  6. Common Mode Voltage is approximately the bus voltage.
  7. Generally, the current is uni-directional.  These are positive DC supplies feeding our SOC, but they are switching supplies.  

The INA219's check most of the boxes above, but they have the following limitations that we'd like to address:

  1. The bandwidth of measurement is ~2 kHz.  So we'd like something with a higher bandwidth so that we're not limited to a 500 micro-second sample window.  We tend to not detect dynamic voltage/current changes that occur because of the limited bandwidth.
  2. The voltage range of the INA219 is 16V/32V.  This is much higher than we need, so ideally we'd have a device in the 8V range where we might gain some resolution.
  3. When reading in "Shunt and bus, continuous" mode, the INA219 readings require alternating the ADC between each shunt and bus measurement.  This is effectively blinding the ADC from current or voltage activity 50% of the time.  
  4. It would be ideal if the device provided some level of buffering/FIFO of readings so it wasn't limited to the I2C ability to pick up each discrete reading.  Since I2C has some overhead for each reading, the payload of the data read has significant overhead.  This becomes a limitation when having multiple INA219 devices in the circuit because it takes time to get back to each device with I2C speeds.  Averaging of the results does help, but also dilutes the results.

Our question is whether TI has an offering of a device with similar features as the INA219, but addresses some of these accuracy and bandwidth limitations.

Thanks for your guidance.

Tom

  • Hi Tom,

    Since the introduction of INA219, there have been some major milestones in our digital power monitor technology. You’ll see improvement in speed and accuracy.

    To start with, you may look into INA226 or INA233. These two are similar in terms electrical specifications though INA233 is newer.

    If you’re looking for even higher performance, INA238 (16 bit) and INA228 (20 bit) are our latest high voltage devices with high resolution and fastest conversion time @50uS.

    Regards, Guang