Why does INT2 in DT_INT-18 stop working after 30-40 interrupts?

[jellis00]

Thanks, Sayzer, for this input. All three of your suggestions are logical and good. However, I implemented all three of your suggestions, and it still stops after exactly 23 interrupts on IN2.
Any other ideas from anyone as to why this is happening? Would really appreciate an input from Darrel Taylor.
John Ellis

[Jerson]

What is the interrupt rate? 1 per second? Or 1 every minute?

I would recommend not to use heavy time consumers like LCDOUT and WRITE commands within the ISR as it can create problems.

[mister_e]

I would try to get rid of the WRITE in the ISR first, but yeah, rule #1... keep ISR as short as possible, set some flags and process the dutys in the main loop.

Another thing.. get rid of all ENABLE/DISABLE DEBUG... for entertainment and testing purpose.

[sayzer]

He has 2 seconds of pause in Int routine.
So I assume that the code is written with a logic that allows such a long pause in there.

Also, since there is an LCD, I would have a step number variable.
And, at each line of code, I would number the step variable and show on LCD.
This way, you will see where the program stops; so you can see the step number and check the code.

[jellis00]

Thanks to all of you. All were good suggestions and I have implemented them all. Replacing the WRITEs in the ISR with a FLAG that is polled in the main loop to determine whether to WRITE the time stamp to EEPROM, fixed the problem.

Sayez's idea of using a step variable and LCD to troubelshoot where in code a problem exists is a great idea I had not heard before. Will definitely make use of it in future. Thanks!

This forum is one of the greatest resources that exists in the microcontroller design world. Thanks to all of you for your great participation and contributions!
John Ellis