• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 81 subscribers
  • Views 1224 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.

MSP430FR5969: ESD stops code execution

Martin Schmidt93
Intellectual 320 points
Part Number: MSP430FR5969


Dear All,

I'm ESD-testing a product that uses MSP430FR5969 as the sole micro controller.
The product is battery-powered and wakes up every 10 minutes, 24/7.
The product fails the ESD test. After the failed test, the MSP is found with it's correct supply voltage and its oscillator running, however it's not doing anything and has an increased current draw.
A test pulse (1/s) that is present on P1.3 during normal operation is no longer present.
The MSP's watchdog is being used but it seems unable to restore normal operation. When I pull nRESET, everything goes back to normal.

What can you recommend to improve ESD performance of the MSP ?
Are there any known weak spots of the device or typical engineering errors ?

It is acceptable that the product passes out for a few minutes, as long as it recovers by itself. Maybe this facilitates a solution.
If an external watchdog is the way to go, which device can be recommended ?

Appreciate any suggestions.

Regards
Martin

  • 0
    Guru 319210 points
    What specific ESD test did you do? What pins were targeted?

    The ESD ratings of this device are shown in the datasheet. Anyway, they are designed only to handle normal manufacturing processes.

    If the MSP (or for that matter, any other complex chip) is connected to the outside world, it needs additional ESD protection.
  • 0 in reply to Clemens Ladisch
    Intellectual 320 points
    Dear Clemens,
    Pins were not directly targeted. The testing was +/-4KV contact and +/-8KV non-contact onto the aluminium housing of the product.
    PCB ground was once connected and once not connected to the housing, no difference.
    Does the MSP430FR5969 have a particularly sensitive area where I could look first ?
    Martin
  • 0 in reply to Martin Schmidt93
    TI__Guru**** 219977 points
    Hi Martin,

    I would not say that the MSP430FR5696 has a ESD-specific sensitive area to be wary of, based on your description of the issue it appears that the device is being forced into a reset state from which it does not recover until a physical reset is applied. This could be from a SVSH low condition issue, please try disabling the SVSH (SVSHE = 0) to verify its involvement in the problem. You could also be in a trap ISR due to an IFG that the MSP430 is not equipped to handle. What LPM do you use and what is the current consumption difference between expected and failed operation? Would you be able to monitor the VCC and RST lines during an ESD event which replicates the issue? Please provide a schematic if possible and refer to the MSP430 System-Level ESD Considerations App Report (SLAA530): www.ti.com/.../slaa530.pdf

    Regards,
    Ryan
  • 0 in reply to Ryan Brown1
    TI__Genius 14165 points

     Hi Martin,

    can you please tell me which Device Revision you use for the test?

    Also how high was the increased current consumption?

    I can also offer to have a look to your layout in case you agree I will contact you via mail. Focus should be on the supply it self, how many layers, GND slices,...  A lot of things which are covered in the Apps Note mentioned by Ryan. Also I would focus on JTAG signals and here maybe the TEST pin. Do you have routed it to any test points on the PCB or the JTAG connector.

    One good experiment would be to CUT the TEST pin trace close at the device and GND it at the device. I know this has impact to JTAG but would give me some idea what's going on.

    Best regards,
    Dietmar

  • 0
    Intellectual 320 points
    Thank you to everyone who looked into it.
    We got it fixed in a common effort of mechanical, hardware and firmware-design. It was a bundle of measures that did it.
    The key issues were unused I/Os that should have been configured as a low output but were inputs.
    Various changes to the GND connections between the metallic housing and the PCB.
    Measures that did not bring any benefits were grounding of the TEST pin, SVSHE=0, GND reinforcement on the 2-layer PCB, additional blocking of one signal line that connects an exposed sensor to the MSP, additional blocking of VCC.
    Before the fix, once the fault condition had occurred, current was approx 500uA instead of < 2uA normal sleep current.
    Regards,
    Martin

**Attention** This is a public forum