# Pulse a 555 astable oscillator every second without an additional IC

I am developing a ranging device to help people with blindness. I need to pulse a 38 kHz Infrared LED every 1 second. A 555 timer in astable mode generates the 38 kHz modulation. I'm currently using two ICs, one to blink every second and another that constantly generates the 38 kHz modulation. However, I feel it may be possible to do it with a single 555 IC. Because the reset pin of the 555 turns off the oscillation whenever it is grounded, I thought it might be possible to use a capacitor and resistor combination to pulse the reset pin every 1 second. Perhaps there is a different combination of doing this without using an additional IC?

The current configuration is a basic 555 astable oscillator as below. I'd prefer to add simple components to the reset pin that would allow me to control the duty cycle of when the 38 kHz signal is triggered without using an additional IC.

R1 = 4.7k R2 = 22k C1 = 680 pF

• However, I feel it may be possible to do it with a single 555 IC. I don't and I think it is impossible without extra active components. But feel free to prove me wrong and show a design that can work. You could use a NE556 which contains 2 timers in one package. However, 555 based timers are not very accurate, what you design as 38 kHz will in practice be between 36 and 40 kHz due to component tolerances. If you need good accuracy then use a microcontroller. Then getting an accurate 38 kHz and an accurate 1 second will be easy (assuming a cystal is used as a clock for the micro.). Dec 13, 2021 at 9:14
• Also realize that making C = 680 pF is below what an NE555 can work with properly. With C = 680 pF the frequency will be very inaccurate (much lower) even if you used 0.1 % accurate components only. I would not make C much less than 4.7 nF for a reliable working 555 timer. Dec 13, 2021 at 9:18
• Also: usually the IR receiver have a quite narrow bandwidth (for light immunity), are you sure that a 555 would be stable enough in all operating conditions (mostly temperature)? Dec 13, 2021 at 11:06
• Thus far it has been very stable, I've not tested it at sub zero temperatures, but usually it is within 1khz. Dec 13, 2021 at 11:27
• 1 kHz of 38 kHz is roughly 3% error. That's pretty terrible. Also, your schematic as is is already very complicated. Do you really want to add more analog circuitry to make it more complex and less exact? Because your whole problem would be solved, at lower total cost, by a 40ct microcontroller and a capacitor for its power supply (two components, in total!). If you want much better frequency quality, you can't rely on 555s or microcontroller-internal RC oscillators, and would have to spend a couple cents on a quartz in addition to that. Dec 13, 2021 at 15:34

You say that you "feel it may be possible to do it with a single 555 IC", so I'll try to provide some intuition for why this might not be the case.

Consider the waveform you want to generate. You have three states: (1) off, in the inactive state, (2) off while pulsing at 38kHz, and (3) on, while pulsing at 38kHz. One might think that states (1) and (2) are the same since the voltage of the waveform is the same, but the two states have very different conditions for when to change to the next state.

Since the circuit then has three distinct states, it will need more than 1 bit of memory. The 555 timer only has a single flip-flop in it, so I feel pretty confident that it would be impossible to implement the circuit you want without introducing another bit of memory in some way.

Other answers and commetns have already suggested mulitiple ways of introducing this extra bit of information. Using a 556 timer (or two 555s), using a microcontroller, or implementing a flip-flop with discreet components.

You could use a 556 which is a dual 555 timer IC. Some are sold for barely 10ct each when you buy at least 1000 of them.

I do not see a simple solution with a single 555 timer instance using only a few resistors, capacitors or diodes.

I could think of some methods not using another IC, just for the "fun" of it, but not really practical:

• Add a spinning disc at $$\\frac{1}{n}\$$ rotations per second, with $$\n\$$ slots to let the light pass only for as long as wanted each second. Possibly use a metal disc and use a contacter reset the 555 when it is not needed.
• Heat a bistable metal plate while the 555 is on which resets the 555 when it became too hot and removes reset when it cools down. This would be very dependant on the local temperature;
• Use a resettable polyfuse - it trips when it gets too hot, so you do not need the heating current to go through it, only heat it enough. So you might be able to hook it up to the reset pin and heat it using a charging resistor. I am not sure that you can heat the polyfuse enough nor that you could find the desired operating periods and pulse widths, and have stable periods over time.

Anyway, I could only think of these wild alternatives, which is a good indicator for me that it's unlikely there is another "electrical" solution. Using a 556 is technically a solution to the question as it's only one IC. Using a microcontroller would be too, but it exceeds the cost of the 556 (and you still need to program it).

• The device is portable and is powered from a battery. But the antique 555 eats batteries, I recommend using a Cmos 555 (LMC555, TLC555 or ICM7555). Dec 13, 2021 at 14:20
• It's true that I didn't verify current consumption when checking for pricing which almost tripples for a CMOS device. To stay with a 556, the TS556CN costs just under $0.30 starting from 100 pieces, the only issue being availability in the current market. At that cost, considering a cheap low power microcontroller might be the better solution: some kind of signature could be introduced in the emitter pattern so that the receiver signals permit determination of the emitter and also the exclusion of any unknown emitter. Dec 13, 2021 at 15:26 • There's a few easily available microcontrollers that are about$0.3. To that needs to be added the cost of programming the component and the NRE (compiler, development) but there's greater flexibility (adjust frequency & period in the code, signal modulation). Dec 13, 2021 at 15:44

Technically, using the dual timer (TLC556 as Audioguru stated) is still just one IC, would only need 1 more resistor and 1 more capacitor to provide the gating pulse. If starting over, that would be my choice. Or a PIC, as Ie_top stated. But if you're committed to using a 555, then yes, there are a few more ways to generate the gating pulse without using another "IC".

The well-known 2 transistor free-running flip-flop (astable multivibrator) first comes to mind. It needs more parts than the other approaches but it's not using an IC :-)

I've found enabling a 555 via pin 4 requires a firm pull-up. Internally, it is the base of a PNP BJT. So I'd first try using PNP BJT or e-mode PMOS transistors to actively-pull pin 4 'high'. Like this:

simulate this circuit – Schematic created using CircuitLab

The exact resistor values may need to change, as they're affected by properties of your particular transistors. For PNP BJTs, resistors R5 and R6 would not be needed.

You didn't specify how narrow or wide the gating pulse should be. If it can be narrow, then this is a simpler "no IC" pulse generator:

simulate this circuit – Schematic created using CircuitLab

Pulse1 is mostly high (555 is On), with a narrow Off time. The added parts in the dotted box invert Pulse1 to create Pulse2. Pulse2 is mostly low (555 is Off). You can tweak the parts somewhat to change the timings, but it's not as flexible as the other circuits. It's also harder find a 2N6027 or 2N6028 PUT device.