# Remote switching via IP

Discussion in 'The Projects Forum' started by Kafter244, Sep 3, 2010.

1. ### Kafter244 Thread Starter New Member

Sep 3, 2010
1
0
I'm working on a circuit whereby I have a contact closure in one room provided by a General Purpose Input/Output (GPIO) connector, and I wish to power a low voltage (c.24v) light in another room.

The contact closure is obviously two contacts (either side of a switch), but the light may be in any room in the building. I was hoping to extend the switch through our vast IT network, using unused RJ45 ports for the ability to patch the switch to anywhere in the building.

The problems...
• using two poles of the switch means only four switches are available via standard 8-core CAT5 cable
• I don't want to risk any voltage travelling through the system in case a cable is mis-patched and throws 24v up a computer!
My theory is that if I have the +ive of a 24v supply at my destination, connected to one side of the light, the other side of the light connected to one terminal of the patch. The switch at the source will switch between ground and the patch, thus, when the contact is closed, a circuit is connected between ground and +ive 24v, via the light. Obviously the -ive pole of the supply would need to be connected to ground also.

I hope that in this way I can use a single cable to control the light (thus having 8 channels via RJ45) and that any electricity will only pass when both ends are correctly connected, thus protecting any equipment accidentally patched from undue voltage.

Is this theory correct? Can anyone suggest, perhaps, a better way, or point out any gaping flaws in my logic? It's been a while since I did some circuit design and my confidence is down! lol.

Thanks in advance for your time and help.

2. ### marshallf3 Well-Known Member

Jul 26, 2010
2,358
203
That would work so long as you aren't trying to drive much current, CAT5 is #24 AWG = 26.17 ohms per 1,000 feet.