• State TI Thinks Resolved
  • +1 person also asked this people also asked this
  • Locked Locked
  • Replies 25 replies
  • Answers 1 answer
  • Subscribers 64 subscribers
  • Views 388 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

UCC256403: Question 1 for LLC_UCC256403

Akira Kurata
Intellectual 280 points
Part Number: UCC256403
Other Parts Discussed in Thread: UCC256303

Tool/software:

Question 1 for LLC_UCC256403

 

Dear TI engineer,

Thank you for your help.

This is Kurata.

I have a question regarding the UCC256403_Maximum frequency control settings.

 

  1. UCC256403 peripheral design conditions

 

1-1) Power supply 2 output: ① 5V 10A, ② 24V 2A (24V uses DC/DC converter for winding output)

 

(1 transformer conversion method with feedback control of 5V output)

 

1-2) Total output capacity: 98W max

 

1-3) Input voltage: 390VDC (input uses PFC control output)

 

1-4) LLC transformer design value

 

(Designed with reference to UCC25630-1EVM evaluation board transformer)

 

① Lm: 840μH

 

② Leakage inductor (external): 50μH

 

③ Approximate leakage inductance of each output winding: 5V_20μH, 24V_40μH

 

④ Resonant voltage input voltage division capacitor constant of IC_VCR terminal: Upper C_150pF, Lower C_0.015μF

 

  1. Key points of product design specifications

 

2-1) Minimize the no-load power of the product. (Currently about 4.5W)

 

2-2) For control at no load or light load, do not use burst mode control, but use continuous mode.

 

2-3) Ensure that the conversion efficiency of the entire product, including PFC, is 86% or more at rated load.

 

2-4) The LLC circuit is based on the UCC256303 design, but there are some problems with the IC's internal control, so UCC256403 will be used instead.

 

  1. Issues and questions regarding the design and evaluation of using UCC256403

 

3-1) Previously, I was told that the maximum control frequency of this device in terms of design is up to 400kHz, but in the prototype evaluation, it stopped at about 200kHz. Is there a way to control it beyond this frequency?

3-2) In the transient response of a sudden load change of 5V output from 0 to 100%, there is a sudden drop and rise to about ±200mV (4%) when changing to 0%, which does not fall within the stability standard of ±3%. How can this be improved?

3-3) The excitation inductance of the transformer currently being used is set to 840μH. Under these conditions, would it be difficult to raise the IC's control frequency to 200KHz or higher and control the control frequency to nearly 400kHz to ensure a stable transient response when the load changes suddenly from rated load to no load? If so, please advise me of specific measures to improve the situation.

  • 0
    TI__Intellectual 1190 points

    Hi Kurata,

    Can you please share the schematic with us to better understand the problem?

    Also, please share the design calculator used for this design?

    Also, please share the waveform when the converter stopped at about 200kHz...Please capture LO, HO, Resonant current, load current in the scope. Also please let me know in which load condition, converter is achieving 200kHz frequency?

    Regards,

    Sougata

  • 0 in reply to Sougata Nayak
    Intellectual 280 points

    hi　Sougata
    Thank you for your reply.
    I have prepared the documents you requested and will send them to you.
    I look forward to your further response.
    Please contact me if you have any questions.LLC_UCC256403 Evaluation circuit diagram.pdfUCC256403_LLC_Waveform when load changes suddenly.pdfコピーUCC25640x Design Calculator Rev2.0.5V.24V.20240703.xlsx

  • 0 in reply to Akira Kurata
    TI__Intellectual 1190 points

    Hi Kurata,

    Thanks for giving the waveforms. 

    1. In the waveforms, HO, LO pin waveform of the IC are not given. Can you please provide me HO, LO waveform of the IC and Gate-source voltage of FETs TR3 and TR5 while performing load transients? 

    2.  can you please provide me Gain crossover frequency of the designed converter and Phase margin of the converter?

    3. Can you please provide ESR of output capacitor of 5V output rail?

  • 0 in reply to Sougata Nayak
    Intellectual 280 points

    Hi, Sougata.
    This is Kurata.
    Thank you for your help.
    I have prepared the answer to your question, so I will reply in the attached document.
    I look forward to your further assistance.UCC256403 LLC load change operation waveform and question and answer.pdf

  • 0 in reply to Akira Kurata
    TI__Intellectual 1190 points

    Hi Kurata,

    Thanks for your reply. I can see that ESR of output capacitor is very high. 

    1. you can select an output capacitor whose ESR is much low to improve the transient pick. With the existing 15m ohm ESR and 10A load change, you will get 150mV change in output voltage for best case scenario. This can be improved by reducing ESR of output capacitor.

    2. You can also increase gain crossover frequency to get better response.

    Thanks,

    Sougata

  • 0 in reply to Sougata Nayak
    Intellectual 280 points

    Hello Sougata
    This is Kurata.
    I checked your answer, but unfortunately it wasn't the advice I expected.
    I agree with what you pointed out that the ESR of the output capacitor causes a sudden change in voltage in response to the load changing from 100% to 0%, which is a transient response.
    However, I don't think this issue is caused by the ESR itself.
    As I previously attached the waveform, I think this issue was triggered by the LLC HO and LO output stopping just before it occurred. I think this happens in the following sequence.
    1. The LLC output HO and LO stop due to some IC control problem. ⇒2. Immediately after, the 5V output goes from 100% to 0% with no load due to the sudden load change, so the voltage rises by several tens of millivolts. ⇒③ After about 150μsec has elapsed since the LLC output stopped, the LLC restarts, but at this time, the fluctuation of the LLC VCR terminal voltage is almost 0 volts, so the HO and LO outputs suddenly resume operation at maximum power. ⇒④ As a result, even though the output has reached the set voltage, the output capacitor is instantly charged with excess power, so I understand that a transient voltage of about 200mV due to the ESR occurs.

    At this time, the control of the error amplifier on the output side is unloaded and the control response is slow, so the fact that high-speed feedback to the LLC is not possible is also a factor that makes it difficult to suppress the 200mV increase.

    Based on your advice, I connected two capacitors with the same characteristics in parallel, reduced the ESR by half, and performed a sudden load change test from 100% to 0%, but the problem did not improve at all.

    Therefore, I think your explanation is not conclusive about the problem.

    I think that the cause of this phenomenon is some kind of abnormal operation of the LLC device.

    One of the reasons for this is that the output voltage was evaluated at 4.7V, which is lower than the set value of 5V. Also, although there is some dummy resistance in the output, I wonder if the control capability of the IC operation (continuous control without burst) when the load suddenly changes to 0% has reached its limit? Or, I wonder if there is a deviation into the burst operation region due to exceeding the dynamic range of the detection voltage of the VCR terminal.
    Taking these points into consideration, I would appreciate your further advice on how to improve this malfunction.

  • 0 in reply to Akira Kurata
    TI__Intellectual 1190 points

    Hi Kurata,

    we are looking into it. In the meantime, can you please confirm that are you using the burst mode or not in this IC?

    Regards,

    Sougata

  • 0 in reply to Sougata Nayak
    Intellectual 280 points

    Thank you for your help, Sougata-san.

    This is Kurata.

    I have checked the burst mode setting of the BW pin.

    The results are as shown in the attached data.

    I would appreciate your further instruction.

    Please contact me if you have any questions.UCC256403_LLC BW pin waveform observation.pdf

  • 0 in reply to Akira Kurata
    TI__Intellectual 1190 points

    Hi Kurata-San,

    Thanks for sharing the update.

    We are working on it. In the meantime, can you please share LL/SS pin voltage and ISNS pin voltage at the same moment?

    Regards,

    Sougata

  • 0 in reply to Sougata Nayak
    Intellectual 280 points

    Hello, Mr. Sougata.
    This is Kurata.
    Thank you for your help.
    As per your request, I have observed the SS pin voltage waveform when a burst occurs, so I am attaching the data.
    Please check it. Also, please contact me if you have any questions.
    I would appreciate your continued advice on how to improve the burst operation.UCC256403_LLC SS pin waveform observation.pdf

  • 0 in reply to Akira Kurata
    TI__Intellectual 1190 points

    Hi Kurata,

    We are working on it to recreate the issue in our bench. 

    Please give us sometime. I will update you the results.

    Regards,

    Sougata

  • 0 in reply to Akira Kurata
    TI__Intellectual 1190 points

    Hi Kurata-San,

    Can you please let me know value of BMTH and BMTL chosen by you in designing this converter?

    Regards,

    Sougata

  • 0 in reply to Sougata Nayak
    Intellectual 280 points

    Dear Sougata
    Thank you for your help. This is Kurata.
    Regarding your request regarding BMTH and BMTL, I am unable to give you a definite answer, but I have attached my answer.
    If you have any questions, please contact me.
    Thank you again for your assistance.Answers about UCC256403_BMTH and BMTL settings.pdf

  • 0 in reply to Akira Kurata
    TI__Intellectual 1190 points

    Hi Kurata San, 

    I am going through the result. Can you please give me the waveform of HO, LO, Vout, Iout of the converter in no load, slight load conditions, mentioning switching frequency?

    Regards,

    Sougata

  • 0 in reply to Sougata Nayak
    TI__Intellectual 1190 points

    Hi Kurata San, 

    What I understood, you have disabled the burst mode in this controller and Set BMT_H as 1.2V. 

    Please confirm this is correct or not?

    Regards,

    Sougata

  • 0 in reply to Sougata Nayak
    Intellectual 280 points

    Sougata-san

    Thank you for your help.

    This is Kurata.

    Regarding the confirmation of the setting value of BMT_H, I have never answered that BMTＨ is set to 1.2V.

    When I was asked to measure the waveform of the LL/SS terminal in a previous load change test from 5V100% to 0%, I remember explaining in my answer that when a burst occurs, the voltage of the LL/SS terminal momentarily drops from about 3.5V to 1.2V.

    However, I did not mention that it was set as a design value.

    I explained that the setting is not the value setting of BMT_H, but the setting to disable bursts, and that option 7 (BMTH/L ratio): 0.4 in the data sheet is set.

    That is what I just uploaded to e2e.

    I have also attached the waveform data you requested, so please check it as well.

    However, I did not measure the output current because it is not easy to measure.

    Please contact me if you have any questions.

    I would appreciate your further instruction.

    UCC256403_LLC primary current, Ho, Lo, control frequency measurement　.pdf

  • 0 in reply to Akira Kurata
    TI__Intellectual 1190 points

    Hi Kurata San, 

    Can you please change VCR pin capacitors to following:

    a. CVCRupper=33pF,

    b. CVCRlower=10000pF

    And then can you please check if it works or not?

    Regards,

    Sougata

  • 0 in reply to Sougata Nayak
    Intellectual 280 points

    Hi Sougata-san

    This is Kurata.

     I made the suggested constant change and checked the operation.

    At the lower limit of the 5V output voltage setting of 4.7V, the transient characteristics due to a sudden load change from 10A to 0A were worse than before the change, and no improvement was seen.

    Before the constant change: +200mV/-100mV, after the constant change: +300mV/-1.4V

    This is probably due to the circuit feedback response being unable to respond properly when the constant is changed.

    At this point, I think it is difficult to make the feedback system respond any faster due to the circuit configuration of the secondary side filter.

    Based on the above, is this the limit of further improvement?

  • 0 in reply to Akira Kurata
    TI__Intellectual 1190 points

    Hi Kurata,

    considering the point that feedback can not be changed now,

    Can you please connect some noise cancellation capacitors in ISNS pin in parallel with the resistor, without changing the CVCRupper  and CVCRlower?

    Regards,

    Sougata

  • 0 in reply to Sougata Nayak
    Intellectual 280 points

    Hello, Mr. Sougata.

    This is Kurata.

    Thank you for your help.

    I have researched and considered your advice from the other day, so I am replying to it in the attached file.

    I would appreciate your further advice and reply.

     About CR connection adjustment of UCC256403_Isns pin.pdf

  • 0 in reply to Akira Kurata
    TI__Intellectual 1190 points

    Hi Kurata San,

    This is due to at lower load condition, ZCS is getting triggered and switching of controller gets stopped due to noise in current sensing in ISNS pin. Therefore I suggested to have better filtering in ISNS pin. When you are also moving resistance in ISNS pin, it helps to reduce noise in ISNS pin preventing ZCS triggering. For overshoot to go down, you need to ensure that at no load, Vfbreplica should be above BMTL, but below BMTH from design calculator to maintain regulation, and accordingly you may have to change your compensation circuit, thus I suggested the solution about changing capacitor in VCR pin. If this can not be done, you can choose faster bandwidth to maintain regulation. More than that, I dont think there is any scope of improvement.

  • 0 in reply to Sougata Nayak
    Intellectual 280 points

    Sougata-san
    This is Kurata. Thank you for your help.

    I am confused because I did not get a specific answer to my question about the contents of my previous inquiry.

    Also, our evaluation board is broken and it will take some time to repair it, so it will take some time to investigate and consider the advice you gave.

    In the meantime, I will ask again.
    I understand that in order to improve the control response when the load suddenly changes by 5V, it is necessary to change the setting of the capacitor on the VCR pin and to review the control compensation, as you said.
    However, since you did not give a specific answer to my previous question, I do not logically understand the situation that is currently occurring when the load suddenly changes from 100% to 0% at 5V, so I will ask again.
    When the load suddenly changes, the voltage on the VCR pin drops and the control frequency increases. In conjunction with this, the feedback voltage on the FB pin also gradually begins to drop from about 5.2V. However, even though the voltage at the FB terminal has dropped, "the control frequency of the VCR pin is kept at around 200kHz, and then the control frequency shifts downward. ⇒ Transmission power increases. ⇒ Output voltage overshoot increases. ⇒ FB signal suddenly drops." So what is the reason why the control frequency does not rise to 300kHz and the power is not reduced?
    Is it because the BMT H/L settings do not match?
    Currently, the BW pin setting has R (Lower) set to 3.3kΩ and burst mode is disabled, so the external resistance settings for the LL/SS pin are R101 (Upper) = open, R104 (Lower) = 1MΩ. Is there a problem with these settings?
    The LL/SS terminal is fixed to 3.5V during operation and burst mode is not selected, so I think the resistance setting of this pin has no effect on the operation of the VCR terminal.
    Am I interpreting this incorrectly?
    Please explain again.

  • 0 in reply to Akira Kurata
    TI__Intellectual 1190 points

    Hi Kurata San,

    please wait for some hours. I will again explain to you.

    Regards,

    Sougata

  • 0 in reply to Akira Kurata
    Intellectual 280 points

    Sougata san

    This is Kurata. Thank you for your help.

    I have readjusted the capacitor connection on the ISNs pin as you advised me the other day, so please check the attached data and give me your opinion and advice again.

    Please contact me if you have any questions.Re-adjust the CR connection of the ucc256403_Isns pin.pdf

  • 0 in reply to Akira Kurata
    TI__Intellectual 1190 points

    Hi Kurata San, 

    I will get back to within 1-2 days. Please have patience !!!!

    Regards,

    Sougata