UCC28782: Increasing RDM resistance causes device to not start

Hi Kaji-San,

Thank you for reaching out.

The choice of RDM is influenced by several other factors like Rcs, Lm turns ratio etc, as defined in the datasheet. So increasing only the RDM will potentially trigger OPP/OCP fault. For the increase in RDM you can try tuning the Rcs/ Ropp resistor to see if the fault goes to confirm if this due to one of the above.

Regards,

harish

Thanks for your reply.
Following your advice, I tried changing the RCS constant to half its current value, but the symptoms did not improve.
I checked, and found that the peak current of the primary winding does not reach the threshold at which protection operates.
From this, it is assumed that the protection is not due to OPP or OCP.

Is it possible that changing the RDM will cause the peak current of the primary winding to increase from before the RDM change, reaching the OPP or OCP threshold and causing protection to operate?
Also, is it possible that changing the RDM will change the OPP or OCP threshold?
Please let me know your opinion.

I also monitored other pins when this symptom occurred.
I would expect the RUN pin to go low when switching stops,
but it remains high, as shown in the attached waveform.
Is this RUN behavior normal?

Best regards,

Hi Kaji-San,

I think increasing the RDM will increase the PWMH ON time, which potentially reduces the PWML ON time and can lead to increased current for the same power level.

If it was OPP or OCP the controller should initiate shutdown with RUN signal going low. Can you please elaborate on much was the load when this was being tested at at? And how much is the increase in RDM beyond the recommended value?

Regards,

harish

Thank you for your reply.
This symptom occurs at startup, but the load current on the output side at startup is zero.
I have also previously attached the waveforms at startup, and the ISW current value is about 2.5A.
The RDM resistance value is 220kΩ. Is this value high?

As additional information, I can start up if I change the bypass capacitor on the CS pin from 100pF to 330pF.
Please let me know your opinion on whether this is related to the RDM value.
Also, please let me know your opinion on the factor that causes RUN to not go low even when switching stops.
Best regards,

Hi Kaji-San,

Thank you for the feedback.

The fact that you are able to startup upon increasing the capacitance indicates to much noise here. I would recommend slowing the turn on by increasing the gate resistor and this should help reduce the requirement for the capacitor and should help with reliable startup.

RDM will have an effect on the load range which the converter can support. Using too high of RDM will potentially restrict the maximum power upto which the output can be loaded. So if this too high, this can affect startup at lower power levels. 220kohm does not exceed the limit however this has to be correlated with the actual conditions.

I would have to check on the RUN signal though.

Regards,

Harish

Thank you for your reply.
We are still continuing to investigate this.
When a bypass capacitor was added to the CS pin and the RDM resistance was increased from 220 kΩ to 270 kΩ, the startup failure occurred again.
This time, the symptom occurred while the RUN pin voltage was rising at startup, so the startup failure occurred without any switching.
I have attached a waveform.
I would appreciate your opinion on the cause of this symptom.
Best regards,

Hi Kaji-San,

What is the effect when RDM is below the 220kohm? Is there any change with and without the bypass capcitor? Is the startup process normal under these conditions?

Could you please share this information too as this

Regards,

harish

Thank you for your reply.
If the RDM is below 220kΩ, there is no problem.
Also, a bypass capacitor is not required and the device starts up normally.
The reason for adjusting the RDM is that the Hi-side switch turns off before the primary side excitation energy is demagnetized.
So, I would like to increase the RDM constant further, but a problem occurs where the device does not start up.
Best regards,

Hi Kaji-San,

Thank you for the inputs. I will check and get back.

Regards,

Harish

Hi Kaji-San,

Sorry for the late reply. I am not sure on why RDM when increased very high causes the PWMH to turn off. One thing which we can try is to adjust the primary side clamp capacitor (increase this) to see if this helps prolong the high side switch ON time and see if this helps when increasing the RDM resistor.

Thank you

Regards,

Harish

Thank you for your reply.
I will look into the information you have provided.
The things I am continuing to consider are listed below.
I have found that increasing the RDM value causes a startup failure, but lowering the value of the pull-up resistor (Rvs1) on the VS pin improves this symptom.
From this, it seems that the causes of the startup failure can be narrowed down to the following:
OVP, OCP, brownout

I have a question related to the VS pin.
Brown-out/brown-in is determined by detecting the current flowing through Rvs1, so I think detection begins after switching starts.
At what point does the brown-in/brown-out function become effective?
Please tell me which timing in Figure 8-32 of the datasheet this is referring to.
I thought that brown-in would not function while switching is halted, but looking at item 8.4.14.1 of the datasheet, it seems to function even before switching starts.
Does brown-in function even when switching is halted?

Also, at what point during startup does Vcst correction by the OVP and VS pins become effective?
Could you please tell me which timing in Figure 8-32 of the datasheet this is?

Best regards,

Hi Kaji-San,

Sorry for my late reply. I checked with the team and it seems reducing the Rvs1 resistor is the correct way to go ahead when increasing RDM as there is a inverse relationship between the two.

Ideally when Vdd hits the ON threshold, there will be 4 pulses given by the controller. The first pulse will the check the voltage on the CS pin and if it is greater than the minimum value, the remaining three pulses are checked for the IVSL > IVSL (start) condition. If this happens, then Brown in condition is met and IC tries to continue to the start up process. 

From figure 8-32, interval F is the region in which these pulses will be counted to check if brown has occurred by checking the first 4 pulses.

Yes, as it is shown in datasheet page 59, the first three pulses during a transient condition when restarted are checked to continue operation rather than created another unnecessary shutdown.

Regards,

Harish

Thank you for your reply.
I understand that RDM and Rvs1 have an inverse relationship, but is there any information on what problems can be expected if this relationship is not met?
I would like to understand the cause of the problem I am experiencing.
Best regards,

Hi Kaji-San,

One thing which I got to know from the desgin team is The RDM resistance is designed by calculator, and the RDM resistance tuning range is around 30% of calculated result to make sure the internal tuner would not saturate. So increasing it too much beyond the recommended value might not cause the optimizer to work properly.

I will try checking other causes in between.

Thank you

Regards,

Harish

Thank you for your reply.
Please let me know if you have any additional information.
Best regards,