1

I have a number of battery powered, low-power consumption zwave sensors installed around the house, such as contact, PIR or temperature sensors. These are most running on 3v (CR123 or 2 AAA) batteries and takes very little power ( one set of battery lasts for 3month to 1 year.. so maybe < 10ma? )

I am thinking of modifying some of them, to put hole in the battery cover so I can wire it to a 12v wall wart with a buck converter outputting 3V near the sensor. The wire may or may not go through attic/drywall/exterior wall, but the wall wart / buck converter will be accessible, and there will be connectors and fuses to break the circuit.

I read about the regulations and safety concerns, and learned that for low-voltage wiring I would want CL2 for in-wall use, and I should worry about voltage drop for longer runs ( which the buck converter should mitigate). It seems like NEC does not specify too much in the class 2 area?

What am I missing? Is there other related NEC for this kind of setup? What are the regulatory and safety concern should I worry about in this setup? Is this a viable plan?

  • If your wiring enters the wall a plug in adapter would not comply with code even though it is low voltage. A power supply in a junction box that is accessible would allow you to run that communications wire. – Ed Beal Nov 7 '18 at 21:38
  • I see. So I should keep the transformer in a junction box? something like this? diodeled.com/lo-pro-junction-box.html Or I should setup terminals (like the banana plug for speakers) for the wall interfacing? – Moe Tsao Nov 7 '18 at 22:09
  • Yes that way you are not violating code but it specifically mentions cases where items are plugged in not being perminant once you go in or through a wall they consider it perminant. – Ed Beal Nov 8 '18 at 0:33
  • I see.. I guess I should continue changing batteries then. Thanks! If only there are more plug in sensors.. – Moe Tsao Nov 8 '18 at 18:17
2

Low voltage wiring to devices that draw less than 55 watts has greatly relaxed rules, and is covered under a different section of (US) National Electrical Code than mains wiring.

You can see the effect of those rules in how thermostat and doorbell wiring is done -- both of which are subject to Code.

Here, the very fact they can get 3 months runtime on two AAAs implies you have exceptionally low draw devices (in the microamps) that they are not using any more power than you might find on telephone or Ethernet cables (disregarding PoE). So telephone style wiring is appropriate.

Of course, telephones have current limiting resistors at the far end of the circuit, even if you're right next to the central office, the telco won't give you enough current for a short to start a fire.

You should do the same: have current limiting resistors on the "head end" (power supply end) to remove the possibility of significant current flowing, e.g. A resistor calibrated so the devices normally receive sufficient current, but if the wires are dead shorted, only a few milliamps flows - would be about right. So you might consider a 3V supply, skip the buck converter, and put a rightsize resistor on the head-end.

E=IR

E=3V, I=10ma (30mw not likely to ignite drywall paper), R=300 ohm. Assuming your device normally draws <<1ma, it will see <<0.3V drop, you'd have to give it some trial and error to make sure the device isn't browning out at times of key activity such as transmitting, to find the right value of R.

  • Thanks a lot for the response! Great to know this could be classified similar to doorbell wiring. Maybe I can just use cat5 to run these. The buck converter is meant to deal with the voltage drop, if we have 3V only, wouldn't the voltage drop be even more pronounced? Will adding resistor be enough so I can skip fuse? – Moe Tsao Nov 8 '18 at 22:18
  • Voltage drop is proportional to current, these devices are so low current that on #18 thermostat wire, even on #22 intercom wire I would not expect appreciable voltage drop. Cat5 is awfully fine wire at #30 or so, that might be low enough to experience some drop. Or maybe not, I haven't run the numbers. You have to check a voltage drop calc. Or if you don't yet know Ohm's Law this would be a fantastic time to learn. . Yes, a resistor at origin would be better than a fuse, for one thing they don't make fuses that small. – Harper Nov 8 '18 at 22:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.