555 PWM Question

Discussion in 'General Electronics Chat' started by DC_Kid, Apr 28, 2014.

  1. DC_Kid

    Thread Starter Distinguished Member

    Feb 25, 2008
    638
    9
  2. ericgibbs

    AAC Fanatic!

    Jan 29, 2010
    2,503
    380
    This configuration of an Astable uses the Output as the charging/discharging source voltage, so the Output is very close to 50:50 high/low

    EDIT.
    Which is required if you are using the 555 for a PWM App.
    The pot will vary the mark/space ratio, but the total period will remain constant
     
  3. Brownout

    Well-Known Member

    Jan 10, 2012
    2,375
    998
    No. Do not connect the red line. The circuit works without it. Connecting the wire would make the circuit fail.
     
  4. DC_Kid

    Thread Starter Distinguished Member

    Feb 25, 2008
    638
    9
    ah, i totally missed it, i didnt see that line from R2 to pin3 was not tied to Vcc (no trace dot, etc). i was scratching my head on a few things there, but now it's all good. my R7 and pin3 are wrong in my pic, etc. always good to check and dbl check the schematic, etc......

    thanks.
     
    Last edited: Apr 28, 2014
  5. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
    I was the one that wrote that article. There are two totally different schematics that use two different methods of operation, as the others have pointed out you have merged them incorrectly.

    The first schematic is based on the 555 Hysteretic Oscillator.

    The diodes steer the current, but the resistor ratio always stays the same, so the frequency remains constant.
     
  6. ian field

    Distinguished Member

    Oct 27, 2012
    4,415
    784
    The discharge and output pins are exactly in phase - you can use 1 less resistor in the timing CR network and get the fastest possible astable configuration by driving the CR timing by (both source and sink) the output pin.
     
  7. DC_Kid

    Thread Starter Distinguished Member

    Feb 25, 2008
    638
    9
    yes yes..... i got it..... thanks for the replies......

    can you ascertain what the duty-cycle adjustment range is on this one, i suspect its that 50-99% side.
    [​IMG]
     
  8. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    5,801
    1,105
    Simulation shows the duty cycle range is ~1%-30%
     
  9. DC_Kid

    Thread Starter Distinguished Member

    Feb 25, 2008
    638
    9
    thats crappy for me, i need a 10-90% range.
     
  10. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
    Then use the schematic I posted as is.
     
  11. DC_Kid

    Thread Starter Distinguished Member

    Feb 25, 2008
    638
    9
    yeah, i am leaning that way, but i also need adjustable freq, and i started with the basic 555 1-99% duty cycle and then added a 8 position rotary to switch in different caps to change the freq, which might be all i need, etc. i might also add in a switch to add some extra uF to each selection so i can have two ranges of freq, etc. i would rather not add more IC's if i dont have to, etc.

    thanks all for your feedback.
     
  12. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
    Lets back up, what are your requirements?

    This circuit (also in that chapter)...

    [​IMG]

    Provides both excellent freq stability and PWM stability. You can also use the spare LM393 gate to make a hysteric oscillator, to eliminate the 555. If I knew what you were after I could draw something up.
     
  13. DC_Kid

    Thread Starter Distinguished Member

    Feb 25, 2008
    638
    9
    sure, i saw it, it needs an additional 393 IC, etc. i need adjustable freq in the 1-200Hz range, with duty-cycle in the 10-90% range. the 555 output is nice because it can sink and source thus allowing me to drive a pair of opto's in flip-flop fashion (one on while the other off, a pair of led's if you will).

    however, the footprint of the 393 may be smaller than having 8 caps for adjusting freq, and, 393 may be more reliable than low cost caps too. so maybe for sake of "easy" i just use 555+393. using a pot instead of a rotary switch will also be cheaper. how much current can the 393 sink and source?

    the on/off times of the output need to be fast enough to not impact 200Hz, etc.

    your thoughts?
     
    Last edited: Apr 28, 2014
  14. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
    This family of comparators (LM393 dual, LM339 quad) is open collector. They can sink up to 16ma, but they need a pull up resistor.

    The big advantage of the above schematic is the two controls, PWM % and Freq, are truly independent, otherwise you could use a 556 (two 555's in one package) and make an oscillator followed by a monostable. It would need an oscope to set up, with simplicity in control you have to have some complexity of circuit.

    When I get a chance I'll sketch up a LM393 version, it was something Sgt. Wookie showed me a while back.
     
  15. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
    OK, I reread post #13, this is not even close for what you want. I was trying out Linux KolourPaint too, so I did not take the time to prettyfie it.

    [​IMG]
     
  16. DC_Kid

    Thread Starter Distinguished Member

    Feb 25, 2008
    638
    9
    Bill,
    the 555/393 combo will be fine.
    with the 393 i'll need an opto driver, the TLP222A-2 opto will be in the 7.5-10ma range, and i cant use the 393 like the 555 output. a small IC with one n and one p fet will do.

    whats the duty-cycle range on the 555/393 combo?
     
    Last edited: Apr 29, 2014
  17. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
    Completely adjustable, you vary R4 and R6 to create the range you want for the pot R5. It can be a full 0-100% if you want. Here is another variation in Chapter 10...

    [​IMG]

    U2a pin5 is a voltage input. 1/3 to 2/3 Vcc is 0% to 100%.

    You could also use a 556 (two 555's) to give yourself the drive you want from U2a from the second 555 in the package..
     
  18. DC_Kid

    Thread Starter Distinguished Member

    Feb 25, 2008
    638
    9
    any reason to not ditch the 1N pair for fet pair like this

    the LED's are the opto's, etc. (ignore pin #'s, etc)
    R3 may need to be bigger than 10, maybe like 500 ohm? even a small charge at 5v sitting on the gate would dump 500mA across R3 (10 ohm) through the 393, albeit for short period, just not sure the 393 can handle that. and maybe R1 down to 5k to help drive the fet gates a tad faster.....
    the fairchild LM393AN lists output current as "typ 18mA", but oddly no max value......
    [​IMG]
     
    Last edited: Apr 29, 2014
  19. ian field

    Distinguished Member

    Oct 27, 2012
    4,415
    784
    String the pot across the tops of the 2 diodes and connect pin 7 to the wiper, you need the usual maximum frequency limiting resistor from pin7/wiper to Vcc.
     
  20. DC_Kid

    Thread Starter Distinguished Member

    Feb 25, 2008
    638
    9
    i am moving to the 555/393 configuration as this will allow me more functionality.
     
Loading...