6 ms·

Nice. I like the lights. You can also slap a reed switch (e.g. a normal door open/close sensor) near your dumb doorbell's magnetic coils that do the ringer and

Nice. I like the lights. You can also slap a reed switch (e.g. a normal door open/close sensor) near your dumb doorbell's magnetic coils that do the ringer and have it send the info to home assistant. I hooked a $0.50 reed switch up to my dumb doorbell and ran it to a digital IO port on a ESP that's powered from the same power source as the doorbell coils. When it senses a doorbell press, it sends me an email snapshot from my local-only front door cam and plays a recording of the doorbell chime on my upstairs stereo (where otherwise I can't really hear the chime). Very convenient and fun. Originally, I tried monitoring voltage on the coils with the analog input, but it was way too unreliable. The simpler reed switch method of detecting current is rock solid.

rainbowzootsuit 3y ago
Reed switches are great like this. I have used them in a similar way stuck to the top case of relays to detect when the coil is energized without probing directly into the control or load circuits.
adolph 3y ago
> same power source as the doorbell coils Beautiful thinking. I wonder what other obscure sources for voltage in a house could be used? The HVAC head unit, analog telephone, ... The POTS [(Plain Old Telephone Service)] phone line, with all phones on-hook, should measure around 48 volts DC. This drops down to the 3 to 9 volt range when a telephone on the line goes off-hook. An off-hook telephone typically draws about 20 milliamps of DC current to operate, at a DC resistance around 180 ohms. The remaining voltage drop occurs over the copper wire path and over the telephone company circuits where there is usually 200 to 400 ohms of series resistance to protect from short circuits and decouple the audiocircuits. https://www.jkaudio.com/article_03.htm https://www.jkaudio.com/article_03.htm
- westurner 3y ago
  If the POTS cables daisy-chained between the telephone jacks after the telco's box are 4-pair Ethernet, there are 8 wires between each receptacle/port less 2 wires for each phone line. If only one phone line has ever been connected and there are 4pair (Ethernet,) cables between the phone jacks, there are 3 spare wire pairs that could be connected in a local private closed loop in order to carry (modulated) DC voltage for other applications; though you'd need to put like a thermal sticker label after the box and behind/on all the phone jacks to warn about the voltage (which is a presumed risk when handling 4pair cabling as an installer, as a (demo) contractor, as a future occupant) Gigabit Ethernet uses all 4 pairs for data, so Gigabit PoE and faster must have data+power on one or more pairs. From https://en.m.wikipedia.org/wiki/Power_over_Ethernet https://en.m.wikipedia.org/wiki/Power_over_Ethernet : > In addition to standardizing existing practice for spare-pair (Alternative B), common-mode data pair power (Alternative A) and 4-pair transmission (4PPoE), the IEEE PoE standards provide for signaling between the power sourcing equipment (PSE) and powered device (PD). This signaling allows the presence of a conformant device to be detected by the power source, and allows the device and source to negotiate the amount of power required or available while avoiding damage to non-compatible devices.
rcarmo 3y ago
I did exactly the same and use an Aqara Zigbee contact sensor (which has a reed switch inside) inserted into the actual doorbell (which has a convenient flat spot above the coil). Works great.
- acidburnNSA 3y ago
  Yes, this is the "consumer grade" version of this that doesn't require any ESP programming or whatever. Easy and simple, integrates with other hubs just fine. Brilliant.
barbazoo 3y ago
> You can also slap a reed switch (e.g. a normal door open/close sensor) near your dumb doorbell's magnetic coils > The simpler reed switch method of detecting current is rock solid. That's really clever! I'm gonna try that. Thanks for sharing.
fullstop 3y ago
Do you have problems with reed switches getting stuck? I've had fantastic luck with hall sensors for this purpose, but maybe I was just looking at low quality reed switches.
- acidburnNSA 3y ago
  Never, at least that I've noticed. I guess it'd be hard to tell if I missed any doorbell rings technically.
  - fullstop 3y ago
    No, in the cases that I'm talking about it gets stuck in one position. If you google "reed switch stuck" you'll see what I'm talking about. I figured that a hall sensor has no moving parts so it should last essentially forever.
- squarefoot 3y ago
  When contacts, including reed switches, normal switches, relays, etc, get stuck, it usually is because of sparks due to voltage spikes that weld the contacts together. this happens when contacts are used with high currents/voltages and/or inductive loads, and can be avoided using so called snubber networks. https://en.wikipedia.org/wiki/Snubber https://en.wikipedia.org/wiki/Snubber
  - fullstop 3y ago
    Ah, that's interesting. So for OP's purpose it will likely run indefinitely since there's essentially zero load.

rainbowzootsuit 3y ago

Reed switches are great like this. I have used them in a similar way stuck to the top case of relays to detect when the coil is energized without probing directly into the control or load circuits.

adolph 3y ago

> same power source as the doorbell coils Beautiful thinking. I wonder what other obscure sources for voltage in a house could be used? The HVAC head unit, analog telephone, ... The POTS [(Plain Old Telephone Service)] phone line, with all phones on-hook, should measure around 48 volts DC. This drops down to the 3 to 9 volt range when a telephone on the line goes off-hook. An off-hook telephone typically draws about 20 milliamps of DC current to operate, at a DC resistance around 180 ohms. The remaining voltage drop occurs over the copper wire path and over the telephone company circuits where there is usually 200 to 400 ohms of series resistance to protect from short circuits and decouple the audiocircuits. https://www.jkaudio.com/article_03.htm https://www.jkaudio.com/article_03.htm

westurner 3y ago
If the POTS cables daisy-chained between the telephone jacks after the telco's box are 4-pair Ethernet, there are 8 wires between each receptacle/port less 2 wires for each phone line. If only one phone line has ever been connected and there are 4pair (Ethernet,) cables between the phone jacks, there are 3 spare wire pairs that could be connected in a local private closed loop in order to carry (modulated) DC voltage for other applications; though you'd need to put like a thermal sticker label after the box and behind/on all the phone jacks to warn about the voltage (which is a presumed risk when handling 4pair cabling as an installer, as a (demo) contractor, as a future occupant) Gigabit Ethernet uses all 4 pairs for data, so Gigabit PoE and faster must have data+power on one or more pairs. From https://en.m.wikipedia.org/wiki/Power_over_Ethernet https://en.m.wikipedia.org/wiki/Power_over_Ethernet : > In addition to standardizing existing practice for spare-pair (Alternative B), common-mode data pair power (Alternative A) and 4-pair transmission (4PPoE), the IEEE PoE standards provide for signaling between the power sourcing equipment (PSE) and powered device (PD). This signaling allows the presence of a conformant device to be detected by the power source, and allows the device and source to negotiate the amount of power required or available while avoiding damage to non-compatible devices.

rcarmo 3y ago

I did exactly the same and use an Aqara Zigbee contact sensor (which has a reed switch inside) inserted into the actual doorbell (which has a convenient flat spot above the coil). Works great.

acidburnNSA 3y ago
Yes, this is the "consumer grade" version of this that doesn't require any ESP programming or whatever. Easy and simple, integrates with other hubs just fine. Brilliant.

barbazoo 3y ago

> You can also slap a reed switch (e.g. a normal door open/close sensor) near your dumb doorbell's magnetic coils > The simpler reed switch method of detecting current is rock solid. That's really clever! I'm gonna try that. Thanks for sharing.

fullstop 3y ago

Do you have problems with reed switches getting stuck? I've had fantastic luck with hall sensors for this purpose, but maybe I was just looking at low quality reed switches.

acidburnNSA 3y ago
Never, at least that I've noticed. I guess it'd be hard to tell if I missed any doorbell rings technically.
- fullstop 3y ago
  No, in the cases that I'm talking about it gets stuck in one position. If you google "reed switch stuck" you'll see what I'm talking about. I figured that a hall sensor has no moving parts so it should last essentially forever.
squarefoot 3y ago
When contacts, including reed switches, normal switches, relays, etc, get stuck, it usually is because of sparks due to voltage spikes that weld the contacts together. this happens when contacts are used with high currents/voltages and/or inductive loads, and can be avoided using so called snubber networks. https://en.wikipedia.org/wiki/Snubber https://en.wikipedia.org/wiki/Snubber
- fullstop 3y ago
  Ah, that's interesting. So for OP's purpose it will likely run indefinitely since there's essentially zero load.