4040 Driven Clock

[Gene]

I need help. I am trying to build an LED clock. I have used a 555 IC for pulses and divided them with a series if 4017 ICs such that I get a single pulse each hour. Now, the idea was to use 12 LEDs and connect them so that one comes (only one) on each hour - the next hour a different LED should come on, etc. I was planning to use a 4040 IC to link the pulse generator to the 12 LEDs but my display is erratic. At hour number 1, LED number 1 comes on (great). But in hour number 2, LED 1 and 2 come on. In hour number 3, LED 1,2, & 4 come on. Does anyone have a suggestion for wiring up a 4040 IC to display 12 pulses and continue. Thanks...

 

[kinjalgp]

4040 is a binary ripple counter and it will display the clock in binary form. You want your output to be in decoded decimal form so you use one more stage of two cascaded 4017 ICs and feed this stage from your 1Hour pulse generator.

Before proceeding let me tell you one thing that this clock won't be accurate at all since it uses RC oscillator. For highest accuracy use crystal oscillators.

 

[Gene]

Thanks. After counting the input pulses, comparing them to (some of) the output pulses, and re-looking the 4040 data sheet I see that you are correct. Since this morning, I have set up a single 4017 and have the desired effect - at least through 10 LEDs. Any suggestions as to where I can find a circuit schematic for adding the second 4017 so that I can get to twelve before the first 4017 starts counting again?

You are also correct - this is an R-C fed 555. However, the circuit is in three sections. The oscillator, the divider, and the display. Once everything is running properly, I would like to re-work the oscillator to be crystal controlled as you suggest.

 

[bogdanfirst]

it is a littel bit to make a 12 counter. but the suggestion is to use a binary counter and then a decoder.

 

[kinjalgp]

Why to do so if already a counter with decoded outputs is available??? and that is 4017.

 

[bogdanfirst]

but it only counts to 10. you need another type of counter, to at least 12, in order to make the leds show the hours.
use a 22 counter, or something else.

 

[kinjalgp]

I hope you know about cascading of counters. If you use 4017 you'll require a total of 2 ICs in cascade mode whereas if we use the method you have suggested then propably 3 ICs namely the binary counter and 2 decoders each having eight decoded outputs (like 74138 or similar). Also you'll have to tie up all the enables and such stuff to respective logic levels.
So which one do you think is better?

 

[bogdanfirst]

i idnt say mine i better. it was just as a posibility. but how do you make a 12 counter with 4017, and have 12 outputs, only one high at the time?

 

[Gene]

Thanks to all who are following this problem. I do not have an issue with using a pair of 4017s in cascade but I have not been able to stop the first IC from recycling after ten counts. I was looking for a circuit that would cascade a pair.

Also, I do not have an issue with using the first 12 counts to the 4040 and then decoding from a truth table but I will need help stopping the count at the right point and arranging the decode logic.

I am even willing to look at other IC options although, I prefer the 4000 series for the operating voltage range. FYI, there are no part suppliers in my area so any ICs that I don't have will have to be mail ordered.

Since the interest in a 12 count/running LED project is still getting some thought here, I should share a little more information about my project. The clock is actually a tide clock with one LED representing high tide and another representing low tide. the other LEDs represent "hours" until the next tide change or approximately 62.5 minutes each. As a last resort, I could divide the lunar day into ten segments of approximately 75 minutes each but the ideal situation (for me) would be 12 divisions. I'm not giving up and appreciate all input.

 

[kinjalgp]

Here is what I was trying to say: Bogdanfirst, post your schematic too so we can compare and find out the best suitable option.

 

[kinjalgp]

Here is one more cascading method which Philips datasheet suggests,

 

[Gene]

kinjalgp - This problem is starting to get interesting. I will work on your first idea this weekend but will need to order the 4049 - so that's a week delay. I have looked locally for the 74HC (that I think I could use with a zenier at Vcc) or the CD4049 (that would be perfect) for the inverter but no luck. I haven't figured out the logic of the second circuit yet.

Thanks for sticking with me on this one,
Gene

 

[kinjalgp]

If your Vcc is 5V then you can even use 74xx04 or 74xx14 inverters instead of CD4049.

 

[Gene]

Good news! I found the 4049, built the circuit (not soldered up), and it almost works. For some reason, Q0 of IC1 does not fire. My solution was to call Q1 of IC1 LED0 which made LED 9 LED 8 and LED 10 became LED 9. Now, I added an LED to Q4 of IC2 which became LED 12. Finally, I moved the reset from Q4 of IC2 to Q5 of IC2. Your circuit now works great. Am I missing something here or is this an acceptable solution?

The next step is to hard-wire the display circuit - after I hear from you about my solution. After that, I will be looking for a circuit for a battery back-up (I'm thinking a 9 volt battery with a diode). When it is all working, I will re-calculate my 555 frequency and working on a crystal driven oscillator up front.