[XCSSA] OT: simple, reliable wireless sensors

[xcssa@xcssa.org]

Is this for a alarm system?

It's commonly believed that the biggest problem with burglar alarms is 
false alarms.  For that reason, police are notoriously unreliable in 
responding to alarm calls from monitoring companies.  In San Antonio, 
all alarms must be registered with the city, or you pay a fine for false 
alarms.  If registered, you are allowed a couple of false alarms a year, 
or something like that.

I've had a monitored alarm system since 1998.  Since then, I've had only 
one false alarm.  The source of it I've never been able to determine, 
however my guess was that a dog jumped up on the sliding glass door, 
causing just enough of the required impulse and crackle sounds to set 
off an acoustic glassbreak detector.  If that were true, perhaps it 
wasn't exactly a false alarm.  There were some paw prints on the door, 
but they could have been there for a long time since I don't clean the 
glass often.

That's pretty good performance.  I've heard of lots of people having 
problems with early generation glassbreaks being set off by 
thunderstorms, motion detectors being set off by pets, etc.  I have 
later generation glassbreaks and dual-technology motions that are said 
to be insensitive to pets, but YMMV.  I didn't need any new open 
detectors because they were installed when my house was built, however 
when I had the new system installed in 1998, the installer checked them 
all out and replaced a few sticky ones.

I had a real attempted breakin (or possibly just vandalism) a couple of 
months ago.  Two windows were broken and one glassbreak set off the 
alarm and apparently scared away the perps.  (I think I had piled boxes 
too close to the other glassbreak, rendering it ineffective.)  The 
police did come out, but that might have been because my neighbor called 
the police himself...I don't know whether they would have come out if 
only called by the monitoring company.  The monitoring company also 
called both me (at a hotel) and my brother-in-law.  I think if you're 
serious, you will have monitoring anyway, though it costs $28 a month or 
so.  A monitoring company will probably only accept alarms made of UL 
approved parts installed by a licensed installer, or anything they 
install themselves.  Most such things are either 100% wired or very 
expensive.  Cheaper to pay for professional wired installation than 
approved wireless.  You can get a basic system "free" with a multi-year 
monitoring contract.

Because of the above, I'd be very worried about using a homebrew 
wireless system.  But cheap wireless systems are not entirely uncommon. 
 IIRC, X-10 used to sell at least one; maybe they still do.  I still 
have the console and sensors from one of those, which I'd be happy to 
let you check out.

Good luck.  And don't forget good locks, strikes, and lighting.  Glass 
doors are notioriously insecure, but can be secured with locking charley 
bar or pin locks.  X-10 is pretty good for automated security lighting.

Charles

xcssa-admin@xcssa.org wrote:

> Hey folks,
>
> Supposing that I could not run wire to all the NC sensors I want to
> deploy without drilling holes and running cable, what would be a
> relatively inexpensive wireless communication protocol that only needs
> to transmit a short-range binary signal (open/closed), without
> interfering with extant 802.11 and 5.2GHz ranges?  I have a ham
> license but don't want to use the spectrum as a trash bin.
>
> I want to avoid having to buy a PC and a wifi NIC, and run off
> batteries for, say, 24 hours,
> so it need to be relatively simple I would think.
>
> I was thinking of a timer circuit and a DTMF encoder that sends a
> "circuit closed" heartbeat every N seconds, and maybe a "circuit open"
> code for immediate notification.
>
> Also, I'd like a small, cheap design for another module where the only
> requirements are that it be battery-powered and listen for a (very
> rare) signal and then activate another circuit.  It would need to
> consume virtually no power unless the signal is sent.
>
> I'm not too worried about jamming or stealth, I just want something 
> reliable.
>
> If anyone has any info on Tesla-style radio-powered circuits, I
> suppose like RFID, I'd be interested in those too.
>
> Let the brainstorming begin!