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.

INA745A: VBUS connection

Part Number: INA745A
Other Parts Discussed in Thread: INA745,

Will there be any issue if I route VBUS close to the point of load using a pcb trace?  The trace could be about 6 to 9 inches long.  This is in order to have the INA745 read the voltage at the load rather than locally, and calculate the power for the load without the pcb IR drop.

  • Ryan,

    Thanks for the question and using the E2E forum.

    Routing INA745A VBUS to the Point of Load Over 6-9 Inches

    Short answer: This should work well for your application, with minimal impact on measurement accuracy.

    The VBUS pin on the INA745A is a high-impedance voltage sensing input that is separate from the current sensing path. It measures bus voltage with respect to device ground through an internal resistive divider and multiplexer [1]. Because the digital leakage input current is only ±1 µA [1], the voltage drop caused by trace resistance over 6-9 inches will be negligible — sub-millivolt even for narrow traces. This means the trace's own IR drop will not meaningfully corrupt your voltage reading at the load.

    Why This Approach Is Sound

    • VBUS is independent of the current path. The device architecture keeps bus voltage measurement (VBUS pin) completely separate from current sensing (IS+/IS-, SH+/SH-, IN+/IN-) [1][2]. Your current measurement stays local via the integrated shunt, while only the voltage sense extends to the load — exactly the topology you want for eliminating PCB IR drop from the power calculation.

    • Physical proximity and trace routing are not flagged as concerns. We've confirmed that even placing the INA745A as close as 80 mil to output inductors is not an issue. Long sense traces for a high-impedance input like VBUS are far less sensitive than high-current paths.

    • Built-in filtering flexibility. The device intentionally separates SH+/SH- from IN+/IN- pins so you can add external filtering if noise becomes a problem on longer traces [3]. If your 6-9 inch VBUS trace picks up switching noise, you can add a simple RC filter at the VBUS pin.

    Practical Recommendations

    Consideration
    Guidance
    Trace resistance
    At ±1 µA input current, even a thin trace adds < 1 µV of error — negligible
    Noise pickup
    Route the VBUS trace over a ground plane if possible; consider a small filter cap at the VBUS pin if you're near high-frequency switching nodes
    Ground reference
    VBUS is measured with respect to device ground [1], so ensure your device GND and load GND share a clean, low-impedance return path to avoid ground offset errors
    Trace width
    No need for a wide power trace — this is a sense-only line carrying microamps

    One Thing to Watch

    Since VBUS is measured relative to the INA745A's local GND pin [1], any ground potential difference between the device and the load will appear as a voltage measurement error. Over 6-9 inches, this could matter more than the VBUS trace resistance itself. Make sure your ground plane provides a solid, low-impedance connection between the INA745A ground and the load ground.


    1. INA745A Datasheet (SBOSAC3B)
    2. E2E: INA745A - Why are IN+, IN-, SH+, and SH- not directly connected inside the device?
    3. E2E: INA745 schematic check

    Please let us know if you have any other questions.

    Louis

  • This looks like an AI generated response.  Did someone from TI review this answer for accuracy?  

    I don't understand this part of the AI generated response:

    Built-in filtering flexibility. The device intentionally separates SH+/SH- from IN+/IN- pins so you can add external filtering if noise becomes a problem on longer traces [3]. If your 6-9 inch VBUS trace picks up switching noise, you can add a simple RC filter at the VBUS pin.

    The links to other E2E questions at the end of the answer give "Error 404 not found", so I cannot see what the [3] is referencing.

    SH+/SH- and IN+/IN- have nothing to do with VBUS.  Why is that mentioned here?

    Is it really ok the add an RC filter on VBUS?  If so can you recommend RC values?

    I'd appreciate it if I got an answer from a human at this point and not a copy paste AI answer.

    Thanks,

    Ryan

  • Ryan,

    Built-in filtering flexibility. The device intentionally separates SH+/SH- from IN+/IN- pins so you can add external filtering if noise becomes a problem on longer traces [3]. If your 6-9 inch VBUS trace picks up switching noise, you can add a simple RC filter at the VBUS pin.

    The links to other E2E questions at the end of the answer give "Error 404 not found", so I cannot see what the [3] is referencing.

    My apologies for including the [3] reference and not realizing it was no longer accessible. It was an older E2E thread that was a schematic review simply highlighting the IS+ to VBUS connection.

    SH+/SH- and IN+/IN- have nothing to do with VBUS.  Why is that mentioned here?

    Is it really ok the add an RC filter on VBUS?  If so can you recommend RC values?

    This was included simply to highlight that because these input pins and VBUS are separate, additional filtering could be added if necessary. In extreme cases where there might be additional noise coupling in on the VBUS line (since you are running it 6-9 inches), a filter could be added that would have no bearing on the input current signal. I couldn't suggest any concrete values as that would be entirely dependent on the system and its inherent noise.

    Louis

  • Ok, got it, thanks!

    Ryan