1
\$\begingroup\$

I plan to build a small timer device that, when triggered, lights up 3 leds. After 30 min, the first goes dark, after another 30 min the second and yet another 30 min later the third.

It's pretty easy to accomplish building three separate monostable NE555 stages that share one trigger button. Yet, only one at a time prevents its being retriggered during its high-level pulse at it's output. Question is: is there an easy way to keep all of the 555s from being triggered while any of them is still activated?

I thought of putting a diode between their outputs and the other's trigger input and putting a resistor between the button and ground, so that an active ouput would pull the other trigger inputs high, stronger than the button to ground would pull them down, and due to the diode prevent them from being pulled down if all outputs showed low-level...

Could something like that work? Something like a only-pull-up-not-down mechanism... Any suggestions appreciated. Thanks!

\$\endgroup\$
  • 1
    \$\begingroup\$ YOu can figure out a way but the Cap leakage time constant for 30 minutes has a large tolerance on the negative side and pushing plastic cap leakage rates with parasitic surface humidity and dust . THe old CD4060 runs at high f and then async binary counters can be decoded easily for ring count type decoding. but takes glue logic to latch off each stage. But very accurate and stable. \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Jul 29 at 0:56
  • 1
    \$\begingroup\$ Using two D-Flip-Flop (DFF) and simply counting in binary sounds more robust than having 3 555 timers. \$\endgroup\$ – Harry Svensson Jul 29 at 1:03
  • \$\begingroup\$ For the project I have in mind, something roughly between 25 and 35 minutes per step would still be perfectly fine. I just wondered if such a retriggering-blocker was possible. \$\endgroup\$ – fdub Jul 29 at 1:23
  • 2
    \$\begingroup\$ Feed the button from the output of the longest-running timer. Then the button won’t have the ability to retrigger while that timer is active. \$\endgroup\$ – Bob Jacobsen Jul 29 at 1:45
  • \$\begingroup\$ @Bob, thanks, that's the hint I was looking for! So simple it was :) \$\endgroup\$ – fdub Jul 29 at 1:50
2
\$\begingroup\$

Could something like that work? Something like a only-pull-up-not-down mechanism... Any suggestions appreciated.

It could work, but you'd need an impressive amount of hand-tweaking things, and very low-leakage capacitors, guarantee stable thermal environment … it's hard to get 555s to work reliably and precise at long periods. It's not what they're designed for

So, you really wouldn't want to build a long-term timer out of 555s; instead, you'd want to build something out of well-defined short periods, and count these reliable periods with a counter.

You'll notice that your own description of the problem reads like an algorithm, "do this, then wait, unless a condition is met, then do this or do that, depending on…".

That's really something that is easy to translate into software. So do exactly that which every EE since the mid-1980s would have done in your situation:

Get yourself any microcontroller, and write a tiny bit of program code. Microcontrollers have been designed specifically for this kind of problem.

Things like the Arduino platform make this easy; but for the same price, you could even have a Raspberry Pi Zero, which would allow you to program it in any language of your choosing (given there's a compiler / interpreter for Linux).
And: even a Raspberry Pi Zero (which in idle consumes 100 mW) would have less power consumption than three NE555 without load.(which doesn't say much positive about the Raspberry Pi, but a lot about the NE555).

\$\endgroup\$
  • \$\begingroup\$ I know I could do that easily but I wanted to build it in hardware to have it powered by two AAA batteries, powering up instantly and also just for the fun of it. So, I think I'll try my luck with a 74HC4060 running at 2.2 Hz, an AND gate and three flip-flops to switch the LEDs on Q12 high, Q13 high and Q12 and Q13 high. But thanks anyway for your explanations. \$\endgroup\$ – fdub Jul 29 at 12:32
  • \$\begingroup\$ Two AAA batteries deliver 3 V (or down to 2.4 V if discharged). There's enough microcontrollers that run at these voltages. \$\endgroup\$ – Marcus Müller Jul 29 at 21:46
  • \$\begingroup\$ Well, I built that thing using a hc4060 and three d-flipflops clocked by the counter's outputs. And from a timing perspective is was perfectly stable (well, it's 1:26:30 and not 1:30:00 but that's ok). Still, I admit it's a hassle and for the next project I surely will use a microcontroller again. Can you recommend a specific one running at 2,4 V? \$\endgroup\$ – fdub Aug 7 at 1:09
  • \$\begingroup\$ @fdub most modern microcontrollers can work at such voltages, what about ATSAMD, for example a ATSAMD10? While relatively pricey (compared to an NE555), it only needs a single external capacitor to work, is lower in consumption, pretty fast, and there's an easy upgrade path to other Microchip ARM microcontrollers, so you don't rewrite software for every project. Also, ARM means you get to program reliably in C with good compilers. Microchip means the ecosystem is proven. \$\endgroup\$ – Marcus Müller Aug 7 at 8:36
  • 1
    \$\begingroup\$ I just wanted to thank you for inspiring me. Recently, I bought a cheap STM32F103 board and it's amazing! \$\endgroup\$ – fdub 4 hours ago

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.