jrlambs
Posts: 2
Joined: Tue Feb 04, 2014 2:48 am

Re: Home Alarm System Retrofit - "Part Deux"

Tue Feb 04, 2014 9:42 pm

Oh, the sensor's that I have that I'm trying to wire up to the pi are Honeywell DT-720B's.

the C/NC wires are listed as "Alarm 500 mA 30 VDC"
the Power wires are listed as "Power 25 mA 7.5-16 VDC"

revell
Posts: 1
Joined: Thu Jan 14, 2016 9:19 pm

Re: Home Alarm System Retrofit - "Part Deux"

Thu Jan 14, 2016 10:11 pm

jrlambs, no expert here, but based to trial and error :) it seems that the NC and C is just a simple circuit with an internal relay. When the alarm is off, the circuit is usually closed (that's what the NC stands for - Normally Closed). If movement is detected, the internal relay opens. If you have a multimeter with a "sound icon" on it (remember, no expert :)), you can test if the two outputs are connected to each other when the PIR hasn't been triggered (no motion).

Based on this, I've been experimenting by applying 12v, 5v and 3v to the NC/C circuit (had no real alarm to test with). Based on my little experiment, I think PIRs (and probably most detection appliance) are supposed to use 3v. I've wired the sensor to the RPI 3v rail and to any GPIO as input (3v should be safe, experts can better comment this topic).

You can also wire multiple sensors in chain, if you want/have to do zones (f.e. if you're short on GPIO pins). Docs state that you can wire C from the first to NC on the next PIR. This would result in alarm trigger if any of the sensors detects motion which makes sense.

I should note that I'm playing with some old Paradox sensors from 1998/2001. Some sites refer to analog and digital PIRs and I've no idea if that's a marketing thing or it's something you should take into account with your sensors.

Naturally you'll get to the point where you want to be able to disarm the alarm (if you think of realworld use). If you consider RFID, that's where C shine but it's a whole new world to explore... Emails to disarm the alarm is not convenient, at least not for everyday use. The traditional keypads are flawed by the fact that most people enter the same code = press the same keys which seems like a weak point to me (not that NFC is bulletproof). What makes more sense to me is have use RFID and generate new hash after every usage.

Return to “Automation, sensing and robotics”