• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 46 subscribers
  • Views 517 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.

AMC1303M2510: Fail-safe output issue

Jason Fan
Intellectual 1705 points
Part Number: AMC1303M2510

Hi  Experts:

My customer is using the AMC1303M2510 for the in-phase current measurement in the servo drives. However, the device usually entered the fail-safe mode like the captured picture below:

you can the data out is hang on "1" before clock for 256 cycles (10MHz) , however, in the fail-safe mode, the CLK will hang on "1" for 70uS , is that correct ?

The device is in the fail-safe mode when the system is normally working , this is not we want to see because It will be false to drive the motor.

I checked the condition to trigger this mode :

a) The missing on the AVDD   b) The VCM > Vcmv (3V,  5V-2V)

My questions are:

1)  We checked the waveform of the AVDD , Vin+,Vin- with respect to the GND1, there is a little switching noise but is not sufficient to trigger the safe-fail mode ,

is there any factor to trigger this mode except for the two ones I listed?

2) We tried the value of the input capacitor on the AVDD , Vin+,Vin- as the datasheet recommended , but it has no effect. How to calculate the input capacitor to eliminate the common/differential mode noise effectively ?  

yellow: Data output    green: CLK out   Blue: phase current (DC current)

  • 0
    TI__Guru**** 168485 points

    Hi Jason,

    The pictures you sent make some sense, the first one is the 256 cycles of the 10MHz internal clock of the AMC1303M2510 while the device is going into fail-safe mode.  Assuming this is the VCM fault, the device comes back into normal operation after the VCM level returns below VCMov (after ~70uS).  If it is because of AVDD failure, then it is taking 70uS for the voltage rail to recover.  Either way, it's a condition that lasts for at least ~96uS so I don't think it's a simple noise issue.  Can you provide more detail on how VINp and VINn are referenced back to AGND?  You can send me a note with better schematic snippets if you have them, the posted images are kind of hard to read. 

  • 0 in reply to Tom Hendrick
    TI__Intellectual 1705 points

    Hi  Tom:

    Thanks for your comments.

    I will give you the capture waveform to indicate the Vinp and Vinn and the clear schematic.

    One question , duration of the safe-fail mode depends on the time when the condition recover  rather than a fixed time, on the other hands, when the condition removed , the device will restore from the safe fail mode to the normal operation ?

  • 0 in reply to Jason Fan
    TI__Guru**** 168485 points

    Hi Jason,

    Yes - once the condition that caused the fail-safe mode to be activated is removed, the AMC1303 will go back to normal operation.

  • 0 in reply to Tom Hendrick
    TI__Intellectual 1705 points

    Hey Tom:

    Please check the below waveform recommend , in sight of the waveform , the device quite from the safe fail mode even the noise is exiting , that means this issue is not triggered by the noise on the Vin.

    1)  Enter the safe-fail mode :    Blue: Vinp to AGND   Yellow: Vinn to AGND    Purple: Dout 

    2)  Quit the Safe fail mode :

    3) Schematic: 

  • 0 in reply to Jason Fan
    TI__Guru**** 168485 points

    Hi Jason,

    What is the PV5U rail doing at the point when the fail-safe is happening?  You mentioned that you checked it initially, but do you have scope shots?

  • 0 in reply to Tom Hendrick
    TI__Intellectual 1705 points

    Hi Tom:

    I didn't capture the waveform of 5V rail, but i checked it before and make sure it's stable.

    Is there any factor trigger the fail -safe mode ? 

  • 0 in reply to Jason Fan
    TI__Guru**** 168485 points

    Hi Jason,

    Avdd and common mode input are pretty much the limits as to what can cause fail-safe mode to trigger.  This is a feature you won't find in some other parts.