• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 49 subscribers
  • Views 355 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.

INA226: Bi-Directional current sense using INA226 with DRV595DAP driving TECs

Mike Garretson
Prodigy 30 points
Part Number: INA226
Other Parts Discussed in Thread: DRV595, , INA260, INA228, INA229, INA237, INA238, INA239

Hi-

We are driving two TECs in series with a DRV595, highest Vbus will be 24V, largest current will be 4A. We'd like to use the INA226 to sense the current in both directions for when the TECs are heating or cooling. There is a 10uH inductor in series in each leg with OUTP and OUTN to the TECs. Wondering if you have a suggestion on where to put Rsense, and what connections to make to the INA226 for Vbus sensing, as the polarity of the voltage will switch with the current? Should Rsense be from pins 22 and 28 of the DRV595 (grounds for the outputs)? In this case, we wouldn't be able to verify heating or cooling by the direction of the current...
If we can't read the Vbus, that is acceptable in this case. Should the Vbus input be tied to either Vss or Gnd, or NC?
Or perhaps we should go with something like the INA260 if we can't read the current direction? 

Thank you

  • 0
    TI__Genius 17990 points

    Hello Mike,

    Thank you for using the TI forum. Using Figure 13 from the DRV595 datasheet as a schematic reference, I would recommend putting Rsense in between the inductor and the TEC device (for example, on OUT+ of the schematic). The INA226 is bidirectional, so it will be able to sense current in either direction in this location.  The reason I recommend putting it here instead of in between the inductor and and the DRV595 is because when the current direction changes it is possible that there are temporary voltage spikes, and you need to make sure that VCM does not go above or below the absolute maximum ratings of the device (in this case we are more worried about negative voltage, as you have some padding on the high end).  With Rsense in this location, it will be slightly separated from GND (and PVCC) by the inductor, and will also have the capacitor (example, C11) to help stabilize the voltage. 

    As you noted, you there is not a good place to measure Vbus.  Since you said this will work for you, I would recommend just connecting that pin to Vss. 

    The INA260 is actually also bidirectional, so you could use this device if you prefer.  On that note, you could even use some of our newer digital current sensors if you prefer. These are the INA228 (20 bit, I2C), INA229 (20 bit, SPI), INA237 (16 bit, I2C), INA238 (16 bit, I2C), and INA239 (16 bit, SPI).  The difference between the INA237 and the INA238 is that the INA37 is less accurate. One advantage about using these newer devices (besides being more accurate), is that they can handle a -0.3V on VCM.  This could help if you have voltage spikes that drive the voltage just below 0V.

  • 0 in reply to Mitch M
    Prodigy 30 points

    Hi Mitch,

    Thank you for the fast reply. The location between the inductor and TEC off OUT+ is where I had originally placed the sense resistor, so I'll leave it there. I'll look into the newer options that you mention also. Appreciate the feedback.

    Best Regards,

    Mike