• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 47 subscribers
  • Views 1086 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.

AMC1301-Q1: How to connect two parallel sense resistors.

Ron Michallick
Guru**** 165316 points
Part Number: AMC1301-Q1
Other Parts Discussed in Thread: AMC1301

Due to high current, the application will use two equal sense resistors. I need to measure both resistor voltages as the current split might be 49%/51% (or similar)

Using just one LMC1301, what would be the correct way to sum (average) the two nearly equal sense resistor voltages?

  • 0
    TI__Guru**** 181021 points
    Hi Ron,

    I've seen what you are working on and I thought there were two AMC1301 devices, one for each shunt. If it truly is two shunts in parallel to one AMC1301, there is nothing you can really do ensure exactly 50/50 current flow through the two resistors. Slight differences in mounting, resistance of the shunt, etc are going to be limiting factors. You can start with the layout example in the back of the AMC1301 datasheet. Picture there being a second shunt in parallel with the one shown there and then route the AMC1301 inputs like a differential pair from the middle of the two shunts.
  • 0 in reply to Tom Hendrick
    TI__Guru**** 165316 points
    Tom,

    I was thinking of sense tapping the resistor(s) just like the data sheet says. Then I would average each sense tap pair by passing them through a series resistor then to the AMC301. The question is how low in value do the resistors need to be to not adversely affect the accuracy of AMC1301 gain. Knowing the input impedance of the AMC1301 would help, so I could use series resistor value lower than 1% or better yet 0.1% of the AMC1301 input Z.
  • 0 in reply to Ron Michallick
    TI__Guru**** 165316 points

    Tom,

    This is my solution for two Rs resistors. It is simulated with imperfect splits and grounds.

    The 47 ohms resistors add 23.5 ohms to the input which will be negligible compared to the AMC1301 input impedance.

  • 0 in reply to Ron Michallick
    TI__Guru**** 181021 points

    Hi Ron,

    In theory, this should work.  In the end, I suppose it really depends on how the FET's and gate drivers are setup.  If they are on separate boards, you may not have any choice.

  • 0 in reply to Tom Hendrick
    TI__Guru**** 165316 points
    Tom,

    They will be close (on same board) as the only reason to have two resistors is the power dissipation limit of the sense resistors. It would be a terrible idea to do this if current split was worse than 40/60 or common mode difference was more than (roughly) 100 mV. Both problems would occur if not carefully laid out because current in this design could reach 1000 Amps. Issue is resolved.