Confusion with series and parallel pssibilities

I have a remote solar installation with six lights that I currently monitor with a GSM voltage monitor. Because of its location, it is subject to the occasional vandalism by bored locals. Specifically, they love to shoot out the lights.

I purchased and installed six DC current sensors and controlled each with one light. The sensors have a relay with a common, normally open contact and a normally closed contact. When current begins flowing (the lights are now burning) the relay changes states.

So I ran the battery feed to the GSM monitor through all the relays in a series circuit. So now if all lights are burning, I properly receive the voltage reading from the battery bank. But when a light is out (any light), the series circuit is broken and I get an alarm because of low voltage.

This is a big improvement. But I wanted to improve it even further by coming up with a design that would let me know not just that a light was out, but how many.

Sometimes they shoot just a single light, but other times they may shoot three or more. It would really help to know how critical my response should be and how many fixtures to take with me since the area is a long way to travel to.

So I started by making a voltage divider to break the 12 volt into six segments of 2 volts each and was trying to figure a way to use the relay contacts and anything else short of a PLC to make this work.

My thoughts were to series the 2 volts in such a way that with all lights burning I would see the 12 volt total, but to somehow also parallel this so that if one relay opened due to loss of a light, I would see only 10 volts. And if two lights were shot, then I would see only 8 volts....etc.

This way I could determine how many lights were out at any given failure. But I have been banging my head against a wall trying to make this work. If it can be done using just relays and basic electronics, I would LOVE to know how to build it.

My thanks for any guidance or information that may help, or even if it is just knowing that I am trying the impossible.

John

Hi,

a simple drawing could clarify things.
To be true, I have no clue about your current setup.

What is your supply, how are the lights connected, the relays and the voltage measurement?

Klaus

Think about this , use your current set up but put a 100K resistor across each switch and a 10K across the input of your GSM setup and monitor the volts there.
All panels OK, N X 12 V + 0 ohms, One light out, (N -1) X 12V+ 100 K, two lights out, (N-2) X12V + 200K. You need to load the circuit with some current so the 100Ks drop some extra volts. I'll let you do the final maths
Frank

et413 said:

My thoughts were to series the 2 volts in such a way that with all lights burning I would see the 12 volt total, but to somehow also parallel this so that if one relay opened due to loss of a light, I would see only 10 volts. And if two lights were shot, then I would see only 8 volts....etc.

Click to expand...

Are your lamps in series? This is a method to get a single voltage reading based on number of lights shot. Put a zener across each light. A zener is idle when the light is good. It conducts when the light is shot. The more lights are shot, the lower the voltage



This illustrates the concept of a current-sensing resistor. Maybe your lamps are in parallel? A single resistor can indicate total current drawn by your lamps. Mount the resistor inline with the power supply. Measure voltage across the resistor. The greater the Amperes, the greater the volts.

The current setup (voltage monitoring only) is shown in the attached image. Each light is individually run from the battery bank with the positive conductor run through the respective current sensor. As you can see, I can check the bank voltage during the day because of the timer, and at night if all current sensors are closed (i.e. lights are burning). There is no feasible way to run these lights in series the way that they are installed.

I will read up on the proposed solution using resistors and see if I can make it work. Thanks.

Your lights appear to be in parallel.

This simulation has the single sense resistor, in line with the power supply. At left is normal operation. At right, one light is bad. Notice the change in voltage across the sense resistor



The sense resistor should be the smallest ohm value that yields a usable volt reading. You can make it from a few inches length of wire, rather than a resistor. Add amplification if necessary.