Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.
Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.
This question of long timer periods seems to come up fairly often here, in slightly different forms.
mechie's got the short answer right, but here's something else to consider.
How accurate do you want the timer to be, over a period of 1 year?
The problem with using the 555 (or any RC based oscillator/timer) as a timebase for long period timers is one of accuracy. Small inaccuracies in the base time period will be multiplied by a factor of the divider count. A few milliseconds can add up over 1 year. (1 second +/- 5 mS amounts to nearly +/- 2 days over a 1 year period)
RCs don't make the most stable oscillators, either. Frequency variations due to supply voltage* and temperature can also add to the inaccuracy of a long period timer. The digital divider chain does not contribute additional inaccuracy, so concentrate on making the initial oscillator/timer stage as accurate (and stable) as possible.
If you can live with wristwatch type accuracy over a 1 year period, start out with a 32Khz crystal oscillator, then divide down the output to get a 1 year delay. Trim the oscillator frequency using a 1-second tap and a frequency counter, if possible.
* the internal design of the 555 makes it relatively immune to power supply variations; not all RC oscillators are as well designed.
ps: Another possibility for use with shorter time periods is to use the power line frequency as a timebase. There are short-term inaccuracies in the power-line frequency, my understanding is that these are compensated for over time by the utilities. The likelyhood of power outages happening over a 1 year period rules out using this method here.