Two switch forward convertor

Discussion in 'The Projects Forum' started by D8666, Apr 27, 2014.

  1. D8666

    Thread Starter New Member

    Nov 15, 2013
    16
    0
    I'm trying to build a two switch forward convertor as shown below, My input voltage as tested was a 30v Lab power supply ( 2A) I'm implementing PWM using the output compare function of a DSPic30f4011 this is working perfectly, wit a duty ratio of 0.45 I was getting 5V across my load.

    My intended output is to charge a 12v battery so roughly 13.7V.

    [​IMG]

    The upper Fet got quite hot, my Fet drivers are Microchip TC1410. After reading on here I have learned about the difference between low and high side drivers. Could anyone suggest a suitable one?
    Edit: I understand the reason the Fet is getting hot is that it is not turning on fully.
    I'm more of a programmer!
     
  2. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    5,801
    1,105
    Why don't you use FETs in the sim? You will get a better indication of circuit operation/problems.
     
  3. D8666

    Thread Starter New Member

    Nov 15, 2013
    16
    0
    Thanks for the reply..I didn't know you could do that! Can I get the actual Exact ones I'm using or will the standard N channel Mos model be ok? I'm new to LTSpice aswell. The FETs I'm using are IRF3710 N Channel 100v 59A

    Edit: Had a look there and that exact model isn't there I guess theres a way of adding it to the library.
     
    Last edited: Apr 27, 2014
  4. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    5,801
    1,105
    The default Fet model gives unrealistic results. Right-click on it and you get a drop-down box where you can select among various types. Choose one with similar characteristics to your intended actual Fet.
     
  5. D8666

    Thread Starter New Member

    Nov 15, 2013
    16
    0
    Thanks for your replies.
    I found a similar one with the same voltage rating and Rds(on)
    My schematic is here. I really cant get an accurate reading for my transformer until I get an LC meter.
    Any suggestions for the highside driver?
    [​IMG]
     
  6. D8666

    Thread Starter New Member

    Nov 15, 2013
    16
    0
  7. crutschow

    Expert

    Mar 14, 2008
    13,022
    3,236
    I'm not clairvoyant so can't help without your schematic showing exactly how you connected everything.
     
  8. D8666

    Thread Starter New Member

    Nov 15, 2013
    16
    0
    Sorry! I hope this is ok.
    [​IMG]
     
  9. D8666

    Thread Starter New Member

    Nov 15, 2013
    16
    0
    I redrew the circuit in tinycad, I hope its easier to understand.
    [​IMG]
     
  10. crutschow

    Expert

    Mar 14, 2008
    13,022
    3,236
    So are you saying the outputs of the IR2210 goes to zero when you connect it to the MOSFETs? There should be at least a small signal.
     
  11. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    What's the purpose of the dual FETs in the primary?

    To get forward current both FETs need to be on (wastes two Rds drops), and to handle primary recirculating current you need at least one FET on, but then it still passes the recirc current through a single diode (so you don;t get the benefit of synchronous rectification).

    It looks to me like you would be better off just replacing one of those primary FETs with a short circuit.

    Or, wiring it properly so one FET is the forward FET and one is the recirc FET to give sync-rect.
     
  12. crutschow

    Expert

    Mar 14, 2008
    13,022
    3,236
    I didn't consider the circuit function previously but I agree with RB, the circuit makes no sense. Is that supposed to be a flyback converter?

    It might make sense as a forward converter if you added a capacitor in series with the transformer primary to allow resetting of the primary inductance by the bottom transistor.
     
  13. ronv

    AAC Fanatic!

    Nov 12, 2008
    3,291
    1,255
    I think the circuit may be okay. Kind of looks like this one. It may be a problem with the big boost cap and the charge path. Seems everyone uses this guys circuit. Remove the 1k from gate to source on the top fet and the electrolytic from the boost cap circuit and see if it will switch.

    http://www.intersil.com/content/dam/Intersil/documents/hip2/hip2100.pdf
     
  14. crutschow

    Expert

    Mar 14, 2008
    13,022
    3,236
    The crucial difference is the capacitor in series with the transformer primary, similar to what I suggested.

    Why remove the electrolytic?
     
  15. ronv

    AAC Fanatic!

    Nov 12, 2008
    3,291
    1,255
    I was thinking about Fig. 1.
    The electrolytic is not needed and with the charge path of his circuit I'm not sure you want it.
     
  16. D8666

    Thread Starter New Member

    Nov 15, 2013
    16
    0
    Thanks for the replies . I have ordered this highside driver I understand the circuit better and it has an internal bootstrap diode
    http://www.farnell.com/datasheets/1700270.pdf
    The Ir 2110 that I was using was acquired from a friend so on the off chance it was damaged before I got it I'll stop asking questions about it..I tried your sugestion there about removing the 1k resistor but it didn't work.
     
  17. crutschow

    Expert

    Mar 14, 2008
    13,022
    3,236
    Okay. I understand now how the 2-transistor forward converter works. When the two transistors are on, energy is transferred from the primary to the secondary of the transformer. The core reset occurs through the two primary diodes when the two transistors are off, returning any inductive primary energy back to the supply.

    But I see no problem with the electrolytic boost cap. It's only used to provide charge to the top transistor gate.
     
    Last edited: Apr 28, 2014
    D8666 and THE_RB like this.
  18. ronv

    AAC Fanatic!

    Nov 12, 2008
    3,291
    1,255
    His circuit may be a little hard to bootstrap since it has to charge thru the inductor instead of like normal where it gets charged directly thru the bottom FET. It may take some playing to get it going depending on the inductance of the transformer. If it was charging in the normal way the 100n would be large enough at 20Khz so there is no need to add another part to the system.
     
    D8666 likes this.
  19. D8666

    Thread Starter New Member

    Nov 15, 2013
    16
    0
    How does this look? I will be not Switching Sw3 or Sw4 on the secondary side so the body diodes on those Fets shall be used for rectifying..Is this ok?
    Can anyone spot any other problems?

    Here's a link to the L6385E data sheet.

    EDIT: Vb on the secondary side shall be replaced by a resistor.

    Thanks for your help. I must say TinyCad is great I only started using it two days ago. Its a lot better than Inkscape!

    [​IMG]
     
  20. ronv

    AAC Fanatic!

    Nov 12, 2008
    3,291
    1,255
    I can't find any info on the L6564.
    Body diode usually aren't to good. (high drops and low speed) You might be better of with good diodes.
     
    D8666 likes this.
Loading...