Problem with 555 clock pulse generator

Thread Starter

szabikka

Joined Sep 3, 2014
92
Hello everyone!

I have recently built a digital clock with 4510 decade counters, 4511 display drivers and an 555 astable as clock pulse generator. However, I have encountered a problem when I tested the circuit. Every now and then the last digit in the seconds display jumps two digits. For example 00:50 -> 00:49 -> 00:47 -> 00:46... The phenomena occurs at random intervals and fortunately rarely (approx. once every 3-5 mins). Today I read about the shoot-through current at the output of the 555 and I think it might be the reason for this "counter glitch". The author of the article I read stated that because of the shoot-through, 555 should not be used with counter chips and microcontrollers. My question is: Is there any way to fix this shoot-through problem without removing the 555 from the circuit? Or if it must be removed what kind of astable should I use instead?

Thank you for the answers in advance!
 

edwardholmes91

Joined Feb 25, 2013
210
It depends if the circuit is time critical. If you are using it as a clock to tell the time, then I'd recommend using a crystal e.g. a 32kHz (32768Hz = 2^15) which would require a 14 stage binary divider and a flip flop for the last stage. This should provide a much more accurate pulse every second.

Let me know if you'd like any more information, I've got a schematic which would work for you.

Kind Regards

Ed
 

DickCappels

Joined Aug 21, 2008
6,531
The recommended solution for shoot-through is to bypass the power and ground pins. When in doubt .01 uf ceramic in parallel with 10 uf or more.

If you post a schematic of that part of the circuit you might receive better suggestions.
 

Thread Starter

szabikka

Joined Sep 3, 2014
92
Thanks for the replies.
Ed, this is a count-down timer. The timing is not so important as it's just a hobby project on breadboard so I will take it apart in a few days anyway. But I can't figure out what can cause the digit "jumping".

Dick, I have both 100 nF ceramic and 470 uF electrolytic caps across my power rails. I usually put a capacitor pair near the supply pin of every second ICs or so, so my breadboards have 16 pieces 100 nF and around 8 pieces 470 uF caps across the power rails.

Here is the circuit. I only added one ceramic and one electrolytic decoupling capacitors here to make it convenient.

counter1.JPG
 

dl324

Joined Mar 30, 2015
11,230
Today I read about the shoot-through current at the output of the 555 and I think it might be the reason for this "counter glitch". The author of the article I read stated that because of the shoot-through, 555 should not be used with counter chips and microcontrollers.
That's a pretty strong statement. Can you give a link to this article?
My question is: Is there any way to fix this shoot-through problem without removing the 555 from the circuit? Or if it must be removed what kind of astable should I use instead?
Without having more context of this "shoot-through" phenonemon, I'd assume it was referring to CMOS current spikes when both output devices are conducting. Your schematic shows a bipolar timer which wouldn't have that problem.
 

AnalogKid

Joined Aug 1, 2013
8,534
There is nothing obviously wrong with the schematic, nothing that would cause a double clock randomly or otherwise, so I agree with others that this is a power quality issue.

I usually put a capacitor pair near the supply pin of every second ICs or so,
Not good enough, especially for chips that can respond in nanoseconds and a breadboard with less-than-perfect ground plane hygene. Standard practice is a 1oo nF cap at every power pin, with the shortest possible connection to each chip's GND pin. This can be backed up with zone capacitors, something like 10 uF to 47 uF every 2 or 3 chips. Power parts like a bipolar 555 get extra treatment, such as its own 47 uF between pins 1 and 8.

ak
 

DickCappels

Joined Aug 21, 2008
6,531
If you are using one of those plastic plug-in breadboards as AnalogKid suspects, then all bets are off.

There is an excellent chance that you are experiencing ground bounce of something similar. Have a look at this document for some ideas.

https://www.fairchildsemi.com/application-notes/AN/AN-640.pdf

One thing that might help is to go to non-buffered CMOS (the so-called A series) and to lower the power supply voltage to reduce the slew rates and response times.
 

Thread Starter

szabikka

Joined Sep 3, 2014
92
That's a pretty strong statement. Can you give a link to this article?
Without having more context of this "shoot-through" phenonemon, I'd assume it was referring to CMOS current spikes when both output devices are conducting. Your schematic shows a bipolar timer which wouldn't have that problem.
Here is the link: http://www.talkingelectronics.com/projects/50 - 555 Circuits/50 - 555 Circuits.html#68

I don't know if I understood it right, but the statement I cited can be found under the sub-title shoot-through.
 

Thread Starter

szabikka

Joined Sep 3, 2014
92
Thanks for the replies everyone! I thought it too, that something is not OK with the power, but I didn't know that "plug" breadboards can't always be trusted. I have experienced te same phenomena when I built a "Knight Rider" running light with an 555 and a 4017, sometimes it missed one of the leds in the row. Now I think the breadboard must have been the culprit in this case too.

I will try to lower the supply power as Dick advised.
 

RichardO

Joined May 4, 2013
2,271
Shoot through current is when both outpu8t transistors are on at the same time. this only happens for an instant.

Yes, a bipolar 555 timer has a _huge_ shoot through current. A CMOS 555 timer does not. As stated, proper power supply bypassing will reduce the effects of shoot through.

About solderless breadboards. They work quite well -- when used properly. You must use short wires and short leads on bypass components. Frequencies into the 10's of MHz are quite possible with close attention to the details of the layout.
 

Thread Starter

szabikka

Joined Sep 3, 2014
92
Thank you for the reply Richard! I have a couple of TLC555CP at home. I never used them as their output current sink/source capability is not as good as the NE555's, but they will be good in this case.
 
Top