The Atmega328P is going too much fast

What CAD package are you using?

If you really want to make robust working gadget, kindly follow these procedures:

  1. Make a setup using Arduino UNOR3 and its IDE.

  2. Connect the following display unit (Fig-1) with UNOR3. Do not connect the big display devices and its drivers at the momet. This is to investigate the source of noise/problem.


    Figure-1:

  3. Connet your input devices with UNOR3.

  4. Upload your applicatio program into UNOR3.

  5. Observe if the system is working alrigt or not.

  6. If you need driving codes for the multiplexed display unit, please be free to ask.

  7. If things are working as expected, then replace the display unit with the big-sized display devices.

i am using Kicad software

I think many of us here use Kicad so if you post the project we can look at the layout.

Ok i will post my kicad project file

New_Display_REV_Final_with_Hole.zip (2.6 KB)

I might not have time to review it today but I'm sure you will receive comments from many others.

Can you post a picture of the back side of the board?

I have seen bad IC sockets, and times when a pin will get bent underneath the chip instead of going into the socket, although I assume you have removed and re-inserted the chip during testing and would have seen if that occurred.

on front sides there are digits attached and the IC base solders are underneath that digits

  1. It may be the problem that IC legs are bent and are loosely attached with base
  2. We plug out controller and then again plug in the frequency of problems becomes low.


This is the back side of the board

Did you protoboard your project first?

When you were writing your code, where in the process did it start this speed problem?

You did write your code in stages, getting each stage working BEFORE going to the next?
You didn't write it in one big go did you?

Tom.... :smiley: :+1: :coffee: :australia:

Yes i prototype my project first and write codes in stages and during that time i have not seen any of the problem.
After successful testing of my prototype we shift to bulk production, and in bulk production we found such problems.

Almost every board has experience this type of problem once but it is resolved automatically, but in some boards the problem persists and coming again and again.

There is no schematic or PCB file in the ZIP file

Did you make a picture if the back, drill a hole and then make a picture of the front?
One hole is missing in picture 2...
By the way: print looks good...

Please show the new code...

If you archive the project it will put all the necessary files in a zip file for you.

Hi, @abd42

Have you got an oscilloscope to check your crystal frequency?

Also this component arrangement for the crystal and capacitors is far from ideal.

The crystal and capacitors should be as close to the controller as possible.

Tom.... :smiley: :+1: :coffee: :australia:

1 Like

Yes you are right
in this 1 capacitor is near and other is far

But i dont have oscilloscope

New_Display_REV_Final_with_Hole.zip (642.5 KB)

Please check this

I ran ERC and DRC on your board and they came up with dozens of errors and warnings. The clearance issues on the PCB can result in unreliable operation.
Your biggest problem is GND. I has 224 vias in it with a total length of 1.4 meters! That need to be fixed. Also the total via count for the board is 502, that is way to much for such a simple design with only 67 nets. Every via is a possible failure point and you have created many of them.

Overall, it’s just a poor layout that will be prone to problems.

emphasized text[quote="jim-p, post:61, topic:1412376"]
the total via count for the board is 502, that is way to much for such a simple design with only 67 nets
[/quote]

Many appear to be “stitching” between front and back ground planes and shouldn’t be problematic.

I’ve never heard of an avr running “much too fast” due to how issues. I suspect the software.

1 Like