Continue to Site

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.

  • 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.

TTL Clock with 7447 7490 and 7400 IC's

Status
Not open for further replies.

RODALCO

Well-Known Member
I have seen various threads with TTL requests regarding how to and build a TTL clock or counter with discrete components.

As I have build around 7 of these clocks in the early 1990's and these TTL, LED clocks are still going strong I decided to have a look in my archives and post the basic diagramme for it.

I have learned building them also by trial and error and from various sources like TTL Cookbook (Don Lancaster) and Digitale klokschakelingen (Herbert Bernstein) written in Dutch origin Germany.

As i like the old TTL technology and with carefull and accurate ratsnest wiring a very accurate clock can be build at very low cost but labour intensive but that doesn't matter for a hobby.

Important is the decoupling of every 74xx IC with a 100 nF disc capacitor.
I omitted these in my earlier clocks and later addition of these was painful but not impossible.

The numbers 6 and 9, can be made with tails by adding an extra 7408 AND gate to drive a BC 547 to sink the A and D segment when these have to light up.
A 74247 IC has those functions built in but i have never seen one of those.
IC's.

Attached is the basic schema which works well and i'm more than happy to give more info to potential TTL clock builders. Also i'm happy to post more photo's if required.

**broken link removed**

:eek:hm: :mu: :eek:hm:
 
Nice diagram!

The first TTL clock I built I used 74393 counters (because I had a bunch of them) 74141 chips and Nixie tubes for display.

After nearly 20 years, it is still operating fine.

TTL logic is still great to work with, reliable, and "oops" tolerant.
 
Hi Folks, I recently knocked up a timebase generator using parts from the junk box. An adjustable 20MHz Xtal and a few 74390's, lovely chips that contain two equivalent 7490 decade counters so you reduce the component count quite a lot. I deliberately clocked them the way I have so that each frequency output is a symetrical square wave. I added the LED as a last minute thing just to let me know that the count chain is working. If you can find an adjustable Xtal like I did, the whole thing can be tuned to a frequency standard.
Les G4CNH
 

Attachments

  • clock.JPG
    clock.JPG
    81.8 KB · Views: 9,875
Last edited:
hello !
Can anyone help me build a frequency generator(60 hz) for my digital clock .i m using 7490 & 7447.but i can't generate 60 hz frquency time base.
 
raiz, you should have started a new thread.Anyways, hope the link could help you.

**broken link removed**
 
I recall purchasing several $5 (USD) kits back in the 70's when several "single chip" clock ICs started appearing on the market. I still have one of these $5 kits laying around here somewhere. They included a little 8-pin DIP oscillator/divider IC, circuit board, 3.579545 crystal, trimmer cap', and a couple discrete components (resistor and cap'?) and they output a 1-Hz TTL level "heartbeat" signal.

Anything like that still floating around now days?

Mike
 
A 555 timer is no way near accurate enough for a digital clock, build a crystal oscillator with the frequency of a multiple of 60Hz then use a divider to divide it down to 60Hz.
 
Using a CD4521 and a 4.194304 crystal will provide a very nice 1hz time base for a clock.:)

Datasheet - **broken link removed**
 
raiz If you look at the schema with TTL clock at the beginning of this post, you have already two 60 counters in there with 7490 IC's.

One is used for the seconds and one is used for the minutes.

If you wire the two 7490's up as in the schema, after the 7413, and use two gates from a 7400 you can use that to decode the 60 Hz from the mains to one Hz for your clock input.

edited for typo's RF.
 
hi guys! I m back but with another problem now I have made the clock it works well but it is a 24 hour clock can any body tell me how to make a 12 hour clock using 7490 ics
 
Get the hour counter to reset after 12?
 
Hi raiz
Attached is an image from the 12 counter.
I posted this in an other thread at this forum too.
Glad that you got it working all right.
I prefer the 24 hour set up for clocks anytime but it is easy to convert to 12 if you like.
I hope that you decoupled all the IC's with a 100 nF capacitor across the 5 volts IC supply points.

**broken link removed**
 
Hi there, another newbie here.

I am building something like this and I found the article very interesting, but I don't know why the 7490 ICs are decoupled with the capacitor. Could anyone explain why this is done and how the power supply is decoupled with a capacitor across it?

Very interesting site btw, I'll be back to check out more.
 
hi,
As the individual transistors in the TTL devices switch ON/OFF they draw a short current spike from the +Vsupply rail.

This spike can pull the Vrail down a few tens of millivolts, this voltage spike then can be detected
by the other TTL gates and cause a mis-operation.

If a decoupling capacitor is connected across the TTL power rails, when this switching current occurs, the capacitor will supply the current and effectively remove the voltage spike.

Also capacitors generally have an impedance [resistance] that decreases with frequency,
so any fast spike on the power lines 'sees' a low resistance path and is suppressed.

This is a simple explanation, you will find as you study the subject of 'noise' its fairly complex.

Its good engineering practice to fit a small capacitor across the power pins of all the TTL ic's.
A low value of about 0.1uf will get the job done, also a 10uF capacitor on the pcb is recommended.

Ask if this is not clear enough.
 
Last edited:
Aah that makes a lot of sense now. Thanks a million for your post, which was very clear:)

I only had a 10 nF capacitor to hand, but I can't get the 7490 to produce a working output yet.

I tried, with a 10K resistor to input a signal into input A (with the crude method of touching it briefly against the +5V wire, but it's stuck on High at QC. Plugging in and out the wire between QA and Input B does vary the output on QB, QC and QD but it is completely random. Sometimes one or two of them flicker and sometimes one or two stay on. QA never lights up and placing a voltage or grounding Input A has no effect. Mabye manually applying a clock is causing problems. I might build a 555 clock later just to see if that works.
 
Input to the 7490 needs to be on the negative edge of a pulse. Trying to manually apply a signal can cause many problems. In effect, you are dealing with extreme switch bounce. Build up the 555 just to see how much better things go.
 
After a bit of figuring out why the device wouldn't work even with a fully functioning 555 circuit, I realised that touching the negative wire changed the state of the QB-QD LEDs so after wondering if I needed NPN or PNP transistors, I just got an ordinary NPN transistor between the input and ground, and applied the signal to the base. That part worked perfectly. So I connected this to Input A and linked QA and Input B together and I got the binary numbers flashing up as they should.

I knew of the negative edge trigger but I thought that it would be on the negative edge of a positive signal into the input. I didn't realise that it was positive already and that applying a positive signal did nothing whatsoever. Output A was at zero potential some of the time I believe, which is why touching the QA and input B wires together produced an output.

Thanks for the help!
 
Thanks ericgibbs for your explanation re the decoupling capacitors.

To be confirmedCheck that the TTL reset gates go somewhere.
if you have a single digit counter, put both to ground, or plus 5 volts.

The inpulse needs to be debounced via a schmitt trigger 7413, otherwise you will get random readings.

I try to put a clearer schema on this site. haven't got software to draw schema's but give it a try in XL.

Regards, Raymond
 
Hero999 said:
A 555 timer is no way near accurate enough for a digital clock, build a crystal oscillator with the frequency of a multiple of 60Hz then use a divider to divide it down to 60Hz.
I disagree, If one uses a cmos version and a good quality film capacitor the
the frequency can be adjusted to exactly 60Hz and it will be very stable. I say a cmos version as it can have much higher value of timing resistors to bring the frequency down to 60Hz.
 
It's still going to be highly temperature dependant, no reason not to use a crystal, they're easy to derive a clock from and very accurate even over wide temperature ranges.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top