• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 63 subscribers
  • Views 340 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.

TPS65950 1.8V DC/DC converter problem

Shashank Unhelkar
Prodigy 40 points
Other Parts Discussed in Thread: TPS65950

I am using two TPS65950 companion devices to power OMAP 3530 processors in my design.

The design and layout of both TPS65950 companion devices (functional and non-functional) are almost identical. The functional TPS65950 companion device does not have its ULPI and Keyboard interface connected. 

 

When I touch the probe to VREF pin, its output starts decaying from around 500 mV to 100 mV over 10 seconds. When I remove the scope probe from the pin, it VREF again goes up sets around 400 mV. 

32-KHz tuning fork crystal for that companion device didn’t work. I tried removing 22pf ceramic capacitors as well.Therefore I connected 32.768 KHz, 1V Amplitude clock from the function generator to 32KCLKIN pin of the companion device. It still has not changed situation.  

I also checked VINTANA1 LDO output of the companion device. It is zero as against 1.5V, expected value. My understanding is VINTANA1 should come up irrespective of 32KHz clock is available or not.

I would appreciate if anybody could please provide more insight. 

Thanks, 

Shashank

 

  • 0
    Prodigy 40 points

    I forgot to mention that when the board is powered ON,

    • 1.8V output rail of the non-functional companion device sits at around 700 mV,
    • there is no output VDD1 and VDD2 core rails,
    • no 32 KHz clock on 32KCLKOUT pin, and
    •  all internal LDO outputs (VINTANA1, VINTANA2, VRTC, VINTDIG,......) are less than 300 mV.

    Companion deivce is configured to operate in Master mode. 

    Please let me know if anybody needs additional information.

    Thanks,

     

    Shashank

     

  • 0 in reply to Shashank Unhelkar
    TI__Genius 15100 points

    I dont know the details of the architecture you are using. I am assuming two TPS65950 is used to power two OMAP processors.

    You mention functional and non-functional. So, I guess one set (TPS+OMAP) is working fine and the other has problems. Is this correct? How are the two sets powered up?

    If both sets are identical then I really dont know where to look for to debug. Are both devices configured as master?

    Do you work with any TI field/sales person via whom you can share the schematics? I wont be able to tell a lot till I see your schematics.

    I guess you have checked some basic things - boot mode pins, VREF, 32K (check if the installed caps are correct value).

     

  • 0 in reply to Gandhar Dighe
    Prodigy 40 points

    Yes, there are two sets (TPS + OMAP) in my design. While one companion device has a problem the other works fine.

    Both companion devices are powered through a common 3.6V VBAT plane. Both companion devices are configured in Master Mode.

    VREF: Non-functional companion device outputs 360 mV VREF output instead of 1.5V, correct VREF value. As I understand, VREF pin is not outputting correct value, since it does not get supply from VINTANA1 internal LDO. I confirmed this by measuring VINTANA1.OUT. It was zero. Not only that, but all LDO outputs are less than 300 mV. Per TI documentation, VINTANA1 LDO should generate output from VBAT upon Power-Up. Boot configuration and 32K clock does not control VINTANA1 output.

    32K Clock: the companion device failed to work with FX-115 crystal (used on Beagleboard). Therefore, I connected 32.768 KHz, 1V amplitude square wave to 32KCLKIN pin of the companion device. I still couldn't see any output on 32KCLKOUT pin.

    I am working with Ottawa FAE. I can make the schematic available to you for your review.

    Thanks,

    Shashank