Controlling H Bridge without uc?

Discussion in 'The Projects Forum' started by Gibz73, Apr 10, 2015.

  1. Gibz73

    Thread Starter New Member

    Apr 10, 2015
    7
    0
    Hi folks, first post/question;

    I have had a search aroud here and t'internet in general and have not been able to find an answer.

    Situation:

    I've built a simple MOSFET H-Bridge to drive a motor forwards and reverse. I want the motor to run forwards for x time then reverse for x time. It needs to be an automated setup, but, without Arduino or any other uc's.

    I've tried setting up a transistor astable multivibrator to flip between each set of gates on the H bridge, which works...but I also want the input to the H dridge to be a PWM signal. I've sent PWM signal to that transistor astable, I've experimented with hooking the H Bridge up to V+ which was a PWM signal all sorts...but nothing seems satisfactory. There must be an easy efficient way to alternate between the two sides of an H Bridge (running motor CW then CCW) whilst also giving the H Bridge PWM(so I can vary the speed)...all without Arduino or other uc's.

    I've simulated a 3 x 555's setup in Circuit Wizard, 1st 555 is a slow astable at the frequency I want, that alternatively sinks and sources power to the other two 555's which in turn power the H Bridge. Results were flaky and I'm not sure I shoud breadboard it.

    I have this feeling that I'm staring the answer in the face but cant see the wood for the trees and would truly appreciate some input here...I need to crack on with this project.

    I hope I've managed to make myself clear as to what I need, please feel free to ask me anything that would help you to help me. Thanking you all kindly in advance.

    :)
     
  2. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,516
    1,248
    To paraphrase Rear Admiral Joshua Painter - Son, an engineer can't take a dump without a schematic.

    ak
     
  3. Gibz73

    Thread Starter New Member

    Apr 10, 2015
    7
    0
    Well that's the thing isn't it...I dont have one. I only have a schematic of the H Bridge I'm using...everything else was discarded because it A) didnt work or B) looked flaky in the simulation :/

    png is the H Bridge. Obviously the two 47 Ohm resistors are where I want the automated PWM signals in (left side, right side, left side right side rinse and repeat)

    Edit: When I say automated I mean I want it to flip back and forth to run motor CW then CCW...and that signal needs to PWM so I can vary the current to the motor in each direction. Make sense?
     
  4. MaxHeadRoom

    Expert

    Jul 18, 2013
    10,492
    2,363
  5. Papabravo

    Expert

    Feb 24, 2006
    10,135
    1,786
    You can design a finite state machine to implement the behavior that you describe. It is simple to describe, but difficult to implement and debug. Are you up for it?
    Google Terms:
    "Finite State Machine"
    "Moore Machine"
     
  6. MikeML

    AAC Fanatic!

    Oct 2, 2009
    5,450
    1,066
    H-bridges that tie the gate of the P-Fet to the gate of the N-Fet invariably generate huge shoot-through currents that usually lead to destruction of the FETs or noise in the power supply that glitches other parts of the circuit.

    The right way to do this is to make a sequencer that has four outputs that drive the gates: GN1, GN2, GP1, and GP2. The gate signals have to be sequenced with slight delays between them such that the bottom N always turns off before the upper P turns on and vice versa. It is easy to do this with a microcontroller, but harder to do it without using one...

    ps. I am saying the same thing as PapaBravo...
     
    Papabravo likes this.
  7. Papabravo

    Expert

    Feb 24, 2006
    10,135
    1,786
    The TS/OP did not put a limit on the complexity of the non processor solution although he might have meant to do so. In practice it could be anything from 3 MSI chips, to an FPGA, up to and including a processor. Your point about shoot through is an excellent one. That is reason enough for taking a disciplined approach to the solution.
     
  8. Gibz73

    Thread Starter New Member

    Apr 10, 2015
    7
    0
    Hi folks and thanks for the input. I dont mind a bit of complexity...I can follow schematics pretty well anyway :) tbf the H Bridge design was billed as being "not too efficient but good enough for your first H bridge"...I welcome links to better/well known H Bridges and any resources that may be of use. *goes to look up state machines*
     
  9. MikeML

    AAC Fanatic!

    Oct 2, 2009
    5,450
    1,066
    Little H-Bridges that operate from a supply voltage of 3V to 5V can get by with tying the gates together; but when you get to a supply voltage of ~8V or higher, then the shoot-through is severe because the Vth of the P plus the Vth of the N is less than the supply voltage, meaning both P and N are on at the same time. The only solution is to delay the P turn on after the N turns off and vice versa.
     
  10. Papabravo

    Expert

    Feb 24, 2006
    10,135
    1,786
    Controlling the gates of the H-bridge independently is a piece of cake for a finite state machine.
     
  11. frpr666

    Active Member

    Feb 2, 2010
    37
    6
    we live in the 21st. century and a $1 chip can replace many glue logic chips. It might be fun to build some stuff with MCUs. OT: I wonder what even the kids are doing with the Raspberry Pi (https://www.raspberrypi.org/magpi/)
     
  12. GopherT

    AAC Fanatic!

    Nov 23, 2012
    5,975
    3,671
  13. Gibz73

    Thread Starter New Member

    Apr 10, 2015
    7
    0
    Hi guys and thanks for the input. Really I'd love to do this with discrete components...for the experience as much as anything. I can power the two sides of the H Bridge with high out from a 4017 but that seems silly to me. Anyway I will revise my post...can someone please suggest a schematic for a good H Bridge to controll motors 6-15v (ish) and handle about 2 amps(peak)?

    Thank you very much.
     
  14. Papabravo

    Expert

    Feb 24, 2006
    10,135
    1,786
    How would that work, and why does that seem silly to you?
     
  15. Gibz73

    Thread Starter New Member

    Apr 10, 2015
    7
    0
    Well, it would be a long winded effort and perhaps a waste of a 4017 and (the way I see it working) 3 x 555's. But then again it wouldnt be a waste if it did the job eh? I'm thinking, first 555 sets the slow oscillation to give the 4017 a slow clock input(I've tested out R1 = 1k Ohm, (V)R2 = 50K Ohm, C1 = 330uf and that gives me the timing I need. The 4017 uses the first 2 output pins to temp power the other two 555's (altenatively obviously)...the two alternating 555's deliver a PWM signal to the 2 sides of the H Bridge.

    How does that sound?
     
  16. Papabravo

    Expert

    Feb 24, 2006
    10,135
    1,786
    Believe me you do not want to power a 555 from the outputs of a 4017. That is not the way things are supposed to work. You have the 2nd and 3rd 555's running all the time and use the outputs to select one or the other.
     
  17. Gibz73

    Thread Starter New Member

    Apr 10, 2015
    7
    0

    That's why I'm here :)

    I dont suppose....yu could perhaps....pop up a schematic for me? I need someone to throw me a bone here tbh :/
     
  18. blueroomelectronics

    AAC Fanatic!

    Jul 22, 2007
    1,758
    98
    Unless it's a homework assignment why not use an Arduino?

    PS what is this device for?
     
  19. Papabravo

    Expert

    Feb 24, 2006
    10,135
    1,786
    In a career that spanned a half a century I never found a use for the 555. Other people swear by them, but to me they're greek.
     
  20. blueroomelectronics

    AAC Fanatic!

    Jul 22, 2007
    1,758
    98
    Aside from a toy organ I built back in the 1970's I've not used them either. The good is they support up to 15VDC, the bad is they're just not that stable. I enjoyed the LM3909 more but they're pretty scarce now.
     
Loading...