Problem with digitalWrite extending interrupt count

@inst-tech 

👍

It could be a none smooth pulse curve, causing two momentary rise/falls which the triggers the interrupt?

This is what happens when coding switches to deal with switch debounce.

I shall think on.

Yes, thanks, you too, have a great day.

Ron B

Posted by: @ronbentley1

@inst-tech 

👍

It could be a none smooth pulse curve, causing two momentary rise/falls which the triggers the interrupt?

This is what happens when coding switches to deal with switch debounce.

I shall think on.

Yes, thanks, you too, have a great day.

Ron B

I have been using an oscilloscope to monitor the wave form.. and its a clean square wave from the signal generator to the interrupt input( pin 2).. nothing suggest any abnormalities in pulse duration or wave signature, and frequency is +/- 1 hz , not enough frequency shift to cause this phenomenon. 

The fact that it is repeatable every time, no matter what frequency is inputted, is what is puzzling..

I'm wondering if because the serial port shares communication with the interrupt pins, the first pass thru it is just out of sync, and then next clock cycle syncs up?  That kinda makes since to me, but still doesn't tell me why it is so..

eventually, the solution will come...lol

Hang in there, we're getting closer now...

regards,

LouisR

@inst-tech 

I have neither a signal gen or oscilloscope so my comments are purely grey matter driven, unfortunately.

Here's an approach - gather stats over a period of time and then report using serial print functions.  You should then be able to rule in or out any interference?

However , I struggle to see if there may be any such inference as external interrupt pins on uno are 2 and 3, whilst the serial functions use pins 0 and 1, I believe.

Worth a try?

@ronbentley1 @inst-tech I have been watching this thread for a while and keep hoping the OP will change pins. Check pinout diagrams and try a pin with NO other function, maybe try many, if they all act the same then we really have a great mystery, but my 'gut' is screaming that this is pin related. If I had the kit I would already be doing that but I don't so can't.

@inst-tech 

Hi Louis,

I put together a short C++ sketch to perform the function of a square wave signal generator, configured for 1 htz cycles. 

I uploaded the sketch to a MEGA (just what I have around) and connected this to the signal input interrupt port of your sketch  (pin 2), which I loaded to an UNO - I did modify the sketch a little to take out the battery check function call (also, this has a delay of 2 secs in it which affects the count reporting if left in).

Anyway, I found that your sketch works as expected, with the interrupt routine correctly clocking the pulses.  So, notwithstanding any specific errors in the author's original post and reported issues, the process works as expected.

I can post the sketches if that would be instructive.  

Cheers

Ron B

PS I did also test the arrangement and config for other frequencies, ie 2 htz, and worked fine.

Posted by: @ronbentley1

@inst-tech 

Hi Louis,

I put together a short C++ sketch to perform the function of a square wave signal generator, configured for 1 htz cycles. 

I uploaded the sketch to a MEGA (just what I have around) and connected this to the signal input interrupt port of your sketch  (pin 2), which I loaded to an UNO - I did modify the sketch a little to take out the battery check function call (also, this has a delay of 2 secs in it which affects the count reporting if left in).

Anyway, I found that your sketch works as expected, with the interrupt routine correctly clocking the pulses.  So, notwithstanding any specific errors in the author's original post and reported issues, the process works as expected.

I can post the sketches if that would be instructive.  

Cheers

Ron B

PS I did also test the arrangement and config for other frequencies, ie 2 htz, and worked fine.

Good Work Ron Bentley!...Yes, I have concluded that the sketch works fine and as expected..all that's left now is to find out why the timing starts in the sketch in the middle of the first pulse..I haven't found anything as yet, but haven't given up.. someone, somewhere, has encountered this phenomenon, and can explain it!..lol  as far as making it work goes, we've accomplished that goal, now to move on to other more pressing problems. 

Thanks again for confirming my test results..

Regards,

LouisR

@inst-tech 

Louis

Could this be the feature being experienced:

When a pin is configured as an INPUT with pinMode() , and read with digitalRead() , the Arduino (ATmega) will report HIGH if: a voltage greater than 3.0V is present at the pin (5V boards) a voltage greater than 2.0V is present at the pin (3.3V boards)26 Jun 2021

The interrupt will fire/trigger as soon as the  threshold is reached and thetefore before fully high if you are detecting this phenomenon on the oscilloscope

Cheers

Ron b