• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 62 subscribers
  • Views 317 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.

LM5066EVK: Wrong measurement values for IIN, PIN

Sean Durkin
Prodigy 40 points
Part Number: LM5066EVK


Hi,

I am probably missing something obvious here:

I bought an LM5066EVK and am using the PI-Commander GUI (also tried the SPM_GUI, same result) to play around with it. The board is detected fine as LM5066EVK; I can set the limits, turn on and off, the measured Vin and Vout is also accurate, temp values seem plausible, all fine.

But the IIN value (and consequently, also the PIN value) is nonsense: IIN is not even close to what I measure with a multimeter; when the switch is turned off, it still measures around 150mA where it should be more like 7mA. If I turn the switch on, the IIN it measures is also way too high, but with a different offset (then it's more like 170mA too high, so it's obviously not a fixed offset that I can just substract).

I can read out the telemetry data via I2C and calculate the values as per the formula with the various coefficients in the data sheet and get the same results the GUI displays, basically.

Now, I could start fiddling around with the parameters, doing my own fit and so on to get the correct m, b, R and so on, but I don't understand why that should even be necessary. Shouldn't all the parameters be set up properly for the EVK already? In the GUI, it has all the correct values for Rs, Rdson etc., so everything matches the EVK BOM. So my thinking is the GUI should be set up correctly and it should in turn also set up the device correctly to work "out of the box". I also have no direct influence on the formula parameters the GUI uses except changing component values to something they obviously aren't.

The EVK is un-altered, meaning I did not change any of the components and all DIP switches are set to their defaults. I am switching a 48V source (which is what the EVK was designed for).

What am I missing here? Is this just an artifact of the +/- 4.5% measurement accuracy when I have a small load?

Greetings,
Sean

  • 0
    TI__Mastermind 40880 points

    Hi Sean,

    Welcome to E2E !!!

    We will get back to you by tomorrow. 

  • 0 in reply to Avishek Pal
    TI__Mastermind 40880 points

    Have you updated the value of current sense resistor here as per the physical RSNS on the EVM? What's the value load current? 

    What's the value load current? Telemetry accuracy depends on the amount of load current also as highlighted in the below screenshots. 

     

  • 0 in reply to Avishek Pal
    Prodigy 40 points

    The value of Rsens is 3mOhms on the EVK, and that is what is also set up on the GUI. As I wrote in my original question, all values for all components match between GUI and EVK. That's what I would expect, since the GUI was made for this EVK.

    There should be no change neccessary here.

    As for the load current: For testing I have a quite small load, I switch only about 6W; I just switch power to a PoE injector with attached load, that's the only 48V load I have at hand currently, but it is close to what the intended application for the device is. I my case it's not hut-plugging, but switching a 48V load with current measurement and some fault management. Worst case is only around 60W.

    But even if the inaccuracy is then up to 6.5%: the measured value is off by > 100%, I measure more than double of what the actual current is. And when the switch is turned off, the chip should measure 0W, but it still measures 5W. Is that just how inaccurate it is when the load is 0?

  • +1 in reply to Sean Durkin
    TI__Mastermind 40880 points

    It is the inaccuracy in power measurement with lower current. In the EVM the programmable current limit is set to 16.7A (CL = GND or HIGH-Z) or 8.7A (CL = VDD). However, you are trying to measure a few mA of current. 

  • 0 in reply to Avishek Pal
    Prodigy 40 points

    OK, understood. I was expecting that it might be more inaccurate for lower currents, but I did not expect it to report 5W of power when the MOSFET is turned off, and values more than 100% off when I switch only a few hundred mA. That was a bit surprising.

    Thanks for the info!