5
\$\begingroup\$

Looking through some 555 square-wave-generator circuits (possibly with variable duty cycle for 4-wire PWM fan control) I've found three different circuits:

  • From the Wikipedia 555 article, a "standard 555 astable circuit":

astable 555 circuit from Wikipedia

  • From DPRG, a "simple PWM circuit":

PWM 555 circuit from DPRG

PWM 555 circuit from Nidec

Functionally, these circuits all seem to do the same thing, but seem to be vastly different in design, particularly the last two (the Wikipedia circuit lacks variable duty cycle, though the article briefly mentions it); short of simulating them, is there a way to determine which is "better" or are they equally valid?

\$\endgroup\$
1
  • \$\begingroup\$ How are you planning to adjust the PWM output? Both of these 555 PWM circuits rely on using a potentiometer to vary the duty cycle. Will that work for you? \$\endgroup\$ Mar 24 '12 at 19:25
3
\$\begingroup\$

The first scheme is the astable configuration as descibed by the datasheet: it's a monostable multivibrator which triggers itself.

  • When the output is low, C is charged through R1 and R2 until it reach the threshold voltage (2/3 Vcc);
  • then OUT is turned high, DIS is turned on and the capacitor starts discharging through R2 to DIS, until it drops below the triggering voltage (1/3 Vcc); then OUT goes low and it starts again.

Note that the duty cycle is limited to 1/2.


The other two schemes have the advantage that the duty cycle can go from almost 0 to almost 100%, making them more suitable for PWM purposes.

The second schematic is very ugly, but basically you change the duty cycle varying the ratio of the resistances that charge and discharge the capacitor C1, turning the potentiometer R1.

For some reason, the charge and discharge of C1 is driven by the output Q, while the real output is through the DIS pin. The bad thing about this is that the output through the DIS pin is open collector, so it will be strongly driven low by the Dis pin but pulled up only by the 10k resistor.

Note that if you turn it fully to one of the rails, you are basically charging/discharging the capacitor through a diode and the output stage of the 555. This is not exactly a good idea.


The third one is based on the same concept, but a resistor is put to the OUTput to avoid the aforementioned problem, and the Vout pin is used properly.

If you need a PWM control, I would suggest looking at the datasheet, where you can find a good PWM modulator:

enter image description here

\$\endgroup\$
4
  • \$\begingroup\$ The PWM reference design requires a failing edge trigger signal - what's the simplest way to generate that? \$\endgroup\$
    – Andrew
    Mar 23 '12 at 13:54
  • \$\begingroup\$ @Andrew Right, you need a clock for that; probably the pulse positioin modulator could work for you... \$\endgroup\$
    – clabacchio
    Mar 23 '12 at 15:29
  • \$\begingroup\$ @Andrew: You could use a 556 (dual-555) chip, and use one of the timers in the traditional astable configuration to clock the other in the PWM configuration. \$\endgroup\$ Mar 23 '12 at 19:54
  • 1
    \$\begingroup\$ @KaelinColclasure Yeah but I think it's getting too complicated for driving a simple fan; probably the third scheme he showed is easier to set up. \$\endgroup\$
    – clabacchio
    Mar 23 '12 at 20:30
0
\$\begingroup\$

I believe the best solution (which I am using and grabbed from https://www.electronicshub.org/pwm-led-dimmer-using-ne555/) is to use the first circuit (in the OP) but use diodes on the potentiometer in order to allow the duty cycle to vary from near 0% to near 100%.

enter image description here

The load shown is LEDs (with current-limiting resistors), IOW, a dimmer for the LEDs.

I am using a Darlington BJT device (http://www.mouser.com/ds/2/149/KSD560-63636.pdf) because I'm going to want to control quite large loads (several amps) so the much higher DC current gain of the Darlington configuration is good.

Looking at the datasheet for the LM555, note that the high-level output is guaranteed to only be 3v or so (for Vdd=5v), which probably precludes using a power MOSFET (most of which don't guarantee Vgs-threshold to be below 3v). But the DC current gain of the Darlington is so high that it almost looks like a MOSFET, except drops a little more voltage (from collector to emitter, versus from drain to source).

I'm using this as a dimmer also. A big issue I ran across is that the LED bulbs I'm using have a capacitor in them. If the PWM frequency is too high, that waveform is filtered out and you can't dim very low. But if you lower the PWM frequency too much (by increasing the size of the capacitor on pin 6 or the resistance of the potentiometer) then you'll get flickering; so it's a compromise.

\$\endgroup\$
1
  • \$\begingroup\$ Does the range of selectable duty cycle depend on the relationship between 1k and 50k pot? If so exactly what is the formula for optimizing or at least determining range given selected components. \$\endgroup\$ Jul 1 '20 at 23:19

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.