I've done a few of the sample projects and looking to start a real-world application. I want a system to protect my shipping container from a variety of break-in attempts. they seem pretty secure, but battery powered saw-zall and even gas cutting torches aren't totally out of the question. I have solar panels to keep a battery system topped off, but would rather put the panels away and only bring them out when I'm there (if possible) I have a variety of sensor ideas and would like to check for any change in state rather than always running a NC or NO circuit. if someone makes an attempt that changes a sensor state and trips the alarm, I'd like the neighbor to be able to hit some kind of a reset button and have the system begin monitoring again with everything in the new state.I did do a little reading on using interrupts to let the processor power down, and I think that's where I want to go with my design. I'm best at starting from an existing project and editing to get where I want, since I'm not yet good enough to create much beyond "hello world" from scratch :( Anyone able to recommend or provide at least a start point??

Many Thanks,

RJ

Yeah a good starting point would be to define exactly what it is you want to do and why you want to do it.

Step it out exactly what you want to do, then maybe we can help you further, but you have to define exactly what you want to do.

In fact once you can outline what you want to do the next step is to write some pseudo code.

Start with some value holders and maybe asks yourself "if this then do what".

We'll help but you have to get clear in your definition of what you want to do and when and why.

Ralph

Ralph, It took a few days thinking about it, then some other distractions, but here's a list of operating requirements, and conditions..

sw1 - main on/off key switch inside the container to cut power to the system

sw2 - NOHC standard alarm magnetic switch mounted inside the door

sw3 - NO alarm magnetic switch mounted against the door that would close when a magnet is placed in a secret spot on the outside of the door (disarm while opening the door) I figure even a keyed switch on the outside could be attacked/disabled.. they can't mess with something they can't see..

sw4 - NCHO spring contacts held open by non-conductive pin or spade connected to a trip wire. Just in case they cut thru the side and try to move something out, pulling the tripwire would set off the alarm

sw5 - NC second trip wire with the opposite intent of sw4. just in case they decide to cut the trip wires this circuit will open (and set off the alarm)

sw6 - NO pushbutton that a trusted neighbor could use to reset the alarm for some period??

operation (assuming open switches show 0v and closed switches show 5v) arming 1. place a magnet on the secret spot on the outside of the door (holds sw3 closed) 2. turn on sw1, powering up the alarm system (monitor sw3 but do nothing else) 3. plug trip wires in for sw4 (open) and sw5 (closed) 4. close and put padlocks on door (sw2 would be closed at this point) 5. remove magnet from door, opening sw3 and fully arming alarm

while sw3=0
power up every 20 seconds and check
if sw2=0 or sw4=0 or sw5=1
do nothing - power down
else wait 5 seconds and begin (alarm)

alarm
power to inside and outside alarm for 5-10 minutes
power to outside strobe until reset button is pushed

assuming door is forced open or one or more trip wires have been disable i'd like the reset button to monitor the 'new' condition and set alarm off only if those conditions change- that way if a thief returns after the alarm was reset, he's foiled again when he trips something else :)

if the neighbor calls me, i have a minimum 2 hr commute, and could fix/reset back to the original setup..

Thanks for asking me to run thru this, it does start to form some code in my head. If something isn't clear, just ask for clarification. It is easy for me to get confused on what 'open' and 'closed' will represent at different states..

i'm OK hacking thru the monitoring code, it's the powersaving stuff i haven't used before. after my initial post, i did read up on powering down features on the chip, but not using any kind of 'sleep' modes

Thanks again Ralph..

RJ

After reading thru my post, it occurs to me that sw4 and sw5 could be made obsolete by a motion sensor inside the container. it would be hard to fool, and wouldn't need to be 'reset' after the intruder runs off (hoping this would be the initial reaction). I'll poke around for a motion sensing circuit to greatly simplify things...

RJ

Oh, i'm lovin' the 'WALK - DON'T WALK' code posted here:

http://www.nerdkits.com/forum/thread/1407/

it monitors the motion sensor and does something... I may have found my starting code (minus powersaving)

rjardo, how about some proximity sensor or capacitive touch sensors.

Set off your alarm before they break in!

You could time the touch sensor so a casual swipe would not set off the alarm but a resting hand would.

Ralph

Also what about sound setting off the alarm? If someone hits your container with a hammer set off the alarm.

Ralph

Ralph,

My big fear is a cutting torch or sawzall.. I have a theory that the toughest lock to open is one you don't even know exists, so I plan to have a locking feature that consists of padlocks (that can be picked by someone with enough time/talent), but also a hidden feature that must be actuated/turned to move a bolt back out of place and allow one of the doors to swing open. with all that extra safety on the door, I worry someone will just give up (after cutting the locks off and STILL not getting in)and just go thru the side.. I like the sound activated sensor idea, but would have to measure several times over time to determine that a tree limb didn't just fall on the top. The sound threshold could be set pretty high since the inside of the container is very enclosed and loud. I'll have to figure out the sound sensing and incorporate it.

I'll work out the initial code and get something up here for some of the real programmers to laugh at..

Thanks,

RJ