• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 3 answers
  • Subscribers 100 subscribers
  • Views 1115 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.

TMS320C6746: c6746 DSP bulge issue after long-time running

Thomas Yang55737
Expert 5026 points
Part Number: TMS320C6746
Other Parts Discussed in Thread: TMS320C6742

Hi Expert

After long time running, several C6746 SOC exists a bulge phenomenon on prototype before production , after check customer'schematic, we found 2 suspicions, customer want to get confirmation from factory side whether they need change these 2 points, if it really needs, they have to do re-design board.

1, In customer design, the pins  M2,N4,P1,P2 (SATA_VDD),L15(VDDA12_PLL1),N15(VDDA12_PLL0),M12(USB_CVDD12) is connected to 1.3V, really test result on these power rails is 1.28V~1.3V. According to C6746 datasheet, we know that M2,N4,P1,P2 (SATA_VDD)  are reserved pins , and they should be either tied to a STATIC 1.2V supply or left unconnected. But according to slva490.pdf, TMS320C6742/C6746/C6748 Power Requirements, the SATA_VDD has a tolerance from -5%~+10% around1.2V.  The test result 1.28V~1.3V should be in the range of -5%~+10%(1.32V), So customer want to know whether 1.3V/1.28V on SATA_VDD is OK and whether could cause possible bulge? The similar questions for L15,N15,M12.

2 The unused pins P14 is connected to 1v8D, P15 is connected to 3v3D, according to datasheet, these 2 pins should be left un-connected, could this possibly cause DSP bulge?

Any suggestions if other possible reason? (C6746 PART Schematic is attached)

C6746.zip

  • 0
    TI__Guru**** 161600 points
    Hi Thomas,

    Regarding pins M2,N4,P1,P2, they should be connected as described in the datasheet: STATIC 1.2V supply. However, I don't think that connecting them to 1.3V will cause dsp bulge... L15, N15 & M12 should be 1.2v FILTERED supply, as far as I can see this is correct on the schematic you shared.

    Now for pins P14 & P15 it is explicitly said that they should NOT be connected. Based on what you shared, in my opinion this is the most likely cause of your problem.

    I am also looping the design team to elaborate if needed.

    Best Regards,
    Yordan
  • 0 in reply to Yordan Kovachev
    TI__Expert 5026 points
    Hi Yordan,

    Thanks for feedback,
    "L15, N15 & M12 should be 1.2v FILTERED supply", from schematic, it show that the 1V2Aa is connected to a 1.3V power supply through CBG2012U121T, so they should also be considered as 1.3V, according to your previous info, can these 3 pins be processed similar as RSV3 (M2,N4,P1,P2) which is static 1.2V supply or it could be put as a tolerance --5%~10% around 1.2V? In datasheet, RSV3(M2,N4,P1,P2) have a clear description on this "STATIC", but L5,N15&M12 don't have a "STATIC" requirement.
  • 0 in reply to Thomas Yang55737
    TI__Guru**** 161600 points
    Hi,

    There is no requirement for tolerance for L15, N15 & M12 pins. The recommendation is 1.2V, however 5.1 Absolute Maximum Ratings Over Operating Junction Temperature Range states that they can withstand up to 1.4V. Of course this is for short periods of time, long exposure to the absolute maximum ratings will eventually damage the pins' buffers and will decrease the SoC POH or cause the chip to work unreliably.

    Best Regards,
    Yordan
  • 0 in reply to Yordan Kovachev
    TI__Genius 15910 points

    Hi Thomas,

    It would be helpful if you could provide some additional details about the boards that are experiencing the bulge.  

    • What percentage of the prototype boards experience a bulge?
    • How long were the boards running before the bulge occurred? 
    • Are the boards with a bulge still working?

    Also, can you send pictures of the bulge?

    Regards,

    Melissa

  • 0 in reply to melissaw
    TI__Expert 5026 points

    Hi  Melissa,

    1, 5 boards issue/ total 50 boards

    2, several days long time running

    3  Bulge DSP can not work as normally (bus error), but it can be connected by JTAG

    Please attachment on Bulge picture

    one more question:

    customer asked the detailed circuit of these 2 pins (P14 P15) to explain why the connected voltage may make the DSP abnormal heat and bulge.Could you help it? Thanks!  Dongfang C6746 issue.zip

  • 0 in reply to Thomas Yang55737
    TI__Guru**** 161600 points
    Hi,

    I see this issue is discussed offline, so I am closing this thread for now.

    Best Regards,
    Yordan
  • 0 in reply to Thomas Yang55737
    TI__Guru* 83825 points
    Hi Thomas
    From the description so far provided /shared - i do not think we can assertively say that it is due to either your suspect 1 and 2 are the cause of the issue.
    We will not be able to provide any detailed circuit for any pins and IOs beyond what is in the public documentation , as if the customer is violating the datasheet recommendations, they need to fix those.

    Having said that, I do not think either of them are strong suspect to be the cause for the issue - so perhaps best would be to have them send a few units back for FA via the customer return process.

    Please continue to also check EVM schematics and other datasheet recommendations to see if they have some other violations or handling issues that is causing this.

    Need to make sure there is nothing in the circuit driving current into the device and no core/IO rails shorted etc (which maybe more related to layout /routing etc).

    Regards
    Mukul