motion detector

[gamblers21]

I am looking for a circuit for an IR/motion detector security light with the following specs.<p>Time on settable up to 12 minutes<p>110 volt operation<p>will not trigger during daylight<p>will only stay on the length of preset time regardless of how it is triggered. i.e., if the power is interrupted, I desire that the unit only stay on say a preset 4 minutes then reset itself. The only kind I am able to buy will remain on all night if triggered by a power interruption.<p>wager2

[Chris Smith]

Because most of the "over the counter" units actually work on a filtered and lower DC voltage to run the sensor board, I would modify a standard unit to have a isolated and buffered voltage comprised of rechargeable batteries and charge circuit to hold the circuit board on during power glitches and power outages. Basically a tiny "UPS". Then you can modify the timer settings on the switch by altering the given resistor values or replacing it with a pot. When its triggered on, the 12 minutes apply no matter what. The lights will not work during the power off, but the timer and sensor will remain timing the 12 minutes. If you want a start up time different than this after a "power off" situation, some sort logic must be added to some sort of board circuit triggered by power on or return. Also most motion/light sensors only turn on during day light hours if the "Test" setting is on.

[gamblers21]

Thanks for the reply Chris. I was hoping to find a simple solution such as being able to disable whatever it is that makes the unit stay on after a power interruption. Surely there is a very simple solution out there somewhere.

[Chris Smith]

Not all motion detectors stay on after a power outage. On some of my units I have to spike the power on fast, [on and off] in order to make it stay on? Im also not sure if this is a function of the unit in question, or just a quirk? Some "stick" on, others work normally? However, if you isolate the power supply to the units board, any brand, it wont stick on.

[gamblers21]

Thanks again. I am not very as well versed in isolation of a power supply. These units seem to have ac input thru a diode bridge to the rest of the circuitry. I suppose if I had a schematic of the unit I might be able to understand it better. <p>Have you any suggestions on isolating the power supply?

[Chris Smith]

Yes,...... after the power supply going into the bridge rectifier, it usually has a semi large electrolytic cap to hold down the ripple and stabilize the voltage, which then leads onto the control circuits. If you cut the trace out at this point, after this cap or general area, and run the power out into a resistor to trickle charge the appropriate amount of batteries in a battery pack, and then add another diode coming out of the battery pack back into the other side of the cut in the trace on the board, as well as another 10uf cap or larger at this point, both with the cap and battery pack negative leads going to ground. Also add back in another diode to bridge the cut in the circuit board to feed and run the standard circuit. The by pass or jumper diode should hold the power to the board on as long as the power is running, while the batteries and second diode will cut in instantly should the power fail, and the current capacity of the batteries will limit the amount of time you have. The second cap holds tiny voltage transients to a minimal, while back flow is removed by the two diodes which would otherwise trigger the circuit into a spike or hold signal that keeps the lights on by accident or design. The dropping resistor should be able to pass a C-1/10 current to charge the batteries. That's 1/10th of the amp hour rating of the battery pack. If left to discharge, the batteries my need time to recover before the circuit works again properly upon further power outages. Most of these circuit boards draw less than 100ma's so a 1000 ma battery pack should last for ten hours or more? You can check for draw and decide how many batteries you need based upon those figures. The solid state relay and opto-isolator will try and turn on but with out the 120 volts to drive the main lights, it wont matter or affect the circuit. I have a Simple drawing of this but I havent done any "Immages" in this forum yet. I'll check the Private message possibility and your email if its there, and try that way if you need the drawing? Let me know?

[slimjim]

After reading the above very useful information I thought I might add a little on the subject of motion sensors.
Here's some info on how I have modified the sensors. I use the sensor head only, supply them with 5 volts and pick off the analog signal from one of the quad op-amp pins with a current limiting resistor. Probing with a scope I found the highest signal coming from one of the 4 amps. I replaced the Photo sensor with a 1meg resistor and it will work 24/7. The analog signal is so sensitive it will pick up a squirrel or small bird. Most of the units have a small 5v regulator and current limiting resistor so running them off a 12v battery with 110ac charger keeps them going. Some of the units require 12 volts. The analog signals used to go to another board that drove relays to turn on the Halogen flood lights and drive the VCR to see what happens at night. 6 of these sensors have been on 24/7 for 4 years. Now I have a surplus AB PLC with analog inputs, AC outputs and the ability to do some fancy programming in my spare time. My current problem is during the daylight hours on a partly cloudy day the sensors go nuts because of the heat change.
A while back I saw a question in the N&V mag about picking up the analog out of these sensors. If anyone still wants the info I have collected let me know. I have only modified 4 different sensors some may not be available today.

[Chris Smith]

About twenty years ago I had the Designer kit and about 50 of those head units. [12 cents each !] As I recall in the text, to fix your day light problem, its one of the comparator circuits that has a timing / bias circuit that speeds up or slows down how fast the hot VS cold timing cycle can be before it sets off the unit. This mode was built in to stop heater vents and moving sun light from setting off the sensor prematurely, along with general motion size, sensitivity and speed. If the changes from hot to cold move at a slow pace the sensor under normal recommended settings or values wont set it off. But these unit heads were built so that all of these settings were able to be customized completely. I still have a lot of the original units but the text is hiding from me? If I ever find the text again I‘ll send you a copy or put in on the web. They are very interesting units full of electronic circuits, about forty different types, crammed into a few square millimeters.

[slimjim]

Thanks Chris,
Well it look like I may have bypassed that part of the circuit. I may be wrong! I run trending graphs @.5 seconds per analog signal sample. Its rides @ 2.2 volts and swings +- about 2 volts depending on whats out there. When probing with the scope I did see some slow voltage changes when the sensor was in operation. I really thing the answer is in the PLC software to detect the extreme changes in all 6 sensors around the house and disable the triggering of the cameras. I tried using the photo sensor in a series voltage drop and sample the light change but a amp is needed to get the signal up to a level to detect cloud/no cloud condition. I enjoy the hobby but the time is limited I spend on it. I do remember several years ago one could get the IR kits but at the time I found no use for them. A schematic would be very helpfull.
Thanks again.<p>Most of the ones I have are from TrueValue when they went on sale.