+ Post New Thread
Results 1 to 6 of 6
  1. #1
    Member level 2
    Points: 258, Level: 3

    Join Date
    Mar 2019
    Posts
    52
    Helped
    0 / 0
    Points
    258
    Level
    3

    Full bridge converter, how to improve crest factor.

    I constructed the following phase shift full bridge converter. Tried various passive filter combinations before and after the rectifier bridge. But the crest factor is really bad. Attached are the simulation results of phase voltage vs. current.

    Any ideas how to improve the crest factor especially with passives filters?.

    Should I use active power factor correction to get an improvement for the crest factor.

    Will active power factor correction solve this problem (crest factor) completely?.



    Click image for larger version. 

Name:	Converter-THD-1.png 
Views:	24 
Size:	98.1 KB 
ID:	151765


    Click image for larger version. 

Name:	fullBridge.jpg 
Views:	21 
Size:	386.0 KB 
ID:	151764

    •   AltAdvertisement

        
       

  2. #2
    Super Moderator
    Points: 51,724, Level: 55

    Join Date
    Apr 2011
    Location
    Minneapolis, Minnesota, USA
    Posts
    12,696
    Helped
    2525 / 2525
    Points
    51,724
    Level
    55

    Re: Full bridge converter, how to improve crest factor.

    Your schematic appears to apply a modified sine wave to a series LC. I've played with similar simulations in a hope to find an easy way to regulate voltage. I found a problem because the LC tries to continue resonating as current is abruptly shut off. Even when it's only an inductor, it wants to continue pulling current at shut-off through all neighboring components. If no freewheeling diode (mosfet, snubber network, etc.) is installed then the inductor generates unwanted high voltage spikes.

    It may be more effective if you gate the H-bridge so that it applies plain AC square waves. Do you want to sense the resonant frequency of your LC combo? Then install a low-ohm resistor in series. This allows you to sense each zero crossing of current at the exact time when current falls to zero. That is the cue to change the state of your H-bridge.



    •   AltAdvertisement

        
       

  3. #3
    Member level 2
    Points: 258, Level: 3

    Join Date
    Mar 2019
    Posts
    52
    Helped
    0 / 0
    Points
    258
    Level
    3

    Re: Full bridge converter, how to improve crest factor.

    Quote Originally Posted by BradtheRad View Post
    Your schematic appears to apply a modified sine wave to a series LC. I've played with similar simulations in a hope to find an easy way to regulate voltage. I found a problem because the LC tries to continue resonating as current is abruptly shut off. Even when it's only an inductor, it wants to continue pulling current at shut-off through all neighboring components. If no freewheeling diode (mosfet, snubber network, etc.) is installed then the inductor generates unwanted high voltage spikes.

    It may be more effective if you gate the H-bridge so that it applies plain AC square waves. Do you want to sense the resonant frequency of your LC combo? Then install a low-ohm resistor in series. This allows you to sense each zero crossing of current at the exact time when current falls to zero. That is the cue to change the state of your H-bridge.
    Sorry I forgot to add the freewheeling diodes. They are actually there. I send the whole schematic.

    Click image for larger version. 

Name:	Converter-schematic-rectified-current-voltage.png 
Views:	20 
Size:	33.7 KB 
ID:	151783

    - - - Updated - - -

    I also send the bridge Voltage and current in the following graph (Readings of the ossiloscope XC3)

    This is a phase shift converter. Each leg has a fixed duty cycle. The power flow is determined by the phase angle between each leg. In this particular case it is in the middle (the phases are 45 degree apart). This means %50 of maximum power flow is flowing through the bridge.

    Click image for larger version. 

Name:	Converter-bridge-current.png 
Views:	6 
Size:	7.3 KB 
ID:	151784


    General waveforms of a phase shift converter. (in literature)

    Click image for larger version. 

Name:	Phase-shifted-FB-ZVS-converter-basic-waveform.png 
Views:	3 
Size:	14.8 KB 
ID:	151785



    •   AltAdvertisement

        
       

  4. #4
    Super Moderator
    Points: 256,096, Level: 100
    Awards:
    1st Helpful Member

    Join Date
    Jan 2008
    Location
    Bochum, Germany
    Posts
    44,654
    Helped
    13587 / 13587
    Points
    256,096
    Level
    100

    Re: Full bridge converter, how to improve crest factor.

    The power factor can be partly improved by using a larger filter inductor. But you get at best rectangular current waveforms with 120° current flow. The best solution is a PFC rectifier respectively active front end. Some topologies have been discussed in this recent thread https://www.edaboard.com/showthread....-or-Boost-type


    1 members found this post helpful.

  5. #5
    Full Member level 2
    Points: 1,896, Level: 10

    Join Date
    Feb 2014
    Location
    Russia
    Posts
    135
    Helped
    73 / 73
    Points
    1,896
    Level
    10

    Re: Full bridge converter, how to improve crest factor.

    here is a kind of passive correction circuit https://www.researchgate.net/publica...Rectifier_Load


    1 members found this post helpful.

    •   AltAdvertisement

        
       

  6. #6
    Member level 2
    Points: 258, Level: 3

    Join Date
    Mar 2019
    Posts
    52
    Helped
    0 / 0
    Points
    258
    Level
    3

    Re: Full bridge converter, how to improve crest factor.

    Quote Originally Posted by FvM View Post
    The power factor can be partly improved by using a larger filter inductor. But you get at best rectangular current waveforms with 120° current flow. The best solution is a PFC rectifier respectively active front end. Some topologies have been discussed in this recent thread https://www.edaboard.com/showthread....-or-Boost-type
    Thank you for your input. I added a big inductor 50mH@10A and 100mH @10A and the result is just as you told.

    The cost of 50mH@10A is slightly lower than the active PFC. But as you said the PFC should be much better.

    But the passive inductor has also a big advantage. It is robust.

    Click image for larger version. 

Name:	Converter-bridge-BigInductor-current (2).png 
Views:	8 
Size:	10.2 KB 
ID:	151807

    - - - Updated - - -

    Quote Originally Posted by Velkarn View Post
    here is a kind of passive correction circuit https://www.researchgate.net/publica...Rectifier_Load
    I come to a conclusion that this kind of passive filter is good up to 500W. But for 5000W it may be ineffective.

    Thanks for your input.



--[[ ]]--