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.

DS90UB935-Q1: Request a circuit review

Part Number: DS90UB935-Q1
Other Parts Discussed in Thread: DS90UB953-Q1

Hi TI expert,

A schematic of the DS90UB935-Q1 is attached.

DS90UB953_Circuit_v0.1.pdf

We refer to the DS90UB953-q1 EVM data. The PoC filter part is also designed the same as the EVM.

It was designed as 953, but changed it to 935 to lower the unit price. Are there any caveats when designing with the 935?

Please review the overall circuit.

Thanks,

Downey Kim.

  • Hi Downey,

    - For the PoC network, I suggest following the datasheet figure 8-2: 

    - The SCL and SDA should have 2.2k pull up to VDD1.8

    - For mode, if you're using synchronize mode, leave Rhigh DNP and Rlow 10kohm.

    - The VDD CAP power shouldn't be supplied

    There's no additional hardware or software change needed to use a 935 instead of 953.

    Best,

    Jiashow

  • Hi Jiashow,

    Thanks for you support!

    PCLK(Max) is  100MHz on the TI web. 

    Is PCLK 125Mhz not applicable? is it spec over?

    Thanks!

    Downey Kim.

  • Hi Downey,

    We usually talk about the 935/953 in terms of CSI bandwidth, not PCLK. What's the CSI rate of your input?

    Best,

    Jiashow

  • Hi Jiashow,

    The POC filter you suggested will be applied to the next PCB.

    Before taking your advice, we made a PCB with an existing circuit. (Refer to EVM circuit as shown below.)

    In the current board, fails at bandwith 800Mbps. If POC is not used, it is pass. Is it related to POC filter?

    Case

    POC

    CSI Bandwidth

    Pass / Fail

    #1

    X

    400Mbps

    Pass

    #2

    X

    800Mbps

    Pass

    #3

    O

    400Mbps

    Pass

    #4

    O

    800Mbps

    Fail

    your advice : For the PoC network, I suggest following the datasheet figure 8-2:

    I want to test by modifying the circuit, but there are currently no parts. 

    Please comment if modifying the POC filter can improve it.

    Thanks!

    Downey Kim.

  • Hi Downey,

    The PoC and CSI bandwidth shouldn't affect each other. Could there by crosstalks on the board? Could you take a look at the noise presented on the PoC network?

    Best,

    Jiashow

  • Hi Jiashow,

    We tested by changing the Ferrite Bead to BLM18HE152SN1 recommended in the datasheet. used POC and passed the 800Mbps test.

    Please tell me why the POC filter configured in EVM B'D and in the datasheet are different.

    EVM B'D used Ferrite Bead in series with 330ohm and 47ohm, and Ferrite Bead is a TKD product.

    Datasheet uses 3 Ferrite Beads 1.5kohm in series, Ferrite Bead is Murata product.

    Please explain...

    Thanks,

    Downey Kim.

  • Hi Downey,

    I would suggest using the PoC filter suggested in the datasheet. While both will work, one on the EVM was the older design and supports less current. 

    Best,

    Jiashow