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.

Compiler/DS92LX1622: CAD Pin and I2C Address Decoder

Part Number: DS92LX1622
Other Parts Discussed in Thread: DS92LX1621, USB2ANY

Tool/software: TI C/C++ Compiler

Hi
I'm trying to use DS92LX1621/DS92LX1622 pair in camera mode. 
i am using below circuit to connect to DS92LX1622 chip via I2C to my HOST.
at first i tied CAD pin to GND with 0ohm resistor and i used "110 0000" ID to assess the chip on HOST side the problem was that DS92LX1622 do not acknowledge the I2C packet frames.(investigated via Oscilloscope)
then i changed the RID to 2.0K and used "110 0001" ID on my I2c_master controller , now it works fine and i can easily read/write DS92LX1622 internal registers.
why it is like this?

  • Hello Hsvz,

    Have you verified that the I2C control is using the 8 bit or 7 bit addressing scheme based on your system SW? For example with setting RID = GND, try I2C addresses of 0x60 or 0xC0 based on 8bit or 7 bit convention?

    Also can you send the scope shots?

    Best Regards,

    Casey 

  • hi

    thanks for the answer

    i am using 7 bit addressing scheme so for RID=GND i use  "110 0000"  to access DS92LX1622

    here are two scope shots:

    RID=2 Kohm  ID:"110 0001" Ack: OK (writing "E0" ro register "27h" of DS92LX1622)


     

    RID=GND  ID:"110 0000" Ack: NOT valid

     

     

     

  • Hello hsvz,

    This is a strange issue - is it happening on multiple boards or just one?

    Best Regards,

    Casey 

  • unfortunately, I do not have another board to test this on it. I do not think it is related to the hardware because I can read/write the internal registers successfully. maybe there should be more considerations when you use RID=GND (e.g t2c timings,....).

  • I checked a new 1622 Chip and assembled a new PCB and the same problem exists:

    I investigated that a weird thing is happening when RID : GND: the chip does not ACK packet sent to ID "11000000" but it ACK the packet sent to ID"11000010" (which it is supposed to work when RID=2Kohm) !!!

    here are two scop shots:

    RID =GND ,M/S=H ,PDB:High  ID:  "11000000"

    RID =GND ,M/S=H ,PDB:High  ID:  "11000010"

    btw here is my chip marking:

    89A4LEC8 LX1622

    and my schematic:

  • Hello hsvx,

    What state are the S1 switches in? Have you tried changing between the two modes for M/S?

    Best Regards,

    Casey 

  • Casey McCrea said:
    What state are the S1 switches in?

    PDB switch : open=> PDB : High

    M/S switch : open => M/S: High

    BISTEN switch : Close=> BISTEN  : low

    Casey McCrea said:
    Have you tried changing between the two modes for M/S?

    yes, nothing changed.

  • Hello hsvz,

    This issue is quite strange and the device in question part of a legacy portfolio so it may take some time to investigate on this. We will check to see if we can find an eval kit for this device to do some testing on our side. Please expect a follow up in ~ 1 week

    Best Regards,

    Casey 

  • Hello hsvz,

    I tried this on the DS92LX1622 EVK: https://www.ti.com/tool/LX16EVK01

    It is working as expected on that HW. With CAD = GND, the address 0x60 is working:

    Is there anything else on this I2C bus which may be influencing the result?

  • hi casey

    thanks a million for the answer and the test you done.

    no there is no other I2C device on the I2C bus and only one Host is connected (a STM32F4 discovery board as Master I2C)

    could you please answer these about the EVK and your test?

    • you have configured I2C bus speed to 100Kbps, right? What Host I2C controller you used for this test?
    • what are the values of R25,R24 on the EV? (note that in "www.ti.com/.../snlu013" document the value in BOM and shematic do not match!)
    • are C56,57 installed?
    • what is VDDIO at that board? 3.3v or 1.8v?
    • what is VDDI2C you are working? 3.3v or 1.8v?
    • R18 is on schematic (page 14) but not on BOM (page10),is is installed on the EVK?
    • what are the state of these jumpers? Switch2-4 , JP9,J8,JP8
    • could you please test whether you can connect to 1622 with ID "11000010" while CAD=GND

  • Hello hsvz,

    • you have configured I2C bus speed to 100Kbps, right? What Host I2C controller you used for this test?

    • what are the values of R25,R24 on the EV? (note that in "www.ti.com/.../snlu013" document the value in BOM and shematic do not match!)
      • It does look like there is an error in the user's guide and again, since this is a legacy device I'm not sure where the discrepancy is coming from by the I2C pull up resistors on my board are 1kohm
    • are C56,57 installed?
      • Yes
    • what is VDDIO at that board? 3.3v or 1.8v?
      • 1.8V
    • what is VDDI2C you are working? 3.3v or 1.8v?
      • 1.8V
    • R18 is on schematic (page 14) but not on BOM (page10),is is installed on the EVK?
      • Yes
    • what are the state of these jumpers? Switch2-4 , JP9,J8,JP8
      • PDB = H, BISTEN = L, M_S = H, JP8 = 2-3,JP9 = 1-2, J8 = connected to I2C host
    • could you please test whether you can connect to 1622 with ID "11000010" while CAD=GND
      • It does not work - only the address listed in the datasheet works when CAD = GND

  • thanks 

    I checked all the circuit again and I came across a missing track in my design, one mistake was there and VDDCML pin was not connected to VDD

    it was interesting that I2c was working for all RIDs but not for CAD=GND