LM25145: LM25145 output voltage is not changing 2.6V,

Hi,

Please fill out our design calculator so we can compare it to your schematic - https://www.ti.com/tool/LM25145DESIGN-CALC

Regards,

Rahil

Thank you very much for your assistance. I have updated my schematic according to the design calculator. The input voltage (Vin) range is 12-24V, and the output voltage (Vout) is set to 2.25V using feedback resistors of 22K and 12K. However, when I set Vin to 12V, the output voltage is not exactly 2.25V; instead, it is 2.33V. When I set Vin to 14V, 16V, 20V, or 24V, the output voltage consistently reads 2.6V. Even when I attempt to achieve different output voltages by changing the feedback resistors, the output voltage remains at 2.6V. I am sharing my schematic and design calculator with you for further review.

LM25145_quickstart_calculator.xlsm

Hi,

Can you check your power supply? It might be current limited

Regards,

Rahil

I have not yet connected the load, and I have checked the voltage using another power supply. However, the voltage remains at 2.6V instead of the expected 2.26V. I used 6x4.7uF input capacitors, but the design calculator suggested using 11x4.7uF. For the output capacitor, I used 3x47uF capacitors, while the design calculator recommended one 100uF capacitor. Could this discrepancy be causing the problem?

Another issue arises when I connect the load and measure the voltage at the load. The voltage drops significantly to 0.6V. This indicates that the compensation is not correct, even though I calculated the compensation network using a design calculator. The design calculator provided a phase margin of 62 degrees and a crossover frequency of 99kHz, with a switching frequency of 400kHz.

Hi,

Cout should be fine. Can you add 5 more 4.7uF input caps and check if you're still having this problem? 

Also please double check the values of your resistors on the board.

Regards,

Rahil

Thank you very much for your assistance. I have double-checked the values of the resistors on the board, and they are all correct. I have also removed the 4.7uF capacitors and added 5x10uF input capacitors, but the problem still persists.

I reviewed the LM25145 design on Webench Power Designer, which suggests a 2.8uH power inductor, while the design calculator recommends a 1.2uH power inductor. Which inductor size is correct?

I am sending you my PCB and schematic files. Could you please check if there is any issue with my PCB design? Thank you very much.

PCB_Project.zip

Hi,

Try changing RC1 to 3k and CC3 to 470p. If that doesn't work it's probably a layout issue. I can do an in depth review

Regards,

Rahil

I have made some adjustments to the compensation network of my circuit. Specifically, I have changed RC1 to 3.3K and Cc3 to 470pF. The other compensation components are as follows: RC2 is 100 ohms, CC1 is 3300pF, and CC2 is 100pF. After making these changes, the circuit appears to be functioning correctly.

However, I have observed an interesting behavior regarding the output voltage. When I connect the load, the output voltage is stable at 1.6V, which is the expected value. But when I disconnect the load, the output voltage rises to 2.6V.

Is this a common phenomenon with the LM25145, or is there a minimum load requirement that I might be overlooking? Alternatively, could there still be an issue with the compensation network that needs to be addressed?

Glad to see we're moving in the right direction. 

Can you please please test at light loads and see at what load it switched from 1.6V to 2.6V? Behavior at no load is unusual.  

Regards,

Rahil

 I have been testing a DCC power supply with light loads, specifically 0.1A and 0.5A. When a load is connected, the output voltage is 1.6V, which seems correct. Without any load, the output voltage is 2.6V. I then connected this 2.6V output to an H-bridge control circuit. However, when I measure the output voltage at points A and B in the schematic of the H-bridge, it drops to 1.6V even without any load connected.

1. Why does the voltage drop to 1.6V just by connecting the H-bridge, even when no load is attached?

2. When I connect a load through the H-bridge between points A and B, I observe no output voltage.

3. I have checked the datasheets of the MOSFETs used in the H-bridge, and their operating V_DS voltage is lower than 1.6V.

Is the issue with the DCC power supply or the H-bridge? Why does the voltage drop to 1.6V when the H-bridge is connected, and why is there no output voltage when a load is connected through the H-bridge?

Hi,

Not sure what could be causing this. Schematic looks fine. Is your power supply limited in any way?

I'll review your layout and get back to you

Regards,

Rahil

The power supply is not limited. According to the datasheet, the crossover frequency should be one-tenth to one-fifth of the switching frequency. Therefore, I selected the following components: RC1 = 5k, CC2 = 100pF, RC2 = 100 ohms, CC1 = 3300pF, and CC3 = 470pF. With these values, the crossover frequency is 51kHz and the phase margin is 65 degrees. Additionally, I have changed the output capacitor to 247uF. However, when I connect the load through the H-bridge circuit and turn on switches Q3 and Q6, the output voltage increases very slowly to the set value. It starts from 0.35V and gradually rises to 1.6V, but it takes a long time to reach that value. Should I increase the output capacitance further, adjust the compensation network, or increase the switching frequency?

I am planning to download the PCB file for the LM5146. Can I use the LM25145 in this PCB design, given that there is no design file available for the LM25145, only the LM5146-Q1 EVM? I already have the inductors, resistors, and capacitors from my previous design, and I intend to use them in the LM5146-Q1 EVM PCB to design the board. Another option is to use a PCB design exported from Webench. Can I use that PCB instead?