• State
  • Locked Locked
  • Replies 6 replies
  • Subscribers 64 subscribers
  • Views 3849 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.

LMZ21701: Failure V2

Dimce Iliev
Prodigy 40 points
Part Number: LMZ21701
Other Parts Discussed in Thread: CSD18540Q5B, DRV8303, LM46002,

Hi,

we are experiencing similar issues as described in this post: https://e2e.ti.com/support/power_management/simple_switcher/f/858/t/528460

System description:

48V system voltage (Input protection, surge, EMI filtering)

BLDC Motor control with DRV8303 + CSD18540Q5B (halfbridges) + AD8210 (current measurement, 20mOhm)

LM46002 48V system voltage to 12V

LMZ21701 12V to 5V (used for uC [third party 3V3, 1V8 converter], current measurement AD8210)

6 Layer PCB

Bottom: 

Mid 4:   (48V sys)

Mid 3:   (IO signals)

Mid 2:   (IO and supply)

Mid 1:   (Ground plane)

Top:   (Vos trace, ground plane)

Experienced problem:

LMZ21701 works ok as long the PWM on the motor is disabled (ripple ok, output load as specified)

After enabling the PWM on the motor (motor in stop condition, same PWM on all phases) --> Short circuit on LMZ21701 output after ~15min

Things tried already (no success):

- decrease feadback resistors by factor 10

- increase output capacitance (DC bias)

- add small capacitance (100nF) on Vos pin (Top side)

- cutting output 5V traces to trace down noise coupling

- external 12V supply

- and various combinations of the above things

Additional info:

We have a version 1.0 of the same PCB where the "Mid 4" layer do have  5V (on a plane) below the LMZ21701 (48V sys voltage is routed further away) and that version does run and do not have described problem.

How sensitive is the LMZ21701 to noise?

  • 0
    TI__Mastermind 18940 points

    Hi Dimce, 

    The control loop traces (feedback and the output sense) can be susceptible to noise. If you have a high voltage switching circuit near those nodes it could affect the control loop.

    The things you tried all seem to make sense for noise mitigation, but the large 48V plane is still underneath the IC. I wonder if the proximity of the large 48V plane in the PCB stackup is still resulting in too much capacitive coupling to the sensitive nodes. Is the motor running off of the 48V plane, causing high frequency noise directly underneath the regulator? Is there any reason for the layout change near the regulator (e.g. routing the 48V plane directly under the IC)?  

    I think it would make sense to submit the failing parts to TI for analysis.

    Regards, 

    Denislav

  • 0 in reply to Denislav Petkov
    Prodigy 40 points

    Hi Denislav,

    yes the motor is running from the 48V. Your explanation with the capacitive coupling sounds plausible. We are just surprised about the noise sensitivity, I think one would more expect unstability of the control loop or introduced noise on the 5V but not that the part breaks.

    The layout change underneath the regulator was performed because of worries about the current transfer of the plane and layout simplification.

    I can provide you with some faulty parts and working ones of the same batch. Just let me know where to send them.

    Regards,

    Dimce

  • 0 in reply to Dimce Iliev
    Prodigy 40 points
    Hi,

    do you have any updates on this?

    Thanks and regards,
    Dimce
  • 0 in reply to Dimce Iliev
    TI__Genius 15810 points

    Hi Dimce,

    I apologize for the drop in communication. I will send you an address soon. 

    Regards,
    Akshay

  • 0 in reply to Dimce Iliev
    TI__Genius 16025 points

    Hi Dimce,

    Sorry for the delay,

    Please send the FA request to:

    Texas Instruments

    Atten: Tara Jamali

    2900 Semiconductor Drive

    Santa Clara, CA 95051

    Please fill out this form and attach it enclosed in your shipment. 5658.Customer Information Form (Rev L).doc

    Thanks,

    Anston

  • 0 in reply to Anston Lobo
    Prodigy 40 points

    Hi guys,

    thanks for the additional information and follow up.

    I'll send the samples in the next few days to you.

    Best regards,

     Dimce