So I finnaly finished a project that I have been working on for quite sometime. It is a 12hr clock. I tried it for the first and low behold it works...not really. My seconds work, but the minutes and hours seem to change randdomly. At first i thought it might be my power supply, but it did the same thing running off batteries as AC. So that is ruled out. I think it could be do to a small amount of current jumping from one trace to another or even electromagnetic noise. Could this be right or might there be other problems?


Thank you,
Kurtruk

 

Without seeing your schematic all I can say it your clock is broken.

Now if I saw your schematic...

 

I know that my schematic is fine... I think. I got ut apptobed by several people on here some time ago. But if you really need it Ill post it tomorrow.

 

No, we don't really need it. It is you that asked for "Assistance please!!"

 

It's reassuring to hear that you got ut apptobed by several people on here. I think.

I have no idea what you are hoping to get from us. Do you expect us to try to track down your other thread where you got your schematic apptobed?

Do you really think we are going to be able to troubleshoot an unknown circuit based on the description that the minutes and hours change randomly?

If you post your schematic AND you provide a clear and detailed description of what it is doing, and perhaps even a couple of pictures of the circuit, then we may well be able to offer suggestions.

 

Oh, well if your schematic is fine then you did not follow it.

You need to find the mistake in your wiring. Or the broken part.

Or a literal insect that crawled onto your board and is shorting together some vital signals.

 

I have attached my schematic and a video of what it is doing. I am sorry I misspelled approved as apptobed.
Last night when I tried it the minutes section and the hours section were changing "randomly" much more frantically then in the video.
Note segment g of the most right digit is broken, and therefor doesn't light up.

I guess I am a newbie and expected to get a magical answer, and therefor received rude responses.

My entire schematic is split up into 3 sections.
1. The displays including the resistors
2. The BCD to 7 segment display decoders (4511s)
3. The 1HZ time base, counters (4518s) and logic circuits.

I am pretty sure the problem is in the third board, because I have tested the other boards fully and they are fully functional.

I couldn't upload the movie to AAC so here is the link:
https://docs.google.com/file/d/0B8Zph6cDsOXVUlluZG1UZXIzeHc/edit?usp=sharing

 

No, we are not being rude. We are trying to set you straight.

1) Post a subject title that is meaningful. "Assistance please" helps no one.
A better title would have been "12hr clock runs erratically"

2) We are not mind readers. No one could offer a suggestion without seeing the schematics and perhaps photos of your component layout.

3) You jumped to conclusions when you say:

I think it could be do to a small amount of current jumping from one trace to another or even electromagnetic noise. Could this be right or might there be other problems?

4) Put yourself in the shoes of someone reading your post. Can they understand your request and be able to provide constructive advice? i.e. you open yourself up to some jovial poking in the ribs. It gets boring around here sometimes and we just like to mess around too.

 

Ok I cant change the title now.

Notice the words "I think" and "could be".

Do you have any advice on my problem or tips to troubleshoot?

 

What voltage regulator are you using? If it's an LM78xx, I would add a 10μF cap across the output.

Segment G in the seconds ones digit doesn't work.

Hard to tell from the video, but it looks like some of the component leads are pretty long, maybe picking up noise. Post close-up photos of the assembly.

 

Thanks for the reply;
my voltage regulator is a LM 7805C. I don't have a 10uF Cap, The closest I have is a 47uF. Would that be fine?

I will attach some pictures. They are pretty low quality but I don't have a better camera so that will have to do. Since 6 is the max number of attachments I will upload the rest in the next post.

I also took another video and did a lot of testing.
The top LED in the video is connected to the first BDC output of the ones digit of the seconds. The bottom LED is connected to the first BCD output of the ones digit of the minutes.
Here's the link:


I think it might be that my OR gate IC is fried. I think I will desolder it and test it on a breadboard. Would this be wise?

The output of the AND behaves as it should as far as I can tell.

 

Oops forget the link:
https://docs.google.com/file/d/0B8Zph6cDsOXVbllfSV96OTNZZVU/edit?usp=sharing

Here's more pics:

 

I tried the 47uF capacitor across the power supply and it did not make a difference.

 

Are there pins on some of the ICs that are not soldered to the PCB? I think leaving gates floating can cause erratic operation.

 

I had one OR gate that wasn't connected to ground. I connected it to ground still the same erratic results. But it got me thinking, what about the switches when open they are out "in limbo"?

I tried it and it works as far as I can tell!!!!

But I need to make it so when the switch is open it is grounded, but wjen it is closed it is not. Is there a way to do this with a only a SPST push button switch? I think I need to do something with a pull down resistor but Im not sure.



THANKS TO EVERYONE WHO HAS HELPED SO FAR IT IS MICH APPRECIATED

 

I have attached my schematic.

Thank you.

A very high likelihood is that your Hour and Minute set switches are normally floating. This is BAD!!! CMOS inputs should NEVER be left floating. They become extremely noise sensitive and may even result in shoot-through currents that can destroy the chip. You need a passive pull-up or pull-down resistor on these inputs.

This is a very strange design and, frankly, I can't see how it works. You have a bunch of counters that you gate the enable pin on, but the clock pin is hard tied to ground! How can any of them every count?

Why do you have resistors on some (but, oddly, not all) of the counter reset pins?

 

I had one OR gate that wasn't connected to ground. I connected it to ground still the same erratic results. But it got me thinking, what about the switches when open they are out "in limbo"?

I tried it and it works as far as I can tell!!!!

But I need to make it so when the switch is open it is grounded, but wjen it is closed it is not. Is there a way to do this with a only a SPST push button switch? I think I need to do something with a pull down resistor but Im not sure.



THANKS TO EVERYONE WHO HAS HELPED SO FAR IT IS MICH APPRECIATED

Ah, so you already focused in on the issue. Good.

If you want the input to be LO when the switch is opened, just tie a fairly high valued resistor, say 47kΩ to 100kΩ, between the input pin and ground. If you want it to normally be HI, just tied it to Vcc instead of ground.

Your switches will also bounce when they open or close, which will make it seem like you are getting multiple clock pulses instead of just one. There are several approaches to switch debouncing that we can go into if you would like.

 

First off how would I construct a pull down resistor? I have verified that my problem is the floatin pins already. I hope it did not destroy my Ic.. That would stink.


My design may be strange...but Hey it works...hopefully. The enable is the same ad the clock except it counts on the down slope instead of the up.

I don't really understand the resistors, someone said I needed them, and could not give a clear explanation of why.

 

Sorry I didn't see your second post until I made my post but anyways. I think the switches are tacticle so they shouldn't have too much bounce. Is it easy to eliminate switch bounce?

 

Pull up and pull down resistors are just high value resistors that tie an input to either +V (pull up) or ground (pull down). Because of their high resistance, they are easily "overcome" when the input is switched to the opposite polarity. Thus, the input is never floating.