Design boost converter

  1. i need help please, now i try to design boost converter that robust to the variation of the input votlage. i'm using PSIM to simulate the design, but i'm kinda confused to make the variation of the input. What i should do? what circuit that i should create to make the input varies in PSIM??..
     
  2. jcsd
  3. berkeman

    Staff: Mentor

    First a cascade buck design, and now a boost design.... sounds like a fun class. Normally I'd move this to Homework Help, but since it's a relatively complicated project, I'll leave it here in EE for now.

    Handling variations in input voltage is called "line regulation", as opposed to "load regulation" which has to do with how steady your output voltage is with variations in output current demand. You test your line regulation by varying the input voltage -- doesn't PSIM let you do that? Try searching for "line regulation" in the Help menus and the manual of PSIM.
     
  4. lol...actually this is for my final project, for the cascade buck, i scrapped it, because it was hard to implemented and complicated for its controller....after month delayed, now i continue do my project, and i need your guidance berkeman..

    currently i design boost converter with specs :
    Vin = 24 V
    L = 2mH
    C = 470uF
    Vout = 48 V
    Iout = 4 A
    IL = 8 A
    Fswitching = 5000Hz
    Vref = 48 V
    i use current mode control,so an error signal is generated by comparing output voltage Vo with reference voltage Vref. Then this error signal is used to generate control signal I. The inductor current is then sensed and compared with control signal I to generate the duty cycle of the switch and drive the switch of the converter.
    and i have create block diagram for controller
    [​IMG]
    a. Diagram block for the whole controller
    b. Diagram block for voltage controller(assumed that current controller had worked well)
    c. Diagram block for current controller(assumed that voltage controller had worked well)
    Gv = PI(proportional Integral) for voltage controller
    Gc = PI for current controller
    Gc* = Vout/Vin
    Vs=Vin
    Vo* =Vref
    what i wanna ask are?:
    1. was my diagram block correct?
    2. How can i determined the PI constant for both current controller and voltage controller?
    thanks very much..i am looking forward for your help...
     
    Last edited: Jun 8, 2009
  5. berkeman

    Staff: Mentor

    Some of those specs look wrong. If output is 48V at 4W, then Iout is not several amps. And what's IL?

    5kHz is too slow for a practical switcher. Should be more like 50kHz at least, especially with a 2mH inductor.

    Where is the triangle wave oscillator in your block diagram? I may be missing it...
     
  6. oopss..sory, i mean P = 192 W since Ioutput = 4 A, IL is inductor current,in boost
    IL = (Vout/Vin)*Iout ;IL = 8 A
    what i want to control in this boost converter are the current input(IL) and the Voltage output(Vo)..
    actually i kinda confuse how to put triangle wave in my block diagram, but this is how i simulate the circuit in PSIM.
    [​IMG]
     
  7. berkeman

    Staff: Mentor

    Looks like you have the tri wave generator in the lower left corner of the circuit. You will want to include a max duty cycle limiter in the circuit in some way (Quiz Question -- Why?). Also, why are you including inductor current in the control equation? It could be included for a cycle-by-cycle current limit, but that doesn't appear to be its use here (I could be missing it tho').
     
  8. yes,this is current mode control(or i thought so),so an error signal is generated by comparing output voltage Vo with reference voltage Vref. Then this error signal is used to generate control signal I. The inductor current is then sensed and compared with control signal I then compared with triangular wave generator to generate the duty cycle of the switch and drive the switch of the converter...
    hmm....is it wrong if i make the controller like that??Is it possible?cause i taught that it will be much easier, all i have to do is find the PI constant by looking the step response of the system,bode plot, effect of disturbance,etc..and also the current controller must work faster than voltage controller...
    nb:sorry my English is not good because i'm still learning :D
     
  9. berkeman

    Staff: Mentor

    Don't worry, your English is fine.

    I have mainly worked with voltage mode feedback. I've only used current sensing at the switch element to use for cycle-by-cycle current limiting, not as part of the main control loop. So sorry, but I won't be able to be much help on that part. Maybe others here have done what you are trying...

    You didn't answer my Quiz Question yet on max duty cycle...
     
Know someone interested in this topic? Share a link to this question via email, Google+, Twitter, or Facebook

Have something to add?