Help with fet - vgs

Discussion in 'General Electronics Chat' started by p_s_maniac, Jan 11, 2010.

  1. p_s_maniac

    Thread Starter New Member

    Jan 11, 2010
    5
    0
    hello everybody, i need some help on how to use fets
    i need to design a driver for a bipolar stepper motor using fets, i'm using two h bridges of n-channel and p-channel fets and a l297 drive. i use it with 5v for the l297 and 12v for the motor. until now everything works fine, but the circuit is intended to work with 48v . thats where my problem starts, i'm using irf530 n-channel fets and irf9530 p-channel fets and in the data sheet says that the max vgs(gate-source V) is of +/-20v and if it surpases that the fet is destroyed. then.... my question is, how can i use it for those conditions without destroying the fet and aplying most at least 46V in the coils of the motor?
    P.S.
    i've already tried with zeners, they dont work
     
  2. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Post a schematic of your circuit how it is now; we'll go from there.
     
  3. p_s_maniac

    Thread Starter New Member

    Jan 11, 2010
    5
    0
    ok, here it is, the zeners there is another h bridge like that one conected to the c and d pins of the l297. i use the bjt circuit to trigger the fets, and thats it. thanx for ur help
     
  4. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    OK, let's just start with a single P-channel MOSFET driver.

    R7 is 10k for Q3's base resistor
    Q3 is a BC548.
    you're using R6 as 10k resistor for a pull-up for the MOSFET gate.

    You'll certainly wind up with problems if you try to exceed 20v for your battery voltage.

    You're already having problems because of the high values of resistances used. 10k is ok for an emergency gate pull-up (I frequently use them as such in case the circuit fails), however it causes a very slow gate discharge time.

    See the attached; it's a simulation of just one improved driver.

    R1 limits the base current of Q1; a typical NPN switching transistor.
    R2 limits the current flow through D2, a 10v Zener diode, to about 21mA. D2 prevents the gate from getting more than 10v lower than V+.

    R3 tries to keep Q2 turned on when Q1 isn't sinking current. Q2 provides a quick discharge path for the gate of M1. D1 prevents R3 from trying to discharge the gate. R4 limits the maximum gate charge/discharge current to keep it from "ringing" due to the parasitic inductance of the wiring (unavoidable in the real world) and the capacitance of the MOSFET gate. R5 is the "safety net" - if something breaks in the rest of the driver, R5 will pull Vgs to zero.

    You also have some problems with your N-channel MOSFET driving; somehow you have those wires connected to the motor instead of the L297 driver IC. It's not going to work like that. Please post the lower side of your schematic so we can get the rest of the picture.

    Anyway, look at the attached schematic and simulation. If you have LTSpice installed, you can experiment with the simulation (also attached)

    If you don't have LTSpice, Google: LTSpice download
     
  5. p_s_maniac

    Thread Starter New Member

    Jan 11, 2010
    5
    0
    thanx for ur help, i really apreciate it
    ok, here is the second part, only another h bridge of the same caracteristics as the first one.
    im trying ur design......... i 'll inform about it a little later...........
     
  6. p_s_maniac

    Thread Starter New Member

    Jan 11, 2010
    5
    0
    mmmmmmm................
    i've alredy tried ur circuit. works fine except for one thing.
    by watching the voltage in the load, 21 of 24 V are in the load, which means 3V will fall on the fet. thas in the case of the p-channel fet. in earlier tests ive seen the same thing happen on the n-channel fet, so that meanstha in the final circuit (in the best case) 5v will fall on the fet and 19 in the load. which will not work as efectively as my boss wants.
    besides. the motor has a resistance of 5 ohms. 19V/5ohm=3.8Amps +/- .so that means there will be 5V*3.8A= 19 watts per line of fets and there are 4 lines, so 76 watts in total for all the transistors to dissipatte. that would cause an innecessary waste of energy.....
     
  7. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    OK, you need to post what you've done. It works in my simulation. If it does not work in yours, there is something wrong, but I need to see what's wrong.

    You're trying to use the high-side MOSFETs to control the lower side MOSFETs. This is defeating the purpose of the L297 controller. The controller is supposed to supply PWM to the low side of the bridge, but you are keeping the low-side MOSFETs turned on by the high side. This will not work very well.
     
  8. p_s_maniac

    Thread Starter New Member

    Jan 11, 2010
    5
    0
    ok, sorry for the lateness,thanx for your help but our boss decided to use an l6203 instead. works fine but itsa little expensive
     
Loading...