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.

Application circuit of UCC2817

Other Parts Discussed in Thread: UCC2817, UCC28050

Dear Specialists

My customer is considering UCC2817.

They have to move return GND.

Typical application circuit is not acceptable in their system

In this case, it has to change current sense circuits.

It seems to be concerned PKLIMIT,CAOUT,CAI and MOUT.

Could you please send me the example circuit.  

Best regards,

Shinichi

  • Dear Shinici,

    Do you know why the ground has to be moved? The circuit will not work with ground moved to the location you show in your drawing. The customer is right to be concerned with the pins you mention. The problem is if you move the ground to the new point, the boost switch current will pass through the sense resistor R14 and either fail or provide a distorted input to the CAI and Mout pin. The node you want to put ground on needs to float to measure the input current and voltage. Can you give more details to explain why the ground needs to be moved? Then an alternate solution might be possible.

    -Chuck
  • Dear Chuck

    Thank you for your reply.

    I confirmed the customer.

    The customer's system is consist of some boards.

    System GND is diode bridge return pin.

    If GND is changed for UCC2817, voltage difference has caused from other board's GND voltage.

    The customer isn't acceptable for changing other board's GND.

    So the customer wants to move GND for UCC2817

    Could you please advise the example circuit for moving gnd.

    It is also ok for adding external device or circuit.

    Best regards,

    Shinichi

  • The only solution for this problem that I can think of is to use a different type of PFC PWM controller. The  UCC28050 is a transition mode controller that does not need a current sense input to maintain unity power factor. The PWM controller gnd and the boost transistor source lead connect directly to the rectifier ground as you have in your drawing. The current sense resistor goes away. The only issue would be power output. Usually the transition mode PFC is used for 100 W or less because of the high peak currents in the inductor. You will need to either increase the size of the boost inductor or the switching frequency, or both, to achieve the output power you need.

    -Chuck

  • I have a better solution:

  • You may not need the sense resistor, you need to check. Also if the current sense transformer is too large you could use a hall effect sensor from Avago or LEM.

    -Chuck
  • Dear Chuck

    Thank you for your reply and offer an information.

    In this case, the customer prefer to CCM mode.

    They say DCM mode makes a large ripple, they couldn't accept their system.

    I think all the CCM device needs to use minus voltage for current signal.

    I appreciate your great help.

    Best regards,

    Shinichi.  

  • Dear Chuck

    Thank you for your reply.

    I'll confirm and feedback to the customer.

    I think PKLIMIT is necessary to change connection to current transformer.

    I appreciate your great help.

    Best regards,
    Shinichi
  • Shinichi, 

    I am glad to be of some help.

    In response to the following comment from your last message:

    "I think all the CCM device needs to use minus voltage for current signal."

    That is correct. You can achieve "minus voltage for a current signal" by wiring the secondary of the current sense transformer opposite to that of the primary.

    If the peak current is too high for current sense transformer, you can use a hall effect sensor.  FHS 40-P/SP600 is made by LEM and costs $ 2.49 in qty of 1000.
    The best solution, of course, is for your customer to isolate the grounds between his systems. He should use a "star" grounding scheme so that the "return" current from his other circuits do not run through the PFC sense resistor. This is probably a problem due to poor layout.
    Chuck
  • Dear Chuck

    Thank you for your e-mail.

    I understand using current transformer is the way to solve the customer's request. 

    In case of the high current value, hall sensor is recommended.

    I'll feedback to the customer.

    I appreciate your great help.

    Best regards,

    Shinichi