Hi Lucas,

Thank you for the query. Can you please let us know the input voltage specifications so that we can verify the calculations?

Regards,

Harish

The input voltage range is the same as the PMP22557.

50Vdc – 150Vdc

Hi Lucas,

Your R1, R3 and R5 look to be ok. But I feel your inductance is on the higher side as per my calculations. The peak current Ipp was around 20A for Dmag of 67% as the converter will operate in DCM mode. Can you share how you got the 100uH value or share your excel calclations to proceed further.

Thank you

Regards,

Harish

Harish,

I didn't have much of a method to find the inductor value other than basing it off of the schematic below from a similar reference design.
https://www.ti.com/lit/df/tidm659/tidm659.pdf?ts=1662339986536. I just chose a value between that and the PMP22557.

What value did you calculate? Is there any collateral dedicated toward designing this kind of buck converter? Any help would be appreciated. 

I have 47uH inductors arriving tomorrow. Do you think I need even smaller?

Hi Lucas,

You can use our calculator tool to get a rough estimate of the inductance which will be required.

For PMP22557, we see they have used 47uH.

This is obtained by setting Dmag = 67% Tr= 2us and Dutymax will be 24%. The peak current will be 1.75A and the corresponding inductance will be close to 50uH. 

This can also be verified by the inductor on time voltage equation. V=Ldi/dt

L = ((50-12)/1.75)*(0.2/83000) ~ 50uH (0.2 on time duty, 83khz is Fsw)

You can refer to the following document which shows an example with a different controller.

https://www.ti.com/lit/an/snva750/snva750.pdf?ts=1662542115301&ref_url=https%253A%252F%252Fwww.google.com%252F

Please let me know if you have any questions.

Thank you

Regards,

Harish

Harish,

Thank you for that. However, I still have questions. 

There seem to be some discrepancies in the linked document. 

Rpk is not defined and Vccr = 260mV (not 540V) as per the data sheet for the UCC28910. Vcste = 540V but this parameter is not defined in the datasheet for the UCC28712. T is also not defined for the equation for calculating the inductor value. 

Where does Dmag = 67% come from as well as the other chosen values for Tr and Dutymax? How exactly did you calculate Ipk? 

The calculation for the inductor suggests that a changing Vout also requires a changing Inductor value. Does this mean this design is not suited for a variable output where a potentiometer can be adjusted to alter the output voltage?

Hi Lucas,

I think you are confusing two things. The design document for non isolated buck which I pointed you to is just a reference for designing a non isolated buck. The design equations mentioned there are correct and it is according to the datasheet of UCC28910. The choice of switching frequency is as per your system specification. This document takes it as 60khz. Use equation 26 in the datasheet of UCC28910 with Nps as 1 you get close to 1.3kohm with a load current of 100mA (given in system specifications). Substituting this in equation 1 and 2 of the document we will get inductance value but this is not close to what is mentioned there.But the procedure to calculate inductance is correct. 

Let's come back to the case of UCC28712, you have to define your Fsw. The max allowable value is 100khz. The excel sheet row D20 can be used to enter this value. Datasheet page 19 mentions why to take Tr as (500khz freq.) as a good initial assumption. The Dmag time recommended in the datasheet is 0.425 for mainitaining CC mode. Ideally for the low line voltage of 50V and maximum output current of 5A, the inductor is designed for this worst case boundary mode operating point. But based on PMP22557, I had suggested Dmag-67% as it will be operating deep in DCM mode based on the waveforms and its configuration as non isolated buck. Peak current can be calculated as shown in excel sheet Vcst(max)/Rcs.

Changng Vout is going to change your output power from 60 to 180W which is going to require change in sense resistor due to increased Ipeak also. Using same inductor as first case is going to increase the ripple also and hence it will be better if the circuit is designed for the two output ranges separately.

Thank you

Regards,

Harish

Harish,

I realize now that the excel sheet you have been referring too is here.
https://www.ti.com/lit/zip/sluc590. I will study it further to get a better understanding of the parameters. 

I have changed to the 47uH inductor. It's closer to the desired 12V output but only regulating to about 10.7V even after changing R1 to 68k which should yield a voltage a little above 12V. There is some amount of functionality it's just not regulating to the intended output voltage. Maybe something to do with the feedback?

1. This is the waveform on Vs. 

2. This is the DRV pulse train that is only happening at about 32Hz.

I am willing to take waveforms at any node you think is useful to look at. Is there some other waveform that would be useful in diagnosing the issue? 

Something still seems off about the values I'm using for my design.

Rcs is calculated as follows based on the datasheet

Since an inductor is used, Nps=1 and efficiency = 1. Vccr = 0.33V and Iocc = 5A. This results in a value of 0.033ohm which is what I'm using.

Now for the peak current calculation. 
Vcst(max) = 0.78V as per the datasheet. I(peak) = Vcst(max)/Rcs which comes out to a value of 23.6A.

Now for the Inductor calculation.
L = ((50-12)/23.6)*(0.2/83000) = 3.88uH

This calculation for inductance seems really low but could be the reality of trying to scale this design for such high output current. Does this seem right?

This also means Rlc = 5.9k rather than 1k as currently in my schematic. 

Referring to the excel sheet for help is difficult because I can't change any of the non-yellow values to adapt it for a buck design. Also, the names of the parameters are relevant to transformers rather than an inductor making it difficult for me to translate to this particular application,  

Hi Lucas,

Thanks for confirming on the calculations which I had mentioned earlier. Clearly your inductor needs to be optimized. I have attached the excel sheet with your design parameters. Please  change the required components as per the calculations before re-running the tests and let us know your observations.

0310.UCC28710-16 Excel Design Tool 9 29 15.xlsx

Regards,

Harish

Harish,

Ok. I will order new components to test. They should arrive Thursday. I hope to have results on Friday. 

Hi Lucas,

Sure, thank you

Regards,

Harish

Harish,

I have tested again with both 3.3uH and 4.7uH inductors. Neither resolved the issues. I have also tried different values of Rlc. 

The output still doesn't regulate properly. The output voltage goes from ~6V to 9V if I sweep in the input voltage from 50V to 120V. So not the desired 12V output.

The DRV pulses are similar to the figure below


https://www.ti.com/lit/an/sluaac5/sluaac5.pdf?ts=1663447270288&ref_url=https%253A%252F%252Fwww.google.com%252F

This seems as if some kind of fault happening. Do you know what I could check or observe further in order to diagnose the fault? 

Hi Lucas,

Thank you for the reply and observations.

Definitely the IC is working in the fault mode, attempting to restart but process continues as the fault is not cleared. 

Can you please check the Vs waveform for any OV event? If greater than 4.6V it is output overvoltage.

Please probe CS pin to check for overcurrent fault (>1.5V).

The ton(min) should be greater than leading edge blanking tcsleb mentioned in the datasheet.

Input undervoltage where controller starts continuous operation. (Input undervoltage).

Could you please share Vs pin and Cs waveforms to help debug this issue?

Regards,

Harish