Generating a frequency without changing duty cycle

Discussion in 'The Projects Forum' started by Zotto, Mar 3, 2016.

  1. Zotto

    Thread Starter New Member

    Mar 3, 2016
    26
    2
    Hey I am trying to design a frequency generator ranging from .25hz to 25hz approximately, while having a constant duty cycle that is as small as possible, preferably ~5%. The design I am currently using includes a 555 timer, but I am not sure if it can produce what I am looking for. Would really appreciate your advice. Thanks.

    Here is the circuit I am using: http://www.instructables.com/files/orig/FJJ/V35I/GBF5KOPV/FJJV35IGBF5KOPV.jpg
     
  2. GopherT

    AAC Fanatic!

    Nov 23, 2012
    5,971
    3,663
    Do you want the duty cycle to remain constant while frequency is changed AND as small (low) as possible?

    Or,
    Do you want the duty cycle to remain constant while frequency is changed AND the circuit to be as small as possible?
     
  3. crutschow

    Expert

    Mar 14, 2008
    12,954
    3,214
    If you just want the duty-cycle to remain low (short pulse) that is fairly easy.
    If you want the duty-cycle to stay at a constant value, that's more difficult but doable with a counter.
     
  4. GopherT

    AAC Fanatic!

    Nov 23, 2012
    5,971
    3,663
    @Zotto,

    Also, when you mention duty cycle, do you mean the ON time (e.g. 10 mSec) or do you mean the percent of time it is on vs. time of a full on/off cycle 40%).
     
  5. Zotto

    Thread Starter New Member

    Mar 3, 2016
    26
    2
    Thanks for the replies, I need the duty cycle to be low at all times, simplicity would be nice. For instance it doesn't need to be a precise 5% duty cycle all time I just need the pulse to be short, and gopherT i mean high 5% of the 100% cycle, so short pulses.

    To clarify my design, I need constant short pulses, where the frequency is variable using a potentiometer, but using the pot has no effect on the pulse width, i.e. changing frequency while maintaining a constant duty cycle.
     
  6. dannyf

    Well-Known Member

    Sep 13, 2015
    1,759
    354
    An RC oscillator with a discharge can have very short duty cycle - it is a typical rc oscillator + a diode.
     
  7. bertus

    Administrator

    Apr 5, 2008
    15,633
    2,341
    Hello,

    You have a contradiction in your requirements.
    You say you need short pulses that remain constant while the frequency changes.
    With a constant pulse width and changing frequency, the duty cycle will change.
    With constant duty cycle and changing frequency, the pulse width will change.

    Bertus
     
  8. atferrari

    AAC Fanatic!

    Jan 6, 2004
    2,644
    759
    If you take time to answer the question above you could then say exactly what you would like to have.

    See if you could get a precise definition of what duty cycle is and go from there.
     
  9. Zotto

    Thread Starter New Member

    Mar 3, 2016
    26
    2
    I am sorry, I am misusing the term duty cycle. Thanks bertus. I need constant pulse width with a changing frequency.
     
  10. bertus

    Administrator

    Apr 5, 2008
    15,633
    2,341
    Hello,

    Take a 556.
    Make an astable oscillator with the first half and a monostable with the second half.

    Bertus
     
    wayneh likes this.
  11. HitEmTrue

    Member

    Jan 25, 2016
    32
    9
    Can this be done with one 555, with a small capacitor (and parallel bleed-off resistor?) at the output to generate the pulse?
     
  12. GopherT

    AAC Fanatic!

    Nov 23, 2012
    5,971
    3,663
    Things are gettingn confused and inconsistant with the answers. Again, when you say "constant" do you mean

    (a) the easy way for a 555, meaning a constant 0.05 second pulse and change the interval of that 0.05 second pulse? (constant time, variable frequency)
    (@bertus' answer will work here)

    -or, do yu mean,

    (b) the difficult way with a 555, meaning constant percent (example, 5%, when the frequency is slow) and stay the same percentage when the frequency is fast. (constant duty percentage, variable frequency)
     
  13. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    5,756
    1,098
    Please specify how many mS, and the tolerance in mS, e.g. 10mS ± 1mS
     
  14. dannyf

    Well-Known Member

    Sep 13, 2015
    1,759
    354
    It can be done with a 555, a resistor, a capacitor, a diode and an optional diode.
     
  15. Bernard

    AAC Fanatic!

    Aug 7, 2008
    4,167
    394
    You could try connecting 555-pin 2 to junction of R1-R2. C1=2 uF, R1= 1k, R2= 2 M pot,, R3 20k.
    Pulse about 2 msec. Clock period about 4 sec to 40 msec.
     
  16. Zotto

    Thread Starter New Member

    Mar 3, 2016
    26
    2
    @GopherT
    The first one you are saying is what I think I need. I dont have a 556 timer, can I implement it using a single 555, and a 10k pot? Sorry for the constraints but I dont have access to other parts at the moment.
     
  17. Zotto

    Thread Starter New Member

    Mar 3, 2016
    26
    2
    Can i use this design with a 10k pot?
     
  18. dannyf

    Well-Known Member

    Sep 13, 2015
    1,759
    354
    It can be done with just 1 555 timer - think of it as an inverter (a NOT gate).

    Why use two 555s if one 555 does the job?
     
  19. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,512
    1,243
    I agree with danny. If Rb is fixed and Ra is variable, then the discharge output, Rb, and C set one pulse width almost completely independent of Ra. (Ra does contribute to the current through the discharge output, changing Vce, changing the discharge period *slightly*.) The other pulse width includes Ra and is variable. A 100:1 span in 1 range will have very poor adjustability, and will be logarithmically compressed at one end of the range, but a single CMOS 555 will meet the requirements in post #1.

    ak
     
  20. Zotto

    Thread Starter New Member

    Mar 3, 2016
    26
    2
    I need decent adjustability with my pot, can I achieve this?
     
Loading...