A couple of questions about a 555 in astable mode and a blinking LED

Question

I would like to ask you guys to comment a bit. Does it work in real life? I would like to use a 555 in astable mode with the same time intervals for on and off (both 1s) and let a Vishay diode flash. I know it is not 100% possible to have the same time intervals because \$R_a\$ would have to go to \$0Ω\$ which would not let the timer work. As most of you already know, I use PSpice for Ti2020 and therefore do not have all the components imaginable. I had to create a "LED" myself. But should fit like that.

My knowledge comes from this website. It is a German text, but the formulas are the same. ✌️

I arbitrarily chose 6V as the input voltage. The series resistor in front of the LED is \$165Ω\$, so that no more than 20mA can flow through – as specified in the data sheet.

\$(3.947V-0.74V)/0.020A = 160Ω\$ And I chose 165Ω for safety reasons.

\$C_3\$ and \$C_4\$ are backup capacitors. According to the website, one should use them.

Can you eliminate the small spikes during the switch-on and off processes? (see images)

I know there is no \$143919.5Ω\$ resistor, this is just playing with numbers.

current through LED

555 Out Voltage

---

Based on Spehro Pefhany's answer, I have now done the following: I simulated a real LED (forward voltage starting with 3.2V and a maximum of 3.9V). The LED gets its 20mA. I was able to reduce the base resistance a bit, but so that the timer does not lose more than 1mA.

Answer 1

The spikes in the simulation are not important. There may be some small parasitic capacitances etc. that are modeled

Your output circuit is a bit strange. The LED has a Vf of 4V at 20mA according to the datasheet, so it's not much like the diode you have used in simulation with a Vf of 750mV. If you want a remotely accurate simulation you would need to have a reasonably accurate LED model. There is no need of a base resistor for the BJT if you are using it as an emitter follower. You won't get anything like 20mA through the LED as shown. The generic NPN model you are using has a fairly low hFE (100) compared to a modern discrete transistor such as a 2N4401 or S8050.

It might be better to drive an NPN transistor to saturation (now you do need the base resistor) and put the LED between +6 and collector with an appropriate series resistor. If Vf is 3V minimum @20mA, then that resistor would be about 150 ohms.

The TLC555 can sink enough current to drive the LED directly, but if you want it to drive 20mA with the 555 output high (sourcing) then you'll need a transistor.

Answer 2

I know it is not 100% possible to have the same time intervals

Yes, it is. The National Semiconductor datasheet for their LMC555 has a 50/50 astable circuit in the apps section. In this version, it is on page 16.

https://www.ti.com/lit/ds/symlink/lmc555.pdf

Answer 3

The spikes are an artifact of the simulation modeling parasitic capacitance and assuming perfect zero ohm, no-impedance wires and components. Real circuits have series resistance and inductance so these spikes are damped out. You could refine your sim to add these, or just ignore them.

Anyway, on to the main issue: 50% duty cycle. A few changes to your circuit will get you closer that goal, with fewer components and less power used.

• Disconnect DISCH from the feedback and delete the pull-up resistor Ra at TRIG and THRESH. We’re going to repurpose DISCH for the LED and use OUT for the feedback.

• Delete the transistor. Connect the LED cathode directly to DISCH. DISCH has more than enough drive on its own for the LED, but like the transistor we deleted, DISCH can only sink current so it needs to be connected on the low side.

• Use OUT as feedback: connect the single Rb resistor from OUT to TRIG and THRESH. We do this because OUT can drive both high and low, so the Ra pull-up isn’t necessary. It’s that Ra pull-up that complicates making a 50% duty cycle, so by eliminating it we (mostly) solve the 50% problem, save a bit of power and get rid of a component.

With these changes the feedback voltage at TRIG and THRESH will alternate between 1/3 and 2/3 VCC, with a time constant set by the single Rb resistor and the capacitor. This time constant sets the pulse width, and it will be close to equal for both high and low.

A tip: CMOS 555 replacements like the TLC555, LMC555, etc. will be better at this since its OUT pin swings rail to rail (they'll be lower power too.)

Here's a sim, modified as above for a 1s (0.5Hz) LED flasher (simulate it here):

Another tip: to adjust the duty cycle, add a pair of diodes and a trim pot to the feedback. Related: 4-wire fan controlled by potentiometer and 555 IC not behaving properly

Learn more about how the 555 works here: Astable 555 circuit always on, not oscillating