400v dc-dc converter

Discussion in 'General Electronics Chat' started by youfa77, Jul 15, 2014.

  1. youfa77

    Thread Starter New Member

    Nov 20, 2011
    5
    0
    Dear all,

    I need some help on my project. Appreciate I can get some feedbacks from you all.

    My project is to design and fabricate a 500w power rectifier system which convert a 230vac supply into 400vdc source using dc-dc converter half-bridge topology.

    The 230vac source will be rectified through a full-wave bridge rectifier circuit which I will then obtain approximately 325vdc. After which then it will feed into the dc-dc converter circuit by mean of MOSFET switching and go through a step-up transformer and through a fast-recovery diode (schottky diode) and to the output.

    I will also need to write a code for the microcontroller for the switching of the MOSFET with a duty cycle of 0.35 with switching freq. of 20MHz.

    The transformation ratio will be abt 1 : 3.6 which will be indented. And how can will determine the number of turns and flux ( i think related to Bmax) for both primary and secondary winding, and also selection of core.

    Please advise the components or guidelines which are needed for this project, and what are the others things which i need to look out for.

    Thanks in advance!
     
  2. shortbus

    AAC Fanatic!

    Sep 30, 2009
    4,015
    1,531
    You seem to be going at this backward. :) A transformer works with AC not DC. Convert your 230 VAC with the transformer to the needed higher voltage, then rectify it to 400 VDC.
     
    youfa77 likes this.
  3. youfa77

    Thread Starter New Member

    Nov 20, 2011
    5
    0
    A pulse tranformer will be used for this dc-dc converter.
     
  4. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,542
    1,251
    Does the output of your power system need to be isolated from the AC line? If not, then a simple boost converter will turn 325 VDC into 400 Vdc with much less effort. Also, why do you think the turns ratio for your transformet is 1:3.6?

    ak
     
  5. crutschow

    Expert

    Mar 14, 2008
    13,019
    3,235
    Trying to switch a power MOFSET at 20MHz will lead to high losses. I think you want to keep it below 1MHz for reasonable efficiency. Or did you mean 20KHz?
     
    youfa77 likes this.
  6. youfa77

    Thread Starter New Member

    Nov 20, 2011
    5
    0
    @ crutschow: Thanks for the reminder. Yupz; Typo Error on my side, Should be around 20kHz

    @ Analogkid: My lecturer prefer half-bridge topology Isolated DC-DC converter; using 2 MOSFETs. Tat's why the pulse transformer is there. Duty cycle (D) set at 0.35.

    Based on the equation: Vout = Vin * D * (Ns/Np)
    with Vout = 400v, Vin = 325v, D=0.35, (Ns/Np) should be around 3.6
     
  7. youfa77

    Thread Starter New Member

    Nov 20, 2011
    5
    0
    Any one have idea on the calculation on the mosfet gate resistor for mosfet?
     
  8. tcmtech

    Well-Known Member

    Nov 4, 2013
    2,037
    1,666
    There are two easy ways to go about this depending on whether or not you need it to be isolated from the mainline power source or not.

    1: If no isolation is needed a simple buck boost transformer could be added to the system to boost the AC voltage up to around 288 VAC by adding 48 volts to the 240 with a simple 24/48VAC buck boost transformer which being it would set up in a autotransformer configuration would mean it could be rather small like a 50 VA size unit.

    2: If isolation is needed the same process could be done using a dual primary 240/480 VAC 24/48 VAC buck boost transformer but you would need to go with a larger unit like a 750 - 1000 VA size and just use one of the primaries as the power input and then use the second primary plus the secondaries all in series to make the 288VAC output.

    Either way once the 288 VAC is rectified and the diode drops plus other small losses are factored in you would get almost exactly 400 VDC in the end.
     
    youfa77 likes this.
  9. youfa77

    Thread Starter New Member

    Nov 20, 2011
    5
    0
    Will like to check with you all,

    When I am using 160vdc power supply to inject into the Board 3 of the circuit, 15vdc power supply to run the MOSFET gate driver circuit, and Desktop USB to run the microcontroller, the circuit is running and I am able to obtain 300vdc at the output.

    However, when I hook up the whole circuit, and apply a 230vac into Board 2 of the circuit, 15vdc power supply to run the MOSFET gate driver circuit, and Desktop USB to run the microcontroller, the main fuse blown, MCB & RCCB /ELCB tripped, Lo-Side MOSFET, Hi-Side MOSFET gate driver and microntroller is damaged. After which then i rectified the damaged part and do some modification and re-test. The circuit is ready.

    After which then, when I hook up the whole circuit, and apply a 230vac into Board 2 of the circuit, using a 3 outputs dc supply pack (+5v / +15v / -15v) from the same power supply to run the MOSFET gate driver circuit at 15vdc, and microcontroller at 5vdc. The circuit cannot function. Some measurement show that there is a 325vdc is measured from L to N; 160vdc from N to E. 160vdc is also present at both MOSFET Gate 1 and Gate 2 when measure from N.

    Some of the component in Board 1 are faulty which include the Cboot, IR2104. Testing of microcontroller is also done but the output waveform is either at flat 5vdc or output which is at +/-2.5Vdc which is abnormal, as it should read from 0 to 5vdc.

    Is there any case or floating neutral or unbalance neutral which caused this? All the DC negative are not earthed.

    My circuit is as shown as attached.

    Please help, thanks!
     
Loading...