RobFromLI

HI All,

I'm trying to use a netduino plus 2 to act as a very simple monitor for a momentary switch.  Basically all I'm trying to do is when the switch changes states, post to a URL being monitored by an HTTPListener in an app on a pc on my network.

To test this, although I realize it's not the ideal test bench, I have simply put a pulldown resistor between GND and D1 and then used a short length of bell wire to short from 3.3v to D1, just in order to test.  Now, I realize this is not an ideal situation and I have some more supplies on the way in order to test in a better bench environment.  However, I am seeing behavior that seems bizarre, even given the bounce I would expect from using a wire as a switch and I feel like I am just doing something very basic very wrong.  I have attached a picture of my netduino physical setup - I currently am using a 1k 5% pulldown resistor, I started out with a 10k.  My code is below.  What I am seeing is one of several things, somewhat randomly:

• If I do nothing but turn the LED on or off in the interrupt code based on the status of D1, everything is wonderful - every time I touch the wire to the D1 leg of the resistor the light turns on, and every time I take it away the light turns off.
• If I create a URL and post to it on each interrupt as in the code below:
  • Occasionally, it will work fairly well.  I may get 15 or 20 hits to the HTTPListener each time I touch or take away the wire, which I would expect without debouncing, but it works.
  • Sometimes, I touch the wire once and I get continuous interrupts, with the light constantly flashing and the HTTPListener getting hit over and over again regardless of whether I then take the wire away or touch it
  • Sometimes, the HTTPListener gets the appropriate number of hits when I touch and take away the wire, but the light stays on constantly
  • When I get a constant stream of interrupts, they all seem to have the same time.  I'm not totally positive of this but I saw it at least once

I've been programming for quite a few years and judging by the random behavior I'm getting, I feel like I must be fundamentally misunderstanding the interrupt process, or missing an initialization, or that I am making some sort of grave hardware mistake.  Again, I understand I should have a breadboard and switch but from what I've read of the "experiment with a Netduino" posts out there, what I'm doing doesn't seem that fundamentally flawed that I should be having this many issues.  I'm hoping someone out there can tell me what it is I'm doing wrong.

Thanks very much!

Rob

using System;
using System.Threading;
using Microsoft.SPOT;
using Microsoft.SPOT.Hardware;
using SecretLabs.NETMF.Hardware;
using SecretLabs.NETMF.Hardware.Netduino;
using System.Net;

namespace FirstApp
{
    public class Program
    {

        private static DateTime LastInterrupt;  //Part of attempt at filter - see below

        //Originally defined these in main, didn't make a difference when I moved them
        private static OutputPort onboardLED = new OutputPort(Pins.ONBOARD_LED, false);

        //Tried glitch filter true, didn't help
        private static InputPort D1 = new InterruptPort(Pins.GPIO_PIN_D1, false , Port.ResistorMode.Disabled, Port.InterruptMode.InterruptEdgeBoth);

        private static int reqnum = 0;
        private static String s;
        private static HttpWebRequest test;
        private static uint lastvalue = 0;

        public static void Main()
        {
            D1.OnInterrupt += new NativeEventHandler(button_OnInterrupt);

            LastInterrupt = DateTime.Now.AddDays(-1); //Attempt at debounce - see below

            // Wait forever
            Thread.Sleep(Timeout.Infinite);
        }

        private static void button_OnInterrupt(uint port, uint data, DateTime time)
        {
            //My attempt at debouncing to see if that helps - it didn't seem to.  I tried 50, 100, 500, etc
            //if (LastInterrupt.AddMilliseconds(50 ) > time)
            //    return;
            //LastInterrupt = time;

            onboardLED.Write(data == 1);

            reqnum += 1;
            String s = (data == 1).ToString();

            //My attempt at a simple filter - didn't help
            //if (data == lastvalue)
            //    return;
            //lastvalue = data;

            HttpWebRequest test = (HttpWebRequest)HttpWebRequest.Create(String.Concat("http://192.168.1.80/this?",reqnum.ToString(), "=", s));
            
            //Thought I was getting random network errors at one point - however try/catch does not seem to make a difference
            //try
            //{
                test.GetResponse();
            //}
            //catch (Exception ex)
            //{
            //    Debug.Print(ex.Message);
            //}

        }

    }
}
 

 IMG_2073.JPG   149.64KB   2 downloads

Hi all,

I am a very experienced programmer who has a little bit of electronics knowledge.  Basically I've fooled around in the past with some simple switches, lights and motors and maybe a resistor or two.

I'm looking to hook my netduino 2 plus to a normally open circuit and trigger it to bang a web service on my pc when the circuit closes.  I have the code roughed out using the on board button and now I want to code/test just using a short wire or switch before I hook it up to the larger circuit. 

I've done a good bit of searching and found this http://arduino.cc/en...utton_schem.pngwhich is part of this tutorial

http://www.arduino.c...Tutorial/Button

I just want to make sure before I try this (assuming I hook everything up properly and protect against static and all that) - if I hook a 10k resistor between digital input 1 and ground as a pull "down"(? or up? but I think down?) and then have a switch that connects the pin to 5v, I won't fry anything, is that right?

Do I have to have the pin set to the proper mode when I connect the wire in order to avoid frying something?  In other words if I have a bug and the pin's set to output but I put 5v on it, will it fry then?

Can I uses 3.3v, is that any safer / better?

TIA

Rob