The hard part of installation is running the wires to the sensors at the door. Wiring to a few screw down terminals will be done before the soldering iron heats up. Besides, I can’t solder worth anything, so you have to include the time of watching videos and practicing in any solution that requires soldering.
I do like the simplicity of wiring your example with the neopixel ring, but it’s brightness just doesn’t compare to an 800 lumen bulb.
The reed switches would be used as inputs to some solid state relays, so they won’t be carrying anymore voltage than in their intended application.
Do you need 800 lumens? A Neopixel ring set to 100% brightness is extremely, intensely bright. If you were to put the ring at driver’s eye level in the middle of the rear garage wall, it would be impossible to miss, even if the sun is coming up behind that garage.
It’s less lumens but all the light is concentrated in one ring, and the individual LEDs have lenses on them.
I really think you should go with this solution because it’s a lot safer. Dealing with live 120v is something you need a lot more components to properly protect yourself, and I fear you want to take shortcuts. You need at least 2 separate metal protective enclosures, they need to be grounded, you need to secure the 14/3 wire, an electrician might use conduit in this location. It’s not cheaper, either.
With my solution, if you use a UL listed 5v power brick - whether it be one that came with the Arduino or a random one you dig out of a drawer - all the danger stops there.
If I’m understanding this correctly, the left switch breaks the circuit when the door is closed. I think I can find normally open held closed limit switch that can be installed.
The right switch has two positions: power to red light or power to green light. I think this is a SPDT ON-ON or DPDT ON-OFF-ON switch. I’m having trouble finding switches of those types that can be actuated by a garage door. I really don’t want to mess with installing and adjusting brackets to trip a switch meant for fingers.
Seems unnecessarily complicated, but I can see it working. One reason I’m suggesting the contactors is that they will handle a lot more current and it seemed like a good idea to provide actual receptacles. And if you use receptacles, you would have 15 A available for whatever lights you want to use.
My parts list:
(2) SPDT spring switches
(2) 12 VDC coil contactors
(1) duplex receptacle
SG box, 22 AWG cable, and some misc. hardware
Spec. the switches according to the power source, 120VAC 3amp for incandescent or, 12VDC 3amp(ish) for LEDs…The $$ will be approximately the same.
All you need is to: install two switches (the hard part), mount the lights, then connect the dots with wire. For ease and simplicity of a safe installation, I’d prefer LEDs. One wall wart wire thru the switches…one wire from the lights back to the wall wart. Done.
BTW: A fuse in the supply side would be a good idea.
No need to overthink this.
SPDT limit switches are about $4 each. A few dollars more if you want something fancy. A 12 VDC coil contactor that handles up to 15A @ 120 VAC is about $25. It’s probably cheaper to get 24 VDC coil contactors, as they are more plentiful.
If you just want to switch 12 or 24 volts for a lower output light, they are around $12 each. These will usually also switch 3A @ 120 VAC.
Your first proximity switch is a merely a single-pole single throw (SPST) that is designed to be normally closed. The second proximity switch is a single-pole DOUBLE throw (SPDT).
Google “proximity switch”…There will be appx. 3000 sites selling what you need, or goto your nearest electrical supply house and they’ll hook you up.
Why not just use the switches that the door already has on the track?
So all lamps are unsafe? I’m just building a lamp with the switch operated by a remote sensor. I’m not hard wiring this into the walls. Perhaps the best cord is not 14/3 romex (4 conductor), but 14/3 extension cord (3 conductor). It is just an ungrounded lamp, so I should be able to use the insulated ground wire as neutral. Off course, using a wire for a purpose different from it’s color can be a problem. The other problem is that 14/3 romex is easy to get at the hardware store, and 14/3 extension cord wire is a specialty part.
As for the brightness, I think there is a big difference between an indicator LED on the far wall, even if easily visible, and a bright light hanging in front of your face. I do like the simplicity of your solution, and I think it is probably cheaper, but all of the soldering really makes it a non-starter for me.
All 120 volt high power circuits are unsafe enough that electrical codes require numerous levels of protection.
Neopixels aren’t indicator LEDs. They produce substantial light in themselves. A sufficiently long strip of them can light a room by itself.
The only soldering you would need to do is to push a wire through a hole, and where the wire is on the other side of the hole, solder it down. You do this by just heating the soldering iron, cleaning the tip, pressing the flat tip against the wire and the metal of the perf circuit board, and waiting a moment.
Then you just touch the solder to the hot wire (not the iron). If you got it hot enough, you will see the solder flow over both the circuit board contact and the wire.
It takes about 10 seconds a wire, and you have 2 switches, so 4 connections there, plus 3 wires for a neopixel, so 6 connections there.
Go watch a video online but this is really, really easy. A little dab of flux from a flux pen will also help but it is not necessary.
There, that few second thing he does? That’s all you do. You just press the tip with one hand and touch the solder with the other. The solder chemistry is what is actually doing 99% of the work.
I must not be finding the same thing, because proximity switches are just leading me back to reed switches, or other similar magnetic actuated sensors. I was thinking that your were talking about a physical switch that the door would move. This is part of my problem. I can imagine the type and shape of switch I want, but I don’t know what to call it, so my searches aren’t returning useful results.
This is where the solid state relays come in. For about $10, they will use as low as 3 volts DC to switch up to 380 volts AC at 25 amps.
This is the question to answer. Your door already knows when it’s opened or closed.
If you insist on your own switching use proximity sensors and magnets. An Arduino is fine for something like this.
Yeah. What echo is missing is that the hardest part of this project so far is not actually soldering or designing a circuit. It’s attaching those damn switches in a way where they will actuate reliably, not vibrate loose when the door runs, etc.
I’m speaking from experience from a small robotics project where I “just” wanted to add a rotary encoder. The mounting system was far more difficult to get to work than the actual electronics or software.
So the thing he wants to do is to find out what voltage the switches use (probably +12V). Tap the switch lines (soldering or crimping). Bring those switch lines to his electronics box. Then either use the switch lines to drive relays or use an isolator board to check the +12V levels safely with an arduino.
Such as this: SparkFun Opto-isolator Breakout - BOB-09118 - SparkFun Electronics
There are no sensors on the track. I have a very odd garage door setup. The screw drive, which is on the wall, does have limit switches to let the opener know when to stop when going each direction. I’m extremely hesitant to tap into those, because they’re already hard wired into the opener. I’d have to cut the wires, and splice in extra connections. If I mess it up, the screw drive will possibly damage itself trying to go too far. That could potentially be a $3-5000 repair.
Actually attaching to the switches will be easiest if they are removed, but they can’t be removed without taking down the entire screw drive assembly. Working on them in place seems much more difficult than screwing a magnet to a wooden door.
Yeah. That. Easy peasy. That’s what I recommend you do, and what TriPolar is voting for as well.
You can easily check, after you make the tap, with a multimeter that the switches still work.
It’s not nearly as hard as you are making it out to be. Also you don’t even need to tap, you can go find where those wires go to a terminal block on the opener, loosen the screw, push another wire in, and tighten the screw back down. Do a quick tug test and you’re good. (maybe twist the wires together and hook them around the screw if you want to be really secure)
I think your biggest problem with this project is your own fear of doing things. It is actually pretty unlikely you’d do 5 grand of damage, as well, I’ve gotten a door replaced and it was about 1.5k for all new everything except for the opener. Moreover, unless the door is on the verge of failure, it’s unlikely the opener will deliver enough torque to actually break it.
I do appreciate everyone’s replies. I don’t want to come off as that kind of person who just argues with good advice, even though I’m doing a lot of that here. The trouble is that I have a very odd door setup, which is almost certainly different than yours, as it is a surprising and odd setup to the garage door professionals that I’ve had come out to work on it. It was a hack job in 1981, or whenever it was installed.
The other problem is my awkward skill set. I’m really bad at soldering. Something that will take somebody with some practice 10 seconds is going to take me 45 minutes, replacement components when I burn the board, and buying the appropriate tip for my long neglected soldering iron. I’m pretty comfortable working with 110 and screw down terminals.
I do prefer the idea of a low voltage solution, but I want something that is like a stoplight, not just an indicator. I’d hang a traffic light from the wall, except those things are huge. I know many low voltage LEDs are extremely bright, but how many $10 LED rings will I have to go through because I destroy them?
It’s probably worth pulling the cover off the opener to see how it looks inside. I’ve already spent $200 replacing the circuit board in there. The $3-5k figure is the cost of replacing two wooden doors with two fiberglass or steel doors and jackshaft lifters. If I permanently kill my opener, that is the repair, because there isn’t a replacement model. Though it looks like there might be screw drive openers on the market again, so it would only be a $400 repair… I don’t think I’ll break the doors, they are pretty sturdy. I might rip a track off the ceiling, though. That would be bad.
Yes, it is my fear of doing things. My fear of destroying $10 components with my non-existent soldering skills. My fear of destroying the garage door opener in a way that requires a capital investment in the house.
Anyway, thanks again for the responses, I do really appreciate them, even if I don’t think they’re right for me.
Yeah, I’m sensing that you don’t want to do any more drilling and soldering than you need to.
I’d still suggest the limit switches. I’d mount them to the track with two screws. They usually have a little roller on a leaf. You can bend the leaf so the roller comes in contact with the edge of the door in the up or down position. This may mean gluing a little block of wood to the door to hit the roller.
Solid state relays would be fine. As an electrical contractor, I always tend to use contactors, but there’s no reason not to use solid state units if you get them.
I do, however, still suggest using a duplex receptacle. You can mount it in a surface box (available at fine handyman stores everywhere) and wire it with a pre-terminated power cable (you know, molded plug and all), Split the receptacles so one is for the green light and one is for the red. You only have to switch the hot conductor. This way, you are not getting into your house wiring at all and you can plug in whatever type of light or device you want to. Shouldn’t need any Romex at all.
Just my suggestions based on your description of the desired results.