• State Resolved
  • Locked Locked
  • Replies 11 replies
  • Subscribers 35 subscribers
  • Views 668 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.

DS90UB960-Q1: Issue of ECC2_ERR from CSI_RX_STS(0x7A) register on DS90UB960

Jim Lee Jr.
Prodigy 110 points
Part Number: DS90UB960-Q1
Other Parts Discussed in Thread: ALP

Dear team,

We are using DS90UB960-Q1 forwarding stream sent by sensor with DS90UB953 serialiser. We only connect 1 sensor to DS90UB960 RX port 0.

I found that a multi-bit ECC error has been detected according to the value in register 0x4E(RX_PORT_STS2)=4d and 0x7A(CSI_RX_STS)=02 of DS90UB960.

This doesn't always happen, so we didn't store all the values in registers. But we can confirm that, when ECC_2_ERR is prompted in the deserialiser, the CSI data received by the serialiser has no errors, and the registers indicating CSI2 errors in the serialiser have no errors either.

My questions is: When ECC2_ERR detected by deserialiser,

1. What is the multi-bit ECC error meaning and why it only happend in deserialiser?

2. Is there any configuration set mistake in serialiser and deserialiser? Which part do we need to check?

3. Is this error indicating any mistake according to hardware working and design?

Any information may be helpful, thank you very much!

  • 0
    TI__Mastermind 24751 points

    Hi Jim,

    Thanks for reaching out with this question. Could you please help to confirm the below points?

    1. Could you please read back register 0x4D (checking link diagnostics)
    2. Have we run the margin analysis program (MAP tool) from ALP to check how much link margin we have?
    3. Could you also read back registers 0x55 and 0x56 to check for FPD parity errors in the link?
    4. Could you please help to provide your initialization script for your deserializer?

    Essentially if we see there are no CSI-2 errors at the serializer, but there are errors at the deserializer side, it is likely that there is some instability in the link. This could be due to hardware configuration, and we can look at the schematic / layout of the ser/des pair to see if there is anything that could cause this link stability issue. 

    Best,

    Thomas

  • 0 in reply to Thomas Bejin
    Prodigy 110 points

    Hi Thomas,

    Thanks for the quick reply.

    1. Please refer to the register values below:

    Error

           0 1 2 3 4 5 6 7 8 9 a b c d e f
    00: 64 00 1e 40 d0 01 00 fe 1c 10 7a 7a 01 00 02 ff
    10: 00 00 00 00 00 00 00 00 80 61 a8 e3 dd 00 04 06
    20: e0 03 00 01 81 00 00 00 00 00 00 00 00 00 00 00
    30: 00 00 01 03 00 01 00 11 00 00 00 00 00 00 00 00
    40: 00 a9 71 01 00 00 20 00 00 00 00 12 01 13 4d 62
    50: 89 00 00 03 00 00 00 00 5a 00 00 30 80 a8 50 00
    60: 00 00 00 00 00 a8 c0 00 00 00 00 00 00 7c 8a 88
    70: 2b 2c 00 02 00 10 00 c5 00 01 02 01 00 f1 00 00
    80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
    90: 23 45 00 00 ff ff 00 00 00 00 00 00 00 00 00 00
    a0: 00 00 00 00 00 1d 00 00 00 00 00 00 00 00 00 00
    b0: 1c 13 1f 08 25 00 18 00 9c 33 83 74 80 00 00 00
    c0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
    d0: 00 43 94 1f 60 f2 00 02 00 01 00 00 00 00 00 00
    e0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
    f0: 5f 55 42 39 36 30 00 00 00 00 00 00 00 00 00 00

    Error cleared when data rate changed

           0 1 2 3 4 5 6 7 8 9 a b c d e f
    00: 64 00 1e 40 d0 01 00 fe 1c 10 7a 7a 01 00 02 ff
    10: 00 00 00 00 00 00 00 00 80 61 a8 e3 dd 00 04 04
    20: e0 03 00 01 00 00 00 00 00 00 00 00 00 00 00 00
    30: 00 00 01 41 01 01 00 00 00 00 00 00 00 00 00 00
    40: 00 a9 71 01 00 00 20 00 00 00 00 12 01 03 04 62
    50: 9f 00 00 03 00 00 00 00 5a 00 00 30 80 a8 50 00
    60: 00 00 00 00 00 a8 c0 00 00 00 00 00 00 7c 8a 88
    70: 2b 2c 00 02 00 10 00 c5 00 01 00 00 00 f1 00 00
    80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
    90: 60 eb 00 00 ff ff 00 00 00 00 00 00 00 00 00 00
    a0: 00 00 00 00 00 1d 00 00 00 00 00 00 00 00 00 00
    b0: 1c 13 1f 08 25 00 18 00 9c 33 83 74 80 00 00 00
    c0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
    d0: 00 43 94 27 60 f2 00 02 00 01 00 00 00 00 00 00
    e0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
    f0: 5f 55 42 39 36 30 00 00 00 00 00 00 00 00 00 00

    2. The ALP software is not available to us because we don't have the EVM board (EVM needed, right?). 

    3. We are going to pay more attention to the stability of the connection.

    Thank you again.

    Best,

    Jim

  • 0 in reply to Jim Lee Jr.
    TI__Guru** 107560 points

    Hello Jim,

    Thomas is OoO and will be back on Tuesday, after the public Holiday.

  • 0 in reply to Hamzeh Jaradat
    Prodigy 110 points

    Hi Hamzeh,

    Thank you for your information!

  • 0 in reply to Jim Lee Jr.
    TI__Guru** 107560 points

    you are welcome!

  • 0 in reply to Hamzeh Jaradat
    TI__Mastermind 24751 points

    Hi Jim,

    Register 0x4D is reading 0x13 in the "error" register dump which indicates that there has been a drop of lock between the SER/DES. When lock is dropped, CSI-2 data may be lost.

    We should be able to run the margin analysis tool as long as we have I2C connection between the computer & DS90UB960 which we can use to run the tool through ALP. Would you be able to run this test?

    By the way, do you know if there are specific conditions that cause this system to be seeing an error? You mentioned that the error cleared when the data rate changed, is this the FPD data rate or CSI data rate? Could you please provide some more background on this? 

    Best,

    Thomas

  • 0 in reply to Thomas Bejin
    Prodigy 110 points

    Hi Thomas,

    Thanks for your guidance,

    1.

    Thomas Bejin said:

    We should be able to run the margin analysis tool as long as we have I2C connection between the computer & DS90UB960 which we can use to run the tool through ALP. Would you be able to run this test?

    We have just got an IIC2USB board(CHI341) and are making a connection between the UB960 and a Windows PC. We can read and write I2C through this board. But the ALP software can not observe the UB960. Maybe FT260S (IIC2USB-HID) working?

    2. 

    Thomas Bejin said:
    By the way, do you know if there are specific conditions that cause this system to be seeing an error?

    We haven't found the specific condition, but it's more likely to happen when the system has just started.

    3.

    Thomas Bejin said:
    the FPD data rate or CSI data rate?

    It is the CSI data rate configured through the UB960 0x1F register.

    Best,

    Jim

  • 0 in reply to Jim Lee Jr.
    TI__Guru** 107560 points

    Hello Jim,

    Someone from the team will follow up shortly, while Thomas is OoO.

  • 0 in reply to Hamzeh Jaradat
    Prodigy 110 points

    Received and thank you, Hamzeh.

  • 0 in reply to Jim Lee Jr.
    TI__Genius 16230 points

    HI Jim,

    Let me review this information and provide feedback this week.

    Glenn 

  • +1 in reply to Glenn Uranga
    TI__Genius 16230 points

    HI Jim,

    In the ALP software there is source code folder where you can get the Margin analysis script. Can you run this on your system?

    C:\Program Files (x86)\Texas Instruments\Analog LaunchPAD v1.57.0010\PreDefScripts\DS90UB960

    Glenn