NE555CN PWM Circuit

Discussion in 'General Electronics Chat' started by tracecom, Dec 1, 2011.

  1. tracecom

    Thread Starter AAC Fanatic!

    Apr 16, 2010
    3,869
    1,393
    I found a circuit on the web that I wanted to build using the parts I had on hand. The changes I made were (1) used 1N4148 diodes instead of 1N5818, (2) used a 10k pot for VR1 instead of a 100k and a 1uF cap for C1 instead of a .1uF, and (3) am driving a red LED instead of the gate of a IRFZ46N Mosfet.

    [​IMG]

    I I breadboarded the circuit and it works; I can adjust the brightness of the LED using VR1. My questions are as follows.

    1. The LED goes very dim, but not completely off. Is that normal? (ETA: When I inserted the 510 Ω current limiting resistor, the LED then would go almost completely dark, so I guess my question is answered.)
    2. I think perhaps the 10k resistor was only a pull-up for the gate of the MOSFET. Does it serve any purpose with the LED or should I delete it? (The circuit seems to work without it.)
    3. I added a 510 Ω current limiting resistor in series with the LED and that seems to limit the current to 20 mA. However, does my DMM measure the current correctly when it is PWM'ed? (I don't seem to be able to measure the voltage drop across the LED.)
    Thanks.
     
    Last edited: Dec 2, 2011
  2. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    It is a very simple circuit that does not go from 0% to 100%, maybe at 5% you can still see the LED. Maybe if the pot is 100k the LED will turn off.

    The 10k resistor helps turn off the LED. Try 1k ohms.

    You are lucky that the battery is probably limiting the current or the LED would have blown up and maybe also the 555. Of course an LED needs a current-limiting resistor.
     
  3. tracecom

    Thread Starter AAC Fanatic!

    Apr 16, 2010
    3,869
    1,393
    Thanks for your response. I did some additional testing and modified my original post. It seems that my only remaining question is if it is possible to measure the voltage drop across an LED that is being PWM'ed.
     
  4. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    An oscilloscope can measure the forward voltage of a pulsed LED. A multimeter averages the voltage from its peak to the times it is off.
     
  5. tracecom

    Thread Starter AAC Fanatic!

    Apr 16, 2010
    3,869
    1,393
    Thanks again. I really need an oscilloscope.

    What about the current? Are my measurements ok using a DMM in series with the LED?
     
  6. wmodavis

    Well-Known Member

    Oct 23, 2010
    737
    150
    Nope it is not for same reason it wouldn't measure voltage correctly.
     
  7. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    A DMM set to measure current has a series resistance that reduces the current being measured. Also, since the LED current is pulsing on and off then the current measurement is an average of the on current time plus the off current time.
     
  8. tracecom

    Thread Starter AAC Fanatic!

    Apr 16, 2010
    3,869
    1,393
    Below is the modified circuit that I used for experimentation.

    [​IMG]

    Although I had no way to measure the frequency of the PWM from the circuit, I calculated its frequency at 144 Hz. In order to see the effect of a change in C1, I removed the 1 uF and installed a 1000 uF. The resulting frequency of the PWM was about .1 Hz (1 cycle every 10 seconds.) Then it occured to me that I could accurately measure the voltage drop across LED1 and the current sinked by pin 7 of the 555. The Vf measured 1.88 V and the current measured 19.47 mA; these are essentially identical to the readings that I got when I had the 1 uF capacitor as C1 and the PWM duty cycle at maximum. Thus, it seems that PWM (at least at a frequency of about 144 Hz) does not significantly distort the measured readings.
     
    Last edited: Dec 2, 2011
  9. bertus

    Administrator

    Apr 5, 2008
    15,648
    2,347
    Hello,

    Why do you use the discharge output of the 555?
    That connection has not so strong transistor inside.
    The max current is about 10 to 15 mA (as I recall what SgtWookie once said).

    Better use pin 3 as output.

    Bertus
     
  10. tracecom

    Thread Starter AAC Fanatic!

    Apr 16, 2010
    3,869
    1,393
    It's not my design, but here is what the explanation is from the web site.

    http://www.dprg.org/tutorials/2005-11a/index.html

    "In this circuit, the output pin is used to charge and discharge C1, rather than the discharge pin. This is done because the output pin has a "totem pole" configuration. It can source and sink current, while the discharge pin only sinks current. Note that the output and discharge pins go HIGH and LOW at the same time in the oscillator cycle."

    For driving the gate on a mosfet, pin 7 is probably adequate, but I may be overstressing it with an LED.
     
  11. bertus

    Administrator

    Apr 5, 2008
    15,648
    2,347
    Hello,

    Take a look at page 4 of the attached datasheet.
    There is the following stated:

    [​IMG]

    Take a special care of note 5.

    Bertus
     
    Wendy likes this.
  12. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
    I have a similar project that hasn't been published yet.

    555 PWM Oscillator

    Bertus, how am I to take Note 5? There is no max current? Or what?
     
  13. bertus

    Administrator

    Apr 5, 2008
    15,648
    2,347
    Hello,

    They state 15 mA at 15 Volts in the table.
    The powerdissipation of the COMPLETE package must be taken care off.

    Bertus
     
  14. nerdegutta

    Moderator

    Dec 15, 2009
    2,515
    785
    I built this last night, had to change the 100uF to a 0.1uF for the motor to run smooth. With the 100uF the motor ran kind of choppy.
     
  15. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Pin 7 does not actually have a maximum rated current, but as current increases, Vsat of the internal transistor increases, which increases power dissipation in the timer. If the power dissipation limit is exceeded, the timer will burn up.

    Somewhere I had found a reference to Vsat increasing to over 400mV if the current sunk by pin 7 exceeded 15mA. Staying below 10mA will lessen the error in the formula used to calculate R1/R2.

    My main objection for sinking a lot of current via pin 7 is when the traditional astable configuration of R1/R2/C1 is used with pin 7 connected between R1 and R2; if R1 is too small, power consumption will unnecessarily skyrocket.

    These facts are not obvious to people who are not intimately familiar with 555 timer circuits.

    So, I came up with the 10mA limit for R1; keeping R1 at a minimum of 100 Ohms per volt of Vcc will limit R1's maximum current.

    Up to a certain point, I prefer to increase the value of the timing resistors rather than the size of the timing cap. If the resistors become too large (over a couple of megs) then the timing cap leakage and threshold/trigger currents will cause timing errors, even preventing the timing cap from charging. On the other hand, large values of capacitance mean physically large capacitors.

    Very generally, the larger you go in cap sizes, the greater the tolerance is, so you're better off using a smaller cap (within reason).

    [eta]
    Gee, didn't see Bertus' post of the datasheet excerpt before writing this, but the Vdis(sat) numbers basically confirm what I wrote above.

    With some circuits, if Vdis(sat) exceeds about 500mV, their operation can be affected.
     
    Last edited: Dec 2, 2011
  16. tracecom

    Thread Starter AAC Fanatic!

    Apr 16, 2010
    3,869
    1,393
    I had actually read the note Bertus referenced (although it's note 4 on my version of the ST datasheet dated November 2008.)

    However, I confess to being very ignorant as to the meaning of what much of the "data" on datasheets means. One thing that continually confuses me is the interpretation of the information in the "Parameter" column as ratings data. I thought that "parameters" were essentially the conditions under which the data in the "Min., Typ., and Max." columns applied. On more than one occasion, I have been advised that the contents of the "Parameter" column were to be considered as "limits" that must not be exceeded.

    For example, in the portion of the table that Bertus posted, in the Vdis(sat) row, under parameters, the highest Vcc listed is +15V (with a corresponding Idis of 15mA). I thought that was just one "parameter" (of two shown) under which the "Discharge pin saturation voltage" was rated. Because the maximum Vcc of the NE555 is clearly stated as 16V, I assumed that the creator of the table could have included a "parameter" of "Vcc = +16V, Idis = ??mA" had he/she chosen to, and included applicable ratings for that "parameter" in the "Min., Typ., and Max." columns. I didn't take the "Vcc = +15v, Idis = 15mA" parameter as a maximum, nor did I take the "Vcc = +5v, Idis = 4.5mA" as a minimum.

    Am I wrong?
     
    Last edited: Dec 2, 2011
  17. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    It always helps to know the manufacturer name and date of the datasheet, as items can vary.

    That is correct.

    That is not necessarily correct; it depends upon the context of the item. For example, for an N-ch power MOSFET, if Vgs(th) (the threshold voltage)
    is given as 1.5v, then in order for the MOSFET to be considered turned OFF, you must get Vgs below the threshold voltage.

    For the absolute maximums, refer to that section of the datasheet. You generally don't want to get close to these maximums. For example, for a transistor Ic rating, use 1/2 of the maximum specification; while you might go up to 80% of the voltage rating. It's subjective.

    It is. While you will usually see the "typical" values with average devices, you must plan on the "max" values, as otherwise your circuit will not be guaranteed to work with 100% of the devices you try.
    Those are just parameters given for the results shown. Some 555 timers have a Vcc limit of 18v, some are 16v, and I think a couple are 15v. You usually would not want to operate an IC right up against its' maximum rated voltage, as if you exceed it, it will likely be damaged by excessive power dissipation - so, it is more meaningful to show min/typ/max results with parameters closer to were you would be actually operating the IC.

    Just because the absolute maximum is 16v, does not mean you should operate it at 16v. If you go by the 80% of voltage rating suggestion, that leaves Vcc at 16 * 0.8 = 12.8v.
     
  18. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
    I have seen folks try to say a large capacitor on pin 7 can exceed the max current, which is basically false. To the best of my knowledge there is no capacitive limitation for this pin, the capacitor will discharge really fast, and pin 7 can handle it.

    Where you get into trouble is in situations like this circuit, where pin 7 becomes a output. Then current matters, a lot.

    Pin 3, the true output, is hard rated for 200ma max., either sinking or sourcing.
     
  19. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    Datasheets also vary, therefore I always look at the datasheet of the inventor of the IC and hope it has good English.

    ST Micro did not invent the 555 and they are Italian, not English speaking.
    Their datasheets are bits and pieces of very poor quality copies (the worst copies I have ever seen) of real datasheets and sometimes the translation from Italian to English does not make sense.
    Signetics invented the 555 then were soon bought by Philips.
    The Philips datasheet is very detailed and their English is excellent.
     
  20. tracecom

    Thread Starter AAC Fanatic!

    Apr 16, 2010
    3,869
    1,393
    Regarding the discharge pin, here are two interesting quotes from AN170, the Phillips Application Notes for the 555; both are from page 6.

    "The discharge transistor, Q14, is current-limited at 35mA to 55mA internally."

    "With extremely large capacitor values, a maximum duty cycle which allows some cooling time for the discharge transistor may be necessary."
     
Loading...