I saw this a while ago:
https://www.instructables.com/Designing-a-555-Timer-on-Discr...
https://hackaday.com/2011/08/05/building-a-computer-out-of-5...
exactly :)
BF901 FTW.
Is that what you had in mind?
So I went to the office hours of a random EE professor thinking they would help me out. Instead I got scolded about how 555 timers are not real engineering and that I shouldn't waste his time.
I never used a 555 timer ever since.
It's excellent teaching material for that reason alone, you do learn about it if you try to understand what makes it tick (there are plenty of articles about it, including blown up versions). I agree it is not the best solution for most applications but I'm happy to admit that I've actually used it in production designs (more than once, actually) where it made good sense to have a component that didn't have to be programmed. If you have a soft component on a board and a spare io line then you are usually better off doing it in some different way.
I've seen some interesting applications of 555s that would have taken a lot more hardware otherwise, one of which was an oven controller with used a thermistor to directly affect the PWM output of a 555.
I was meaning to add footstep-activated lights to my stairs using vibration sensors and 555 timers, but then I learned that if I tried to operate them from a battery, the 555 would drain it in hours, while a much more sophisticated ESP32-c3, would last a month in sleep mode on two coin batteries or one li-ion in the same form factor all while being part of a development board that greatly increases idle current.
Some people get into Science, Software, and Electronics for the wrong reasons.. And end up miserable teaching after failing in the private sector.
A few 555 can teach people a lot, and burning out parts is part of the learning process. Most fold the DIP legs under like a "dead bug", as that is the tradition to prevent its accidental re-use.
In terms of component cost, ATTINY or PIC10 mcu have internal RC oscillators with configurable PWM pin hardware. Thus a single component is usually better than the accumulated precision error in discreet parts around a 555.
I usually recommend an RC car/truck build, https://eater.net/6502 , and or an LDOmotors Voron kit. Getting your Ham Radio technician license will also introduce you to an intuitive understanding of EE component model limits.
This covers a lot about discreet analog circuits, and I recommend trying to figure out how every circuit works on your own:
https://archive.org/details/encyclopediaofelectroniccircuits...
Simulators are not perfect, but they are a lot cheaper when starting out. =3
Tutorials:
https://www.youtube.com/@FesZElectronics/videos
Tools:
https://www.analog.com/en/resources/design-tools-and-calcula...
https://web.archive.org/web/20200218212700/http://spectrum-s...
This requires a GPU on Windows, but is a more advanced newer Spice simulator:
https://www.qorvo.com/design-hub/design-tools/interactive/qs...
Obligatory link to Forest Mims' book: https://archive.org/details/555-designs
Then take a look on https://yusynth.net at some of the VCF designs, and build one of them.
You won't have a synthesizer, you'll have some crazy homebrew drone machine that you can make scary movie sountracks with.
So, yeah, 555 timers are cool and doing things with analog ICs is groovy but there's a reason everyone just stuffs a small microcontroller in places where we used to just stuff a 555, and it's maintainability.
1. Low-noise applications. I’d naively expect the 555 to be less noisy than a clocked digital microcontroller, though it’s been awhile since I’ve worked in this space.
2. Low power applications. How does latent power draw compare between a 555 and a typical low power microcontroller?
The 555 is very power hungry compared to a typical cheap low-power microcontroller. IIRC there are lower power variants but the 555 still fundamentally does timing by draining current through a resistor, which is going to result in losses.
TTL ones were exceptionally noisy because the output transistors "shot through" - both output transistors would conduct for a moment shorting the supply rail to ground and crowbarring ridiculous interference onto other parts of the circuit.
Then in production, you need another stage to flash the FW, which add time and complexity.
Then security, cheap MCU usually has bad software protection, that means your software can be read out easily, not a big deal since the FW replacing the 555 would be dead simple anyway, but try to explain it to a non-technical CEO when he read about it on his morning's newspaper.
