Dead accurate 1Hz clock signal???

[bobleny]

I'm building a clock and would like to produce a steady 1hrz signal. Obviously, the more accurate it is, the better!

Now, I know it's almost impossible to get exactly 1 pulse per second, but I defiantly want it to be close.

Any suggestions?

Thanks!

 

[MP1]

The most accurate way to do this is to use a crystal oscillator and a divider circuit. It can also be done accurately with a micro, since you program the timing. This is basically the same as a programmable divider since all functions in the micro are based off the clock cycles of the micro.

MP

 

[bobleny]

Would a micro controller work good for this?

 

[mvs sarma]

Hi,

Lovely TOPIC. there are time signals broadcast  terestrially. like 10MHz, 15MHZ and many other. they provide time ticks at 1sce-- these are controlled by atomic clock .

second , you should have DCXO- which provides a good standard clock and derive 1sec from this clock-- appears foolish is it not-- but when you used dead perfect 1sec pulse-- you may appreciate perhaps

 

[MNA]

The most accurate way to do this is to use a crystal oscillator and a divider circuit. It can also be done accurately with a micro, since you program the timing. This is basically the same as a programmable divider since all functions in the micro are based off the clock cycles of the micro.

MP

http://www.electronics-lab.com/projects/oscillators_timers/015/index.html

I built a clock with 555 timer and dual decade counters......But it wasn't accurate obviously because of 555....As MP said crystal oscillator will be better.....the above project can be extended to make hrs/min/sec clock.....u will have to use gates to reset the hours section when it strikes 24.......

 

[Fowkc]

If you're powering your device from the mains, another option is to divide down your 50/60Hz signal to get 1Hz.

Not particularly accurate, as the mains frequency wavers quite a lot. If you assume a +/- 1Hz deviation, that could be 30 minutes off in a day!

Of course, most of the time the deviation is a lot less than that. It's quite interesting to see what happens the the frequency based on demand:
http://www.dynamicdemand.co.uk/grid.htm

 

[bobleny]

Ok, crystal oscillators it is! Does any one know where I can look at some? Does anyone have any suggestions for what I should look at?

 

[MNA]

Ok, crystal oscillators it is! Does any one know where I can look at some? Does anyone have any suggestions for what I should look at?

u need a crystal of some frequency,then divide that frequency to suit your need......didn't u look at the project of the electronics-lab project section? .......

 

[mugsby]

see:

http://en.wikipedia.org/wiki/Phase-locked_loop

 

[mvadu]

The simple most circuit for this would be to take the circuit chip out of a dead watch or a table clock. They use a crystal oscillator and dividing technique. You need to cut off the coil winding and add two diodes (IN4148) on both the outputs (P side of it) and join the N-side to get study 1-Hz clock.

 

[mvs sarma]

The simple most circuit for this would be to take the circuit chip out of a dead watch or a table clock. They use a crystal oscillator and dividing technique. You need to cut off the coil winding and add two diodes (IN4148) on both the outputs (P side of it) and join the N-side to get study 1-Hz clock.

Hi
it is working with a single cell-- you need to level shift to TTL or CMOS in order that it is useful


sarma