UCC28019A: Input Current Distortion when a load is connected

Shinichi Inoue

Guru 19701 points

Part Number: UCC28019A

Dear Specialists,

My customer is evaluating UCC28019A and encountering the input current distortion.

I would be grateful if you could advise.

---

I am evaluating TI's PFC controller (UCC28019ADG4), but the input current waveform is distorted immediately after the load is turned on.

Could you please see attached excel file (UCC28014ADG4_Input_Current_Distortion_20200915).

The waveforms at 100V 60Hz and 264V 50Hz are shown in Figures 1 and 2, respectively.

At 100V, some distortion is seen in the first pulse, and then distortion is seen near zero cross.

At 200V, a large distortion is seen at about 8 pulses, and then distortion is seen near the zero cross.

Fig. 3 shows the circuit diagram.

Could you please let us know the cause and the solution.

UCC28019ADG4_Input_Current_Distortion_20200915.xlsx

About the shematic,

C15 is originally 1000pF, but to 4700pF for the noise suppression.

Current Sense Resistor (R54) is calculated 34mohm, but change to 27mohm because current limit is active earlier than expected.

Regarding circuit condition, could you please see the attached file (UCC28019A condition.xlsx)

UCC28019A condition.xlsx

---

I appreciate your great help in advance.

Best regards,

Shinichi

over 3 years ago

0 Mike O' over 3 years ago

TI__Guru 66135 points

Hello,

I notice in the excel design tool that you had entered the actual inductance used was 600 uH. The

HKBS-24D160-7510 is a 750 uH inductor. You might to want to re compensate the voltage loop and current loop based on that inductance value.

The input current distortion looks like the PFC boost might be hitting cycle by cycle peak limit when the load is applied. The other issue could be there is too much voltage droop on the output capacitor during load transients. You might want to study the rectified input voltage, VCOMP, ISENSE and the boost output voltage during startup it may point you to why the input current is distorted.

When it comes to crossover distortion this generally due to amplifier offsets in the PFC controller. Using a smaller current sense resistor/signal will make this cross over distortion larger. If you want to reduce this distortion you can increase your current sense resistor.

Regards,

0 Shinichi Inoue over 3 years ago in reply to Mike O'

Guru 19701 points

Hi Mike,

Thank you for your reply.

I attach a DC Bias specification of the inductor.

The customer meitioned inductance value at 20A is 190uH.

They use in series, so 380uH at 20Apeak.

I'll make a request the customer to obtain the waveforms you mentioned, input voltage, VCOMP, ISENSE and boost output voltage during startup.

When I obtain, I'll sent to you.

I appreciate your great help and cooperation.

Best regards,

Shinichi

0 Shinichi Inoue over 3 years ago in reply to Shinichi Inoue

Guru 19701 points

Hi Mike,

I could obtain the the waveforms, input voltage, VCOMP, ISENSE and boost output voltage during startup.

Could you please let us know the solution.

Also the customer has a question about current limit. d

The input current peak at 100V exceeds 10A, but there is no distortion.

On the other hand, the input current of 230V is distorted even though it does not exceed 10A, it seems to inactive current limit

Please see the input current waveforms of fig.1-2, 2-1 and 1-10, 2-10 in the attached Excel.

UCC28019ADG4_load transient waveform _20200916.xlsx

When we change the phase compensation, could you please advise how to change it.

I appreciate your great help.

Best regards,

Shinichi

0 Shinichi Inoue over 3 years ago in reply to Shinichi Inoue

Guru 19701 points

Hi Mike,

I could obtain the the waveforms, input voltage, VCOMP, ISENSE and boost output voltage during startup.

Could you please let us know the solution.

Also the customer has a question about current limit. d

The input current peak at 100V exceeds 10A, but there is no distortion.

On the other hand, the input current of 230V is distorted even though it does not exceed 10A, it seems to inactive current limit

Please see the input current waveforms of fig.1-2, 2-1 and 1-10, 2-10 in the attached Excel.

0131.UCC28019ADG4_load transient waveform _20200916.xlsx

When we change the phase compensation, could you please advise how to change it.

I appreciate your great help.

Best regards,

Shinichi

0 Mike O' over 3 years ago in reply to Shinichi Inoue

TI__Guru 66135 points

Hello,

I looked at the schematic and notice that the design does not have a bypass diode to handle inrush current. I wonder if this startup behaivor has to do with the inrush current. Could you put an inrush diode across L2 and D5 to see if the issue goes away?

I am thinking during inrush the inductor might be saturating and the FETs are turning on into inrush current. Add the bypass diode and hopefully that removes the issue.

Regards,

0 Shinichi Inoue over 3 years ago in reply to Mike O'

Guru 19701 points

Hi Mike,

Thank you for your reply.

I'll share your suggestion with the customer.

I'd like to confirm regarding the diode between L2 and D5 you mentioned,

is it the diode surrounded by the red line in the figure below?

Figure 26. Design Example Schematic of datasheet

---

I appreciate your great help and cooperation.

Best regards,

Shinichi

0 Mike O' over 3 years ago in reply to Shinichi Inoue

TI__Guru 66135 points

Hello,

That is the correct diode and placement.

Regards,

0 Shinichi Inoue over 3 years ago in reply to Mike O'

Guru 19701 points

Hi Mike,

Thank you for your reply,

I could obtain the waveform with adding bypass diode.

The distortion was not improved.

Could you please see the attached excel file.

UCC28019ADG4_with bypass diode(L2-D6 ).xlsx

Could you please advise where to improve.

---

I appreciate your great help and cooperation.

Best regards,

Shinichi

0 Shinichi Inoue over 3 years ago in reply to Mike O'

Guru 19701 points

Hi Mike,

Thank you for your reply,

I could obtain the waveform with adding bypass diode.

The distortion was not improved.

Could you please see the attached excel file.

2465.UCC28019ADG4_with bypass diode(L2-D6 ).xlsx

Could you please advise where to improve.

---

I appreciate your great help and cooperation.

Best regards,

Shinichi

0 Mike O' over 3 years ago in reply to Shinichi Inoue

TI__Guru 66135 points

Hello,

So we eliminated it was inrush causing the issue.

You had checked the VCOMP and we did not see an issue there.

You might want to check ICOMP to make sure it is compensated coretly?

Regards,

0 Shinichi Inoue over 3 years ago in reply to Mike O'

Guru 19701 points

Hi Mike,

Thank you for your reply.

I'll ask the customer check ICOMP pin and confirm conpensation..

When I get this, I'll feedback to you.

I appreciate your great help.

Best regards,

Shinichi

0 Shinichi Inoue over 3 years ago in reply to Mike O'

Guru 19701 points

Hi Mike,

I could obtain the waveform of ICOMP pin at load connected.

Could you please see attached file.

UCC28019A_load transient Icomp waveform_20200929.xlsx

Could you please advise the problem and solution.

I appreciate your great help and cooperation.

Best regards,

Shinichi

0 Shinichi Inoue over 3 years ago in reply to Shinichi Inoue

Guru 19701 points

Hi Mike,

This is a reminder that I am looking forward to waiting for your advise.

I appreciate your great help and cooperation.

Best regards,

Shinichi

0 Mike O' over 3 years ago in reply to Shinichi Inoue

TI__Guru 66135 points

Hello,

It looks like the current behavior is tracking the Icomp output.

Maybe Icomp is not compensated correctly. Could you have the customer double check it with the excel tool?

Regards,

0 Shinichi Inoue over 3 years ago in reply to Mike O'

Guru 19701 points

Hi Mike,

Thank you for your reply.

I'll confirm the customer's design with the customer by the excel tool again.

By the way, is there a possibility to damage that the distortion of the input current at the load transient happen.

Because the customer don't have much time to degug.

In some cases, it mey be mass production as it is if there is no possibility to damege.

I appreciate your great help and cooperation.

Best regards,

Shinichi

0 Shinichi Inoue over 3 years ago in reply to Shinichi Inoue

Guru 19701 points

Hi Mike,

I'm still looking forward to waiting your advise.

Regarding this input current distortion, is there a possibility to damage at the load transient happen.

We do not believe that damage will occur.

Is it correct?

I appreciate your great help and cooperation.

Best regards,

Shinichi

0 Shinichi Inoue over 3 years ago in reply to Shinichi Inoue

Guru 19701 points

Hi Mike,

I'm still looking forward to waiting your advise.

Regarding this input current distortion, is there a possibility to damage at the load transient happen.

We do not believe that damage will occur, it only reduces power factor and efficiency.

Is it correct?

Please respond at your earliest convenience

I appreciate your great help and cooperation.

Best regards,

Shinichi

Power management

Power management forum

UCC28019A: Input Current Distortion when a load is connected