• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 63 subscribers
  • Views 75 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.

TPS2595EVM: Noise from the eval board

Jaehyeok Kim
Prodigy 10 points
Part Number: TPS2595EVM

Tool/software:

Good evening TI,

I'm stuck with a noise problem with the TPS2595EVM (TPS259541, U3) and would appreciate any help!

Situation:

My team wants to build a cable that prevents short circuits at the output port (LED HEADLIGHT). We want this cable to be compatible with two different battery packs.

  • Battery Pack A: Pin 3 of the USB-A connector is V+, and pin 4 is GND. Pin 1 is connected to the shield, and the shield is connected to ground.

  • Battery Pack B: Pin 4 is V+, and pin 3 is V-. However, V- is connected directly to the battery and is not a ground. Pin 1 is also connected to the shield and grounded.

To solve the polarity issue, we connected an H-bridge on both sides of the eFuse, and we confirmed that it works! However, we are seeing significant noise and are wondering if this is inherent to the eFuse itself or if it's just coming from the evaluation board. Is it possible that this noise will disappear once we move to a custom PCB?

Another issue is that when Battery Pack B is connected, the eFuse's ground will be connected to a floating ground, not a true ground. I'm not sure how problematic this will be. If you have any alternative solutions, I would be very grateful.

(Initially, we tried connecting the eFuse ground to the shield, but this caused a short circuit when using Battery Pack B, so we couldn't do that.)

Initially, I thought the noise came from the transistor only, but even if I connected the cable to the TPS2595EVM directly, I saw very similar noise.

For your reference, the attached picture was taken with Battery Pack B, and the LED headlight is compatible with this battery pack.

Normal operating condition

- Normal operating condition. No noise.

Cut the cable in half and directly connect it to the input and output of the efuse eval board.

No LED headlight is connected, but the same noise was captured.

LED on - 259541 noise

eval board only - USB-A noise

noise at the usb-a port.

Below is the noise measured with the H-bridge on each side of the efuse.

Efuse noise with 2 h-bridges

measured at U3.

H-bridge noise

Measured at the H-bridge

eval board only - USB-A noise - 2 

Measured at the USB-A port.

  • 0
    Prodigy 10 points
    Jaehyeok Kim said:
    No LED headlight is connected, but the same noise was captured.

    I turned on battery pack without LED headlight

  • 0
    TI__Mastermind 38445 points

    Hi Jaehyeok,

    Jaehyeok Kim said:
    are wondering if this is inherent to the eFuse itself or if it's just coming from the evaluation board. Is it possible that this noise will disappear once we move to a custom PCB?

    No, the eFuse or EVM both should not cause this noise. This issue looks more like how you are connecting or using the devices.

    Jaehyeok Kim said:
    Another issue is that when Battery Pack B is connected, the eFuse's ground will be connected to a floating ground, not a true ground. I'm not sure how problematic this will be. If you have any alternative solutions, I would be very grateful.

    I feel like this can be a reason for this noise. Also when the eFuse ground is floating, how are you probing to see this noise. Is it between the device ground (floating) and out or system ground and device out.

    eFuse needs stable ground and we cannot float it for proper functioning of device.

    Since you are designing this system, you understand it in much better depth. Can you share a block diagram with how you are making the connections. I will try to see if we can do something to prevent this scenario of floating ground but again I think you understand this better than me and you must have already through of ways to prevent this issue.

    Best regards,
    Arush 

  • 0 in reply to Arush Gupta
    Prodigy 10 points

    Thanks for the response, Arush Gupta.

    Arush Gupta said:
    Can you share a block diagram with how you are making the connections.

    I confirmed that this block of diagram works as intended. (It limits current to the current that I set when I short the output)
    But, I am skeptical about the return path of efuse, and noise spectrums were detected as the pictures that I attached above.

    Arush Gupta said:
    This issue looks more like how you are connecting or using the devices.


    I double checked the connection, and I made the USB cable as short as possible, and I could see improvement in the noise spectrum on both type A and B. 
    However, Type B still generate some noise spectrum.


    The main purpose of this circuit is to prevent the short at the output end. The battery pack has traditional efuse but I want to add efuse to prevent the frequency of changing traditional fuse by hand.
    I am trying to achieve this without modifying the battery pack board. Therefore, I am updating the cable side.

    Arush Gupta said:
    eFuse needs stable ground and we cannot float it for proper functioning of device.

    is there better way to prevent the short other than the efuse? If there is no way, then I guess I need to update the battery pack board.


    Sincerely,


    Jaehyeok