Home
LED Digital Clock





I built this clock a long time ago when I had nothing better to do. This page is just me recording it for posterity I suppose. If I were to build it again today, I would do it very differently. This was before I had a computer or the internet, so I had no easy way of researching how to achieve what I wanted to achieve. Which was, to build a digital clock with a seconds display
using discrete logic chips. So some of the design is quite quirky and probably unnecessary. The original circuit diagram is shown below, drawn on paper, as was in those days!





The clock
takes its input from a 1 Hz pulse generator (explained later) shown entering top right. This drives a set of 7 segment decade counters (ICs 1-6) which in turn are reset at the appropriate count by another set of decade counters (ICs 7-10). To set the time I employed an unconventional method of using 'one way biased' SPDT toggle switches. This was my way of overcoming the nuisance of contact bounce. The clock inputs require a HIGH to advance the count, but have to be returned to LOW before accepting the next pulse. The toggle switches, being physically biased to ground, always achieve this condition before the next press. It doesn't matter how much contact bounce there is, any amount of '1s' are not going to have any more effect than the first, and so are ignored until the next press of the switch

           
                Rear showing the set switches and DC socket                                                 Front with LEDs aglowin'

The anti bounce method described above is really just for setting the minutes, but I used three identical switches so they would match! One press advances the minutes by a count of one etc. The seconds are zeroed by taking the reset pins of the seconds counters HIGH. If the switch is held down, the seconds will stay at '00' (and the whole clock is halted) until it is released. This is handy for starting the count off when you hear the radio time signal. For setting the hours, there is a continuously running multivibrator (IC12 c&d) whose relatively fast output is injected into the clock input of the minutes counter. This advances the minutes and hours at an accelerated rate to quickly set the time. There is no reverse count though, so if you overshoot you've got to go round again!

Basically then... the seconds counters reset at 59 (60th count) which clocks the minutes counters which also reset at 59 (60th count) which clocks the hours counters which reset at 23 (24th count) when all the counters will be at zero. The display shows '00.00.00' and the whole cycle starts again. An electronic clock is only as accurate as its timing reference... in this case, a 32.768kHz quartz crystal. I cheated a bit here by purchasing a cheap quartz analogue clock from poundland (yes it was a pound) and modifying the timer from that. The original paper drawn circuit is shown below





The block on the right is the module from the analogue clock (I called it a divider IC).
As the original clock was powered by an AA battery, the power supply is now derived from the forward voltage across a red LED which is just under 2 volts. I also replaced the 'cheapo' crystal with a better one from an old VHS recorder. This module divides the crystal frequency down to 1Hz, but the pulses alternate positive and negative and at a low level. This was resolved by feeding it into a bridge rectifier to flip the negative pulse positive, then into a transistor amplifier, and finally through a gate to make it into a clean logic signal


           


The left photo above shows the layout. Everything is assembled on a piece of blank circuit board which also serves as the rear panel of the clock. The main components are mounted on a perforated board and hard wired on the opposite side as shown in the right hand photo. It took me ages... no wonder they invented printed circuit boards! The small veroboard on the left is the timing circuit. The clock is powered from 5 volts at not much more than 100mA and so will run quite happily from a USB socket
. I made notes at the time of the circuit board connections and IC pinouts if needed for future fault finding, but as of yet, this little clock is still working flawlessly after all these years!





The case was made from a length of pine battening, mitred to make a frame. The glass was from a cheap clip frame sprayed black on the back with a rectangle masked off for the display with a red filter behind. I suppose I could have just bought a picture frame, but by making it myself, everything fits and and looks how I want it to




The finished clock