• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 21 subscribers
  • Views 703 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.

Intermittent Configuration Failure

Tony Flynn18
Prodigy 50 points
Part Number: LMX2492


I am working with this LMX2492QRTWTQ1 part configured as a ramp generator on a custom board, 95% of the time the part comes up, gets configured and behaves as expected.  Sometimes the part comes up and fails to operate as expected leading us to reconfigure the part and go again.


I have taken a register dump of the part in the good vs failed operations and in both cases the registers that we program as part of our configuration of the part are programed correctly.  In the cases where the part fails to operate there is a bank of reserved registers (which we do not write to)) that all contain 0x00 values but in the good case they are non zero values.  The datasheet provides no info on these registers, see below for the offending reg bank.  There seems to be 2 failure modes, in some of the fail cases only register 0x31 is different, in the failure case its value is 0x11 and in the good case it is 0xb1.  The other failure mode observed shows the registers posted below.


Is there any errata for this part?  Can you provide any information on what these registers are used for or any information on what might be causing these differences?

good case:

< register 0x2e val:0x30
< register 0x2f val:0x51
< register 0x30 val:0x15
< register 0x31 val:0xb1
< register 0x32 val:0x0e
< register 0x33 val:0x0a
< register 0x34 val:0xce
< register 0x35 val:0x0e
< register 0x36 val:0xad
< register 0x37 val:0x86
< register 0x38 val:0x03

Bad case:


> register 0x2e val:0x00
> register 0x2f val:0x00
> register 0x30 val:0x00
> register 0x31 val:0x00
> register 0x32 val:0x00
> register 0x33 val:0x00
> register 0x34 val:0x00
> register 0x35 val:0x00
> register 0x36 val:0x00
> register 0x37 val:0x00
> register 0x38 val:0x00

  • 0
    Prodigy 50 points
    Adding a follow up to this issue:

    When I apply a software reset to the device by writing 0x04 to reg 0x02 and read back registers 0x2e to 0x38 I see that they take the same values as the good case reported above.

    If I loop on software resetting the device followed by reading back this register set, I sometimes see that they are read back as all zeros which is what I see in the failure case.
  • 0 in reply to Tony Flynn18
    TI__Mastermind 28021 points
    Tony,

    This part has been in production for some time and there is no known issue with the ramping, especially of this sort.

    With the readback, I see that you are reading registers R46 to R56, which are not disclosed in the datasheet. These bits should not change for a given device. However, they should not be all zeros. One bit, PFD_EN, which is R49[6], so if this is low, then I would think that the whole charge pump is powered down.
    So I would not expect this.

    Regards,
    Dean
  • 0 in reply to Dean Banerjee
    Prodigy 50 points
    Hi Dean

    Thanks for the info, I'm not sure where else to look with this issue. When I loop on resetting the device followed by a register dump I intermittently see that this bank of registers is all zeros, looking at the other registers they are either in their POR state or the state that we have configured them to.

    It looks like something is causing this part to enter an unknown state directly after a software reset, it seems to only happen 1 or 2 times per hundred tries.