Before microcontrollers, blinking an LED took one 8-pin chip, two resistors, and a capacitor, and the circuit is still the best first lesson in how time constants work. Today we build the NE555 astable in de:volt, predict its frequency on paper, and then let the oscilloscope grade our arithmetic.

The build

Open a guest session at sim.devoltapp.com and place:

• Breadboard (400) and Bench PSU at 5 V
• NE555 Timer, straddling the centre gap
• Resistor RA = 10 kΩ from pin 7 (discharge) up to 5 V
• Resistor RB = 100 kΩ from pin 7 down to pin 6 (threshold), with pin 6 tied to pin 2 (trigger)
• Electrolytic Capacitor C = 10 µF from pin 2 to ground
• A 10 nF ceramic on pin 5 to ground for decency, a Red LED plus 330 Ω from pin 3 (output) to ground, and supply wires to pins 8 and 1

The palette search finds everything by name, and pin numbers appear when you hover the chip. Press Run and the LED settles into a slow, confident blink.

Predict, then measure

The astable's datasheet formula:

f = 1.44 / ((RA + 2·RB) · C)

With our values: 1.44 / ((10k + 200k) · 10 µF) = 1.44 / 2.1 = 0.686 Hz, with a high-side duty of (RA + RB) / (RA + 2·RB) = 52.4%.

Now make the scope check it. Open the Analysis panel; the interesting channels are the capacitor node and the output node. Set TIME to 5 s so several cycles fit in the window, open the display-settings menu, and switch on Measurements. On our bench the strip read:

FREQ 0.684 Hz against a predicted 0.686 Hz: agreement to a third of a percent, and the difference is real physics rather than sloppiness, since the formula's 1.44 is itself an approximation. VPP on the capacitor node reads 1.66 V, which is exactly the ⅓ Vcc to ⅔ Vcc excursion the 555's comparators enforce: the entire mechanism of the oscillator, visible as one number.

Watch the shape too. The capacitor trace is a shark-fin: an RC charge curve through RA + RB, then a faster discharge through RB alone. That asymmetry is why the plain astable cannot reach 50% duty, and why our high side runs longer than our low side.

Two honest instrument notes

If FREQ reads Unavailable on a shorter time window, that is the measurement engine being truthful: with fewer than a couple of full cycles in view, it declines to guess. Give it a longer window. And note the sample-rate line above the plot: acquisition happens inside the solver at up to 10 kSa/s, and the instrument reports the effective rate it achieved rather than pretending.

Make it yours

Everything is live while the simulation runs. Drop C to 1 µF and the blink becomes a 6.9 Hz flutter. Swap the LED for the 8 Ohm Speaker part, push the frequency into the hundreds of hertz, and the astable becomes a tone generator (wire a real loop; since July the audio path needs both legs connected, like hardware). Push RB down toward 1 kΩ and watch the duty cycle climb, then check your reasoning against the formula.

The monostable cousin of this circuit, a button-triggered timer, is coming later this week. The 555's full pin behaviour, including the control-voltage input we decoupled and ignored, lives at docs.devoltapp.com/parts/timer-555.