flyback boost converter

Discussion in 'General Electronics Chat' started by id_ruben, Mar 11, 2013.

  1. id_ruben

    Thread Starter New Member

    Oct 27, 2009
    7
    1
    My project DC2DC converters, with specifications
    Inputs from 6V DC to 50V DC.
    Output DC 220V.
    Power 1KW

    I chose the flyback boost converter.

    But I do not understand how to make a transformer :
    - The cable must be used.
    - How much for the primary and secondary windings.
    - Can be made from the flyback flyback CRT transformer
    - Is there a match between Bobin following:
    - Bobbin Case EE22 10 pin
    - Bobbin Case EE25 10 pin
    - Ferrite Core Bobbin EE22PL3K/SET
    - Ferrite Core Bobbin EE25R3K/SET
    - Bobbin EI28 10 pin (core case)
    - Ferite Bobbin Core PC44PQ35/35Z-1

    Thank
     
  2. JDT

    Well-Known Member

    Feb 12, 2009
    658
    85
    Start by deriving some rough values:

    You want an output power of 1KW. Assuming an efficiency of say, 75% how much continuous input current is that with a 6V supply (or whatever the supply voltage is)?

    Work out what the what the on/off ratio of the switching transistor will be. It will be roughly Vout/Vin.

    The peak inductor current will be at least twice the continuous input current x the on/off ratio.

    Depending on the switching frequency you want to use, the inductor has to store a certain amount of energy every cycle. Faster frequency = less energy each cycle. Example: 1KW is 1000 joules per second. So if your inductor switched at 1KHz then it will have to store 1J in its magnetic field. 10Khz = 0.1J etc. Allow for losses.

    Now you know approximately how much energy the inductor needs to store in what time. You can work out the inductance.

    And you approximately know the peak and average inductor current. You can work out a thickness of wire.

    You need to specify a magnetic core that will not saturate.

    I haven't done the calculations but I'm guessing the inductor will be BIG. 5 inch cube? Bigger? Laminated iron core or ferrite core depends on frequency.

    Now think about the transistor to switch this current...
     
    Last edited: Mar 11, 2013
    bug13 likes this.
  3. JMac3108

    Active Member

    Aug 16, 2010
    349
    66
    Couple comments fo JDT...

    Flyback duty cycle is not Vout/Vin. Its (Vout x n) / [(Vout x n) + Vin], where n is the transformer turns ratio.

    And there is no inductor in a flyback. Unless you are referring to the inductance of the primary winding of the transformer. In fact, the transformer in a flyback is not really a transformer in the classical sense, its a set of coupled inductors. Energy is stored in the inductance of the primary when the switch is on, then it "flys back" to the secondary when the switch is turned off. So its technically correct to refer to the primary winding as an inductor, but it will probably confuse the OP. Better to refer to it directly as the inductance of the primary winding.

    Also, you will likely use a ferrite core for a flyback transformer, at typical SMPS frequencies of 50kHz up to 400KHz.


    For the OP ...
    Flyaback transformer design is not trivial and you probably won't get simple answers here that let you quickly cookbook it. You're going to have to do some reading. I suggest a google search on flyback transformer design and start there. Perhaps someone here has a good link to share.
     
    bug13 likes this.
  4. bountyhunter

    Well-Known Member

    Sep 7, 2009
    2,498
    507
    bug13 likes this.
  5. JDT

    Well-Known Member

    Feb 12, 2009
    658
    85
    Sorry, I was being confrontational. I think a push-pull converter, possibly with a H-bridge and a double wound transformer will be a much better choice at this power level. It will also lead to a much smaller transformer because you are not storing energy in the magnetics.

    Also, for a 1kW converter don't consider a supply voltage below 24V.
     
  6. takao21203

    Distinguished Member

    Apr 28, 2012
    3,577
    463
    6V, 1kW, 75% eff. = 6V, 1.25 kW.

    1250 / 6 = roughly 200 Amps.

    I'd like to see the actual cables and terminals for that before giving any advice, not to speak about the battery.
     
  7. takao21203

    Distinguished Member

    Apr 28, 2012
    3,577
    463
    Not really.
     
  8. bountyhunter

    Well-Known Member

    Sep 7, 2009
    2,498
    507
    We designed a 800W DC-DC converter that ran off a 12V battery to power a field missile launcher and it took some serious current out of that battery. Had to run 4 FETs in parallel on each leg of the primary side (push pull) to handle the current. We were running it off a large marine usage (deep discharge) 12V battery.
     
  9. takao21203

    Distinguished Member

    Apr 28, 2012
    3,577
    463
    Wow. I only have a 500VA toroid here currently and a 2kW welding transformer.
    I know the heat and the currents for about 300W or 350- wires starting to heat up and all this.

    I assume it only runs for a short time, not continously for hours.

    It is often really similar, people get obsessed with details instead to understand what is realistic and doable, and improve on that top-down.

    It might proove very difficult to start a design for 2kW without having experience from smaller SMPS circuits.

    2kW would be very hard for me, and a big effort as well. I'd solve it with banking for a single, temporary circuit.

    To use this for hours or even 24/7- big effort just for the cables.

    You need to know so many details from experience.

    Terminal screws not thightened properly, or cables not terminated properly- arc discharge + cables melting + fires = only a question of time.

    These a problems you'd see from only a few 100 Watts already.
     
Loading...