• State TI Thinks Resolved
  • Locked Locked
  • Replies 1 reply
  • Answers 1 answer
  • Subscribers 33 subscribers
  • Views 342 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.

TCA9539: TCA9539RTWR

RITESH RAJ
Prodigy 20 points
Part Number: TCA9539


MARS Controller Front Panel.zip

Hi,

     I need a design verification for the I/O expander part to be used for controlling LED segments of alphanumerical display. The schematic is herewith attached. The I2C bus is pulled up with 4.7K resistor on another interface board. Please verify the Schematic and Layout for this part.

regards

Ritesh Raj

  • 0
    TI__Guru* 76991 points

    Hi Ritesh,

    The schematic looks okay. 

    • Decoupling caps present
    • Schematic pinout matches datasheet
    • SDA/SCL nets match schematic and signal nets
    • I2C device addresses appear unique
    • No floating inputs

    Layout looks like there's room for improvement. 

    • Ideally want to place ESD diodes closer to the connector, it looks like you can still move everything closer than what you have currently
    • SDA/SCL net seems to run a little close in parallel and then overlap each other in some areas (top and bottom traces overlap)
      • This can induce cross coupling/crosstalk
      • A lot of strange angles going into a via and test pads, generally bad practice due to potential reflections
    • the GND from the IO expander pins have to go through narrow paths to reach the common GND connector
      • More inductance/impedance, not to mention the large current you will be driving when you drive the GPOs low
        • This increases GND shifts and the via you have currently introduces more inductance to the circuit 
      • If this were my design, I would have segmented the top plane to have some 5V nets areas and made the rest of the top plane as GND to via up to in large quantity
    • D1 NETD1_1 has a super long trace that is skinny and very far from the connectors
      • If you expect large current to pass through this, you may want to increase the width, same goes for any other LEDs like this
    • C17 pad 2 has a trace that changes widths
    • May want to consider using a smaller decoupling cap (0402) and try to scoot the cap closer to the IC 
      • This one isn't a huge issue though
    • The IRSB LEDs you have all have GND pins that connect to the actual GND plane on the bottom through a Via (inductor)
      • If you are going to do this without a top plane GND, I would recommend a second via near the pad to GND this way you can at least cut the inductance in half. 

    -Bobby