LM5156-Q1: External circuit design

THE AUX-BUS voltage is 16V

 The green line is the output of 12V. It has 0.3V ripple voltage. Blue line is the AUX-BUS input. Purple line is the waveform of COMP pin. We want smaller ripple voltage of 12V output.

Hello Zhexi,

Thanks for using the e2e forum and sharing the schematic.
If the output voltage ripple is too large, and you see oscillation on the COMP pin voltage during steady-state operation, adjusting the compensation values is a valid approach.
I do not know the parameters for VIN, output power and inductance, so I cannot re-calculate the compensation for you.
However, I recommend to use our quickstart calculator tool to do the calculation instead:
https://www.ti.com/tool/download/SNVC240

Next to the compensation, another option is to increase the capacitance on the regulated AUX rail.

Please let me know if you have additional questions on this topic.
Best regards,
Niklas

We have set the Vin as 23V and Current as 0.53A. The parameters of transformer showed in the picture below.

Thanks for the quick feedback.
Is the winding ratio 1:1?

Best regards,
Niklas

Firstly, if I just want to know the Rcomp and Ccomp values, is it necessary to fill step 1 to 6?

Secondly, We dont use photo transistor so what should I fill in the related blank?

Hello Zhexi,

1. Yes, it is necessary to fill in step 1., 2., 3., and 4., as these parameters influence the required compensation.
Step 5. and 6. can be ignored.

2. You are using non-isolated feedback.
Please select 'non-isolated' in line 12 and put in the parameters for the regulated auxiliary rail into the "Load 1 Specification" (Cell 7N-11N & 40H-43H).
Load 2 and Load 3 will then be the isolated output rails which are indirectly regulated.

Best regards,
Niklas

Sry for the mistake picture, the winding ratio is 3:6 and the inductance is about 3uH, Rdc is 10mΩ. It would be great if you can provide theory values of the COMP pin RC.

Also， I want to know the principle about the role of specific RC values. It is related to output stability or something?

Hello Zhexi,

Thanks for the correction.
I filled in the calculator tool based on your parameters and received this result, which looks unstable:

The tool says a maximum of 21.9kHz crossover frequency should need be exceeded, as this could cause instability as well. (1/10 of the switching frequency)
But it is always possible to go lower to achieve higher phase margin.
In this example, I selected 10kHz crossover frequency as target in line 83.

We generally recommend a phase margin of 60 degree or more for a fully stable system. More on this can be found here:
https://www.ti.com/lit/an/slva381b/slva381b.pdf

I also attached the filled in calculator here:LM5156_Excel_Quickstart_Calculator_for_Flyback_Regulator_Design_e2e.xlsx

Please note that the output capacitance strongly affects the bode plot. I cannot see the capacitor values of the output rail, so I selected 10uF total capacitance as default.

Best regards,
Niklas

And in what situation, the gate pin will generate this waveform.

Hi Zhexi,

The are multiple possibilities for a gate driver signal like this.
In light load condition, the duty cycle can become very small down to minimum on-time of the device.
On the other side, if the load is too large, the device may trigger cycle-by-cycle overcurrent protection and the duty cycle gets limited because of this.

Can you give additional background for this question?
Is the output voltage not regulated properly, or is there another problem why this waveform causes problems?

Thanks and best regards,
Niklas

Hi Niklas,

Thank you for your instant reply. Does it mean the chip is broken or it is just working condition problem?

Best regards,

zhexi

Hi Zhexi,

This type of operation is a normal working condition. The IC is not broken.
Therefore I asked what kind of problem this operation mode is causing on customer side.

Best regards,
Niklas