• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 62 subscribers
  • Views 177 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

TPS61235P: Does Not Start Up Under Load

MStirlingC
Prodigy 30 points
Part Number: TPS61235P

Hi, everyone. I'm looking for some insight from others with more experience than I have with the TPS61235P. My design is a simple concept: take a 3400mAh Li-ion cell and boost its voltage up to 5.1V, so that it can drive a color 64x64 LED display.

Disclaimer up front: I have never done a boost regulator design, so I took the "TPS61235P Typical Application" from page 1 of the datasheet and went from there.

Here are the schematics of the circuit I used in my design:

Notes:

1) L2 is DNP

2) L1 is a Taiyo Yuden NRS8040T1R4NJGJV. Its inductance value is 1.4uH; the datasheet specifies the maximum inductance value as 1.3uH. Its DC current rating is 7A and has a saturation current of 11A and a DC maximum resistance of 9.1 mohms.

3) The Vin capacitor is a 22uF X5R MLCC and has a working voltage of 6.3VDC

4) The output capacitors are the same type as the input capacitor.

5) I currently have AGND and GND tied together at a single point through R9, a zero-ohm resistor. In the redesign, I will eliminate the resistor and replace it with a small trace.

Below, I show the layout I have on my 2-layer board:

What I am seeing:

Under a no-load condition, I get the expected 5.1V output. However, when I use the output to try to power my LED display (which only requires 350mA at startup), the TPS61235P's output drops from 5.1V to 0V and then goes back to 5.1V at a rate of about 1 cycle per second. It seems that the load is triggering an overload circuit, and then after the TPS61235P resets, it tries again (and fails).

Any help you can give me would be greatly appreciated. Thank you!

  • 0
    TI__Genius 9046 points

    Hi @

    I just come back from holiday. Will reply you by end of this week. Thanks1

    -Wenhao

  • 0 in reply to Wenhao Wu
    TI__Genius 9046 points

    Hi @,

    Please kindly see my inputs below. Anymore questions please kindly let me know. Thanks!

    1.Did you try to use 1uH inductor? TPS61235P is optimized for 1uH inductance and other inductance may cause control loop instability;

    2.Did you measure the input voltage when output fails? Maybe the input fails when you were pulling load. Just 350mA load at output is not likely to trigger OCL;

    3.Please make an intact GND layer on 2nd. The AGND to PGND connection overlapped with Cin to PGND loop. As switching current is inside Cin to PGND loop AGND and related analog circuitry may be disturbed by switching noise;

    4.After used intact GND layer, use some small vias near Cin and Cout GND pad;

    5.Please extend connection between inductor and SW. You could do this by cut part of GND polygon.

    6.After revised layout, you could share me with updated one so that I can give a double check.

    -Wenhao

  • 0 in reply to Wenhao Wu
    Prodigy 30 points

    Thank you, Wenhao. I have made the changes as you suggested. Please see revised layout below.


    I don't understand what you mean about adding the vias:

    Wenhao Wu said:
    4.After used intact GND layer, use some small vias near Cin and Cout GND pad;

    Can you explain a different way, please? What is the purpose of the vias?

    I have 3 types of 1uH inductors arriving tomorrow to test in my current layout. I hope these will help.

    Thank you again.

  • 0 in reply to MStirlingC
    TI__Genius 9046 points

    Hi @,

    You should add at least one layer as a whole GND layer. This is what i mean for below. Using vias can help capacitors conducting switching current and dissipate heat through internal GND layer.

    3.Please make an intact GND layer on 2nd. The AGND to PGND connection overlapped with Cin to PGND loop. As switching current is inside Cin to PGND loop AGND and related analog circuitry may be disturbed by switching noise;

    If you could look at the connection between PGND pin and C3, AGND pin to PGND pin, you will find they are overlapped. This technique is not ok and would introduce switching noise to analog circuitry as I have said above. A GND layer is a good technique and a strong recommendation to resolve this. Thanks!

    -Wenhao

  • 0 in reply to Wenhao Wu
    TI__Genius 9046 points

    Hi @,

    Any updates? If no further questions, I will temporarily close this thread and if you have more questions you are welcome to create another new thread on E2E. Thanks!

    -Wenhao