Tool/software:
Hello Team,
As shown below gate pulses observed in our power converter during startup with no load condition. We operating in peak current mode control, Could you clarify on CH-3 & Ch-4 PWM output nature?
In PSFB , we should get 50% fixed duty at primary side right. Could you help us out?
CH1- QA gate , CH2-QB gate , CH3 - QC gate & CH4 - QD gate.
Hello Mike,
It seems CS measurement within limit as attached below images, But lot of variation observed in COMP node.
Hello,
Your current sense transformer signal does not look correct. It should be symmetrical every cycle. Your signal is trapezoidal then triangular. This incorrect CS signal is causing the OUT D to be less than 50 % D sporadically. The UCC28951 is actually OUT D and OUT C to correct the duty cycle based on the CS signal compared to the COMP voltage.
The following waveform shows the voltage across the input of PSFB transformer (CH1) and the voltage at the CS pin (CH4). This is what the current sense waveform should look like. I think you might have an issue with how your current sense transformer is setup. CH2 is the voltage at the gate of QA and CH3 is from the voltage at the gate of of QD. These waveforms were taken on a 600 W reference design using a 2 to 1 gate driver transformer to drive the FETs on the H Bridge.
I believe there could be something wrong in how you are sensing your current and/or in your schematic. The following link will bring you to an application note that goes through the step by step design process of a phase shifted full bridge using the UCC28950 and UCC28951. You can use this application note to check your schematic and current sense transformer setup.
https://www.ti.com/lit/pdf/slua560
Please note the UCC28950 and UCC28951 are pin to pin compatible. The only difference is the UCC28951 is design for duty cycle greater than 90%.
The following link will bring you to an application note that describes the differences between the two devices.
https://www.ti.com/lit/an/slua853/slua853.pdf
I do have another recommendation. It might be useful in the debug process to have a working design to evaluate and compare to your design.
You might want to order the 600 W phase shifted full bridge evaluation module UCC28950EVM-442. You can order this evaluation module from ti.com.
The following link will bring you to the evaluating modules User's Guide which contains critical waveforms, schematic, bill of material (BOM) and layout.
https://www.ti.com/lit/ug/sluu421a/sluu421a.pdf
Regards,
Hello Mike,
As per previous thread measurement taken with 100 ohm load. We will just Ignore for a moment.
Refer following images for no load startup , even though CS signal less than 0.2V , still observing Out D duty variation, Anything are we missing here.
Hello,
Your current sense transformer signal does not look correct. It should be symmetrical every cycle. Your signal is trapezoidal then triangular. This incorrect CS signal is causing the OUT D to be less than 50 % D sporadically. The UCC28951 is actually adjusting OUT D and OUT C to correct the duty cycle based on the CS signal compared to the COMP voltage.
The following waveform shows what the current sense signal should look like (CH4).
I believe there could be something wrong in the way you are sensing your current and/or with your schematic. The following link will bring you to an application note that goes through the step by step design process of a phase shifted full bridge using the UCC28950 and UCC28951. You can use this application note to check your schematic and current sense transformer setup.
https://www.ti.com/lit/pdf/slua560
Please note the UCC28950 and UCC28951 are pin to pin compatible. The only difference is the UCC28951 is design for duty cycle greater than 90%.
The following link will bring you to an application note that describes the differences between the two devices.
https://www.ti.com/lit/an/slua853/slua853.pdf
I do have another recommendation. It might be useful in the debug process to have a working design to evaluate and compare to your design. The differences in performance, layout and schematic could help you figure out why your design is not behaving as expected.
We do have a 600 W evaluation module using the UCC28950, UCC28950EVM-442 that is a working design that you could use to compare to your design as mentioned above. You can order this evaluation module from ti.com. The following link will bring you to the evaluating modules User's Guide which contains critical waveforms, schematic, bill of material (BOM) and layout.
https://www.ti.com/lit/ug/sluu421a/sluu421a.pdf
Regards,
