I am a musician and not an electronics genius but wanted to ask a question and see if anyone could help me. I am looking for the materials and plans to make a VERY simple LED blinking device. It needs to be very small, (probably powered by a watch battery or smaller) and just blink a tiny, red, LED at a specific tempo.
Basically this is similar to a metronome except that it is very small and there is no input for tempo, it just has an on/off switch. It is like a metronome that only blinks one tempo. I would want the ability to make different units blink different tempos but each one need not be adjustable.
Does anyone have a circuit design or idea that would help me? Thank you so much!!
...nor even slightly technically savvy by the looks of it.
Easy.
formatting link
It just got difficult. ...also see "technically savvy" (above):
formatting link
Not a big deal. I've seen these gizmos on checkout-counter displays (e.g. for candy).
Well, that's EXACTLY what I'd call it. The smallest solution is essentially a one-chip system (a microcontroller). The first project you do with one of these is *Blink a LED*:
Is this for commercial sale? What size, overall? Are you willing to consider stacking two 3V lithium button batteries? (One is doable, just curious.)
I'm thinking of 4 resistors, two caps (one for timing), three 1-cent transistors, and two 1n4148 diodes. Plus the 2mA LED. Constant 20uA draw with no heavy load pulses from the button, at 2 beats per second. Also, there was the LM3909's I think... not as efficient and they required heavy pulse capability from the battery. Also, of course, the micro option. But then the pulse current requirement comes in (which some buttons can do okay on.)
Any oscillator capable of running at a really low frequency will do it, so a 555, a unijunction transistor, a pair of transistors in a multivibrator, the list is endless.
The real issue is that how precise the tempo is. When there's a variable control, one can fiddle with the tempo, but when it's fixed, if the tempo changes due to components changing value with temperature or age, there's nothing there to adjust.
The obvious solution is to get start with a crystal oscillator, and then divide it down. Some math would be required, to balance an easily available crystal frequency with a divider that won't require too many distinct ICs. Often it's a tradeoff, you can get by with a ripple counter that has lots of dividers inside, but then you may have few choices for the crystal (and hence the crystal frequency may turn out require a custom grind). Or, pick a readily available frequency, and then the divider will have to follow, and that may not be in one convenient package.
Done properly, ie pick the right crystal frequency, and one could even create a tone standard as well as a metronome out of this thing. Just have a second suitable divider chain feeding a speaker.
I assume someone has suggested a microcontroller because it basically does the work of the divider chain, and they tend to be run off a crystal clock (since even if absolute clock speed is not required for some applications, it can often be easier to put a crystal in there rather than an RC network). It will keep the number of ICs to an absolute minimum compared to a crystal oscillator and divider chain.
But there again it's a balance. You may pull a divider chain out of some scrap piece of electronics, but have to go and buy a suitable microcontroller. And then you have to learn how to program it, and may need to collect software tools to assemble or compile the program, and get it into the microcontroller. Or pick a bulkier microcontroller because it's easier for single quantity work.
It can divide by pretty much anything, manage the LED's blink rate as well as duty cycle, and even allow for changing the rates through software, and drive a piezo buzzer directly. Plus, unlike a 555 design, it requires almost NO discretes. Oscillator, led, led's resistor, mcu, liIon battery - five parts. Add any pushbutton switches you might want, for controlling it. Some MCUs don't require the oscillator either, but then you get into temperature sensitivity. Crystal and two caps if you want really accurate timing.
Plus, companies are putting a LOT of effort into making MCUs that run on Li-Ion batteries for long periods of time, or even a single NiMH AA battery. Even so, they're packing more and more functionality into smaller and smaller packages.
It's gotten to the point where my *first* thought is "can I do that with an mcu chip?". They're cheap enough that I'm using FIVE mcu chips in my furnace controller, just as smart peripheral controllers - they're programmable, small, and cheap, almost in the same category as PALs or CPLDs.
My concern is that people aren't learning how things *work* any more. They're just assuming they can put an MCU to work and be done with it.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.