So how should we design a DIY security system?
So how should we design a DIY security system?
Every time you rationalize in your planning stage for price or convenience STOP.
To be fair there are two main reasons people DIY.
Price. They think they can do it cheaper. Maybe they can, if their time isn't worth the difference between doing it themselves, and having it done.
Quality. They think they can do it better. Maybe they can if they don't rationalize about one camera with a wide angle lens covering an entire parking lot, or assume that when the box on an "affordable" camera says it has an IR range of 60 feet they meant a real world useable 60 feet.
I've found for video surveillance like people really want they need anywhere from 3 to 6 times as many cameras as they settle for, and usually 2-5 times better quality.
Example: An armored dome instead of a soda can or bullet camera costs more, but nobody is going to throw a rope around it and jerk it off the building. (There are cheap domes too.) A centralized system requires running some wire or doing better planning.
With hi-rez cams, for starters. That'll capture the info.
The planning and the wiring are the reasons I want to do it myself, slowly and deliberately with testing before I drill holes and run conduit. I can design electronics of any required complexity. What I don't know is what works and what doesn't.
I've been moving my cheap ones and trail cams around to figure out the views and illumination. Where to put them is not so simple, the best locations for coverage face the sun.
Yeah, that's an easy way to tell, with actual data.
So build roofs for them to keep them in the shade, except for the one or two hours it gets direct sun. It ain't rocket surgery.
Yes, but not the same market.
25+ ears ago, I was called to repair a very low end video surveillance system. The one camera had a bad vidicon. The monitor had a CRT that wouldn't focus. No recorder. The second camera was a dummy. It was why we were called. The NE2 pilot light had died. I repaired both 'cameras', added a second camera and monitor along with a timed video switch to alternate the cameras between the monitors. I couldn't talk them into anything better.
I remember vidicons... with a great deal of distaste. LOL.
Video has gotten affordable in the last decade or so, but people still want to get by with the cheapest thing possible. I've lost a lot of jobs by quoting people what they really want. LOL.
================= An ad just hit my in-box that may be of interest to you [and others thinking of installing a video surveillance system]
I have bought several items from Tiger Direct, and my only complaint is they seem slow to send you the rebates they offer.
I was a Broadcast Engineer and I remember the early Sony digital color studio cameras with pure disgust. An old RCA TK46 with S14,000 worth of new plumicons blew away just about any camera I've seen since then. :(
Yes, I've had people tell me they only wanted an alarm system with a switch on two doors, because no one ever came in through a window. It's quite sad that they are so ignorant.
Even better were the splitter mirrors to "see" in two places with one camera.
I love how even these days, most security cameras can only be used to verify that yes, somebody did something bad. As two who it was, you just can't tell.
Even if they have good cameras, the time lapse recording is so bad that you lose all detail.
I have to be honest. I do still have two time lapse VRT/VHS recorders on the stock shelf, but I have not installed one in atleast ten years. I use all DVRs and NVRs. The recording quality, and frame rate is settable on a per camera and per event basis on my current favorite models. Still, there are limits.
Still, on most DVRs even if you are running a fairly low frame rate the individual frames are not affected by the issues inhernet in old school time lapse recorders. Each frame is as sharp as the source, and your selected storage resolution.
My theory was just catch them instead of taking their photo.
My rear wharehouse door had a bar on the inside with two padlocks, one each end. Overhead door had burglar bar gate with padlock. Front office window and glass door had burglar bars bolted to the frames. Only real way in was the front door. The alarm was silent; direct dial to police. Ten seconds after the alarm tripped, a solenoid bolted the front door. It unlocked the door when the alarm company remotely reset the alarm. That was done only when the police were on scene. It was only needed once when a former employee broke in. And he knew how the alarm worked. The solenoid was not provided by the alarm company. There were many no and nc terminals in the control box. Simple matter to wire into a no.
I'm thinking of using an OR'd array of motion detectors to power up a DVR that records all cameras sequentially or continuously until some timeout after the detectors stop. Motion-sensing floodlights or remote wireless cameras with (separate?) PIR detectors would also turn it on.
An isolated sensor to detect floodlight current:
It would likely run off 12V or an inverter to catch someone looking to steal a generator during an outage, but wouldn't drain the battery by running continuously. Camera placement is determined later, independent from the basic design.
Even with the game cam it takes quite a while to scan through the pix looking for anyone else but me nosing around out back.
On Sat, 13 Dec 2014 12:43:34 -0500, Jim Wilkins wrote: [...]
An alternative to electrically OR'ing sensors is attaching several to an Uno or Mega Arduino and writing a program to detect and report which ones activate. You could put together an inexpensive system with some PIR sensors and try it out. Probably would be entertaining, might be useful. I've used Arduinos (including Pro Minis) in several projects and have always had adequate room for code.
(ATM, miniinthebox.com (which I have no info about) has $3 PIR sensors, $10 Uno Arduino, $15 Mega Arduino, $2 Pro Mini Arduino, $1 radio link. Eg: )
My thought is that OR'ing the sensors and recording everything removes the need to match their view areas to the cameras. The cameras themselves will provide the same movement information without any detection or turn-on delay. I still need to be able to isolate the detectors for setup and to troubleshoot false or small animal triggering, but that's hands-on and doesn't require automation.
I've played with an Arduino but not found any tasks for it I couldn't solve better some other way, for example I'm using these meters with a laptop for a data acquisition system with better resolution, signal conditioning and optical channel isolation.
They have already shown me that my baseboard electric radiators still had enough rug fibers hidden inside the fins to make the internal overtemp sensor trip, which causes rapid cycling, after I had opened them up and vacuumed them "clean". Compressed air did the trick.
PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.