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.

TPS54062: Vout drops when load is connected

Part Number: TPS54062

I am using TPS54062 to design a buck convertor with Vin (nom) = 24V, Vout=5V, Iout=20mA max.

After designing and simulating the entire circuit on Webench, I physically assembled the circuit with the exact same component values. 

The output voltage comes out to 5V as designed but it drops to 0.5V when load (220 ohm resistor) is connected.

I am attaching the Webench design report to show the component values that I have used. Please help.

WBDesign2_Steady State-2.pdf

  • Hello,

    Sorry to hear about the issue.

    I went ahead and evaluated your schematic in webench and noted that for the BOM I had (approached your comp values), my phase margin was approaching non-ideal (47deg). 45deg is bare minimum. Webench would not allow me to reduce the rcomp value to yours perhaps because it would create unstable design. 

    You can take a look. https://webench.ti.com/appinfo/webench/scripts/SDP.cgi?ID=8A0F6B61C22EE400

    What is the part number of the output capacitor you are using? I am assuming you are using a ceramic.

    I can help evaluate your compensation network further to get it to become a stable design. Stability seems to be the concern here.

    My first recommendation would be to change 

    Rcomp to 133k, Ccomp1 to 2.7nF, and Ccomp2 to 22pF for using ceramic 47uF output cap.

  • I have changed the component values to Rcomp to 133k, Ccomp1 to 2.7nF, and Ccomp2 to 22pF 47uF output cap, but it is still behaving the same way, i.e. 5V output drops to 0.5V when load resistor of 220 ohm is connected. 

    I even changed the entire circuit to the example circuit given in figure 20 in section 8.2.1 in datasheet of tps54062, but the issue persists. 

  • This is confusing to me, Ahsan.

    Please share with me an oscilloscope capture of regulation without load and if possible, right before output drops. 

    If I don't get any answers from that, might need to then evaluate layout.

    IF you can share all of it, that would be good.