Hacker News new | past | comments | ask | show | jobs | submit login
Innovative wireless light switch could cut house wiring costs in half (ualberta.ca)
47 points by thunderbong 11 months ago | hide | past | favorite | 72 comments



Don't think it's particularly innovative really - a lot of electricians in the UK are now installing the Quinetic - https://www.quinetic.co.uk/ - system and it works really well in terms of wireless switches.

Mostly it's not to save the cost of the actual cable - but to save the install cost on an old brick building, where installing a new switch requires fairly extensive work to the house and brickwork.


I haven't seen Quinetic before. This is ever smarter than the solution in the article. Just use the kinetic energy of the press, rather than collecting ambient energy. Seems more reliable. OTOH limits what kind of switches can be used.


What sort of force is required to activate them, is touch enough? Do they make noise to click?


yeah exactly, I was like "what's new here?"

I already have these in my house


I dunno, I believe in hardwiring things where you can. Stuff in an immobile structure like a house can and should remain actual wire.


Right? Light switches work when the microwave is on, when the light is a long way from the switch, or whatever inevitable gremlins find their way in. When you're talking about something that needs to just work Keep It Simple, Stupid!

A few bucks in house wiring is nothing over the lifetime of the structure. Do it right.


Where are you getting the "few bucks"? Just the cost of the wire itself? Electricians cost a couple hundred bucks an hour. If the answer to that is "just learn how to wire your house yourself" then I'd say that is not a feasible solution for most people. If the answer is "no, I didn't mean rewire an existing house, I meant running wire in a new construction" then I'd say that is not a feasible solution for most people either.


I think it's a few bucks, relative to the cost of a house. If you're paying for an electrician, you probably own the house. Compared to the cost of a house, a one-time visit by an electrician is a small amount.

And running wire in a house as it's being constructed is going to be far less expensive than doing it after the fact. I'd much rather add, for e.g., the ability to charge a car in the garage while the house is getting built, than after.


> Compared to the cost of a house, a one-time visit by an electrician is a small amount

Okay but can you fold the electrician's bill into your home loan ? No ? I thought not. It comes out of your current cash flow. Which might be hand-to-mouth for many people.


Yes. Construction loans include electrical.


Sure, the electrical that you know when you plan the house. But additional work later on too ?


Yes, you can get loans for home renovation if you so choose. Most people would just get in the attic and run the wire though, it's not actually that much work.


You need a power line to the switch and a separate power line to the fixture.

If this is rework or an addition, it's exactly as much work as just running a line between the switch and the fixture.


Perhaps. Both are more than most people want to do. I infer this from the existence of a professional class of electricians.


You need continuous power anyway to allow for app or remote-based control.

It's really nice to be able to use a remote to turn lights on and off from my bed or desk, or use voice control when my hands are full.

Plus, I can have dimmable lights even in locations with a three way switch like a hallway.

Color temperature control is not really possible with hard wired controls either. I'm actually considering replacing all my switches with in-wall remotes.


Those are always nice in the first few years, but then a decade or two and three owners later, the switch is now hardwired to a height that was appropriate for shorter generations, controls some track lighting that isn't commonly used, doesn't support dimmable LEDs, or just plain isn't where you want it, next to the new dining table...

They become the electrical equivalent of tech debt. A few dollars to install, thousands to move or change.


Aside from if you want to physically move a light switch, all the rest of the things you mention are cheap and easy to change.


Not as easy as a wireless switch


I'm actually amazed that we're expected to damage and then repair the walls and ceilings just to move switches, lights and outlets around. Or connect the lights to a 4-wire RGB controller where they were previously connected with just 2 wires. Or run ethernet cables.

Some of this is solved in commercial spaces using modular dropped ceilings, but I never saw that in residential spaces and never anywhere for walls.


I mean, no matter what your walls are made of, you have to physically cut a hole for the switch or outlet to sit in.

Then it becomes buying a whole new wall panel instead of just patching the drywall.

Beyond that, you have the problem of insulation. If your wall isn't a contiguous piece of plaster, it will leak air. Then you have to contend with your insulation material releasing fiberglass fragments or whatever into the room. You also have to deal with some way of fixturing the wall panels to the support structure, or they'll all rattle any time you open a door. You also have to just hope that someone is still making the exact same panels decades after the fact. Most rework is done in houses that are several decades old. I don't think that many building materials suppliers that built my house in the 1920s still exist.

Any sort of modular system will always be more expensive than drywall and plaster. If you want to address all of the problems and make it as good as drywall, no one could afford it.

The electrical system is just as integral to the structure as plumbing or ventilation is. You are not going to be replacing things often, and the effort of patching drywall every decade or so is not that much. If, for some reason, you want to move your outlets every six months, we already have solutions for that. You can put conduit and boxes on the outside of the wall.

All in all, it's an awful lot of trouble and expense to make it very slightly easier to do work you only do every decade or so.


> I mean, no matter what your walls are made of, you have to physically cut a hole for the switch or outlet to sit in.

I don't think so - just have some spare panels with pre-made holes (perhaps off center, so the holes can be positioned more precisely by rotating the panel).

> Then you have to contend with your insulation material releasing fiberglass fragments or whatever into the room.

Hm.. Ok, where I live it's usually going to be concrete on the other side, not insulation material.

> You are not going to be replacing things often

I think the popularity of wirelessly controlled lights and speakers says otherwise. Not to mention various decorations (especially if they require electricity - wall hanged bookshelf with decorative lights, paintings with a light above it, or maybe you simply want to move TV to the other wall). And this would allow renters to customize their apartments much more easily.

It might be that you (possibly also most people?) just change their mind about these kinds of stuff less often.

I grant you that it might end up being pretty expensive because of the mounting structure.


Is there an argument to be made that this could be retro-fitted in older homes to reduce fire risk? Or does fire risk mostly come from outlet wiring?


These days it's about finding the labor and the project time. All those extra runs of wire add up to a lot of expensive time. If you're doing a remodel it's also a lot of wall that doesn't need to get ripped up.


I agree, but why does the wire to the switch have to be 14/2 high voltage? Would it be more economical to run smaller gauge wire to a controller near the receptacle/fixture? Or possibly to a central controller?


> why does the wire to the switch have to be 14/2 high voltage?

That way, it's easy to extend the circuit later if you make modifications to the structure.


More interesting would be if you told us why you think that is.


Why invite complexity when the situation doesn't call for it? Wires are simple, robust, easy to work with and easily understood.


Sticking a wireless controller onto a wall is a much simpler operation than cutting a hole in the drywall, fishing wire, patching the drywall, and repainting.


Unless you want to move a switch, or add a new one, or the route is blocked by a post, or you need to RE existing wiring


It feels like this is already a mostly-solved problem.

The wireless switches/sensors/bulbs are already handled by various things in the Zigbee ecosystem. RF energy harvesting is already a thing[0]. I've even got a Zigbee switch[1] that uses the energy from pushing the button to power its transmitter. Turning lights on and off based on occupancy is also a solved problem, as is HVAC zoning.

The only innovation here is to combine the various pieces into one package. Hopefully it uses an existing standard like Zigbee, ZWave, or Thread/Matter instead of inventing something new and incompatible.

I am 100% on board with anything that eliminates the need to change batteries in all of these devices.

[0] https://www.mouser.ca/new/e-peas/e-peas-aem30940-rf-energy-h...

[1] https://meethue.co/products/philips-hue-tap/


> The only innovation here is to combine the various pieces into one package.

Isn't this how any and all innovations work? (At least in the consumer product area)


It is, but honestly this feels pretty weak.

There are already a bunch of Zigbee energy harvesting switches on the market that work well. There's the Hue Tap that I mentioned, plus other commenters have pointed out other ones.

If you replace the "one or two RF power transmitters to power up all switches inside the house" with a Zigbee gateway you've got a packaged solution that you can buy right now.

The rest of the article talks about sensors, "seamless automation", etc and feels like future thinking (i.e., they don't have a solution here but they're working on it), but also feels like more already-solved problems.

For example, lighting driven by occupancy is something you can easily do by adding a Zigbee motion sensor to the system.

HVAC zoning is also a thing, but a lot more complicated as it requires dampers and needs to be designed as a complete system. You can't simply open the dampers for just the occupied rooms, you have to keep in mind that the furnace needs to maintain a minimum amount of airflow so that it doesn't overheat. You could also go with a mini-split heat-pump system where each major room (e.g., bedrooms and living rooms) will basically have its own HVAC system. I doubt that the costs of a zoned HVAC system would ever pay for itself in energy savings vs, e.g., an Ecobee thermostat plus sensors (another off-the-shelf solution that you can buy right now).


Integration is not innovation.


I’m not convinced that the RF harvesting is that helpful. The excellent Lutron Pico costs about $30 (and surely much less to produce), is extremely reliable, and lasts for quite a few years on a very inexpensive coin cell.

Also:

> Imagine heating a 3,000-square-foot house, but you’re only using a room at any given time. If your house has wirelessly controlled vents, you can close the vents outside that room, or maybe a couple of bedrooms,

The problem with a damper-based zoning system (which this is) is that the system needs to be fairly carefully designed to work well, that there are lots of moving parts, and that failures are common and suck. By the time you’re designing a system to work well with zoned dampers, you might as well use wires. The actuators take non-negligible power when moving, too.

I suppose one might be able to retrofit something with wireless controls, but this is IMO a terrible idea for new construction.

(Also… ductless systems are inherently zoned and are usually considerably more efficient.)


I have over a dozen picos in my house and haven’t replaced any batteries yet (5+ years now).

I really wish they’d make a Pico in the Maestro shape though.


I love the Hue switch, but the size is just terrible. If this thing would fit in Decorator Plates i'd have replaced every switch in my house with them. A switch that looks and feels like a built-in but is held on by magnets and you can take with you? That's a win.

There used to be a switch made by Lutron that's close. It's held in the wall plate with a little tab and it works with Hue, but they discontinued it a few years ago, and it's exceptionally hard to find them now.


I found some on/off switches on Amazon that appears to use the motion of your hand pressing the switch to generate the power to transmit. It makes a fairly loud CLICK when you press it, so I assume it's some sort of piezoelectric mechanism?

It is basically this (different brand but identical housing): https://www.amazon.com/Self-Powered-Wireless-Receiver-Linpte...


I had a Philips Hue switch that works like that. It worked perfectly but it’s not aesthetic (chunky, clicky, and oddly shaped), so it’s for bachelors only. Lol


I don't understand whats novel? I already have Zigbee energy harvesting switches in my house from Jung [1]:

> JUNG wall-mounted transmitters for Philips Hue work according to the principle of “energy harvesting”. As only small amounts of energy are needed to transmit the wireless signals, the kinetic energy arising from the switching action is sufficient to operate the transmitter without batteries and thus virtually maintenance-free.

[1] https://www.jung.de/en/10245/products/technology/transmitter...


Betting it’s a module for EnOcean in that, they’ve been doing energy harvesting switches for over twenty years now.

I think the new part of this proposed system is that power is provided wirelessly so each switch can be cheaper than the fancy mechanically powered transmitters. I do wonder how much energy it ends up wasting though.


I dont understand why the little solar panels that were included on every calculator from my childhood aren't used more broadly. I would imagine a little solar strip combined with a coin cell could extend the life of the coin cell considerably in applications like an AirTag, especially if premium functionality is only offered when sola power is available (like software updates or LED status displays, etc. seems like there should be a world of potential combining lithium cells, combined "fire and ice" cores, and little solar flaps.

I guess obviously not or we would see a lot more of it?


Your comment got me curious, so I did some googling. The answer is pretty straightforward: those calculators use ridiculously little power. They have bytes of memory, use the simplest LCD displays, and are idle essentially 100% of the time. Moreover, the assumption is that you're not working in the dark. The solar strip is essentially exactly the size needed to run the calculator: as soon as the light goes away (or even gets dim) the device powers off. For a light switch, that's kind of exactly what you don't want. A cross country flight in the cargo hold of an airplane would be enough to pretty much drain a solar air tag: a thief would just need to put your luggage in a black garbage bag to render the air tag useless.

I suspect you'd need a panel that's a few inches large for it to charge faster than a device like an air tag could draw power. But again, that assumes it'll routinely see some fairly bright light.


Right but that's missing the key point of combining the solar cell with a rechargeable coin cell, where the lithium battery is intended to be capable of powering the device for several months on end. If the coin cell provides 4 months of operation from some device, maybe a coin + solar could provide indefinite service in the right scenarios?


It would not, because the solar strip offers less power than the device consumes—by a lot. The cell would not really ever charge. You'd need a solar strip that's pretty darn large (by my rough math, 1-2 index cards in area, at least). At that point, how is it a useful form factor? And again, that's assuming the best possible lighting, which is not the case for either a light switch or an air tag.

I suppose you could use a little solar strip to make a button battery last longer? But you're talking about a difference of like 100 days of life versus 102 days of life.


The remote control that came with our Samsung TV has a small solar panel on the back. You just leave it face down when not in use and then never seem to have to charge it via the USB.


If the battery doesn't die, you won't buy another air tag. :-(


Air tags have replaceable batteries


I was about to say something like "oh those smart devices, what if the power goes down?" then caught myself.


Hehe.

Seriously, though, with smart lights you probably do actually want to consider what happens when the power goes down, or more precisely when the power comes back on.

With a dumb light controlled by a dumb switch when power comes on the light goes back to the state it was in when power failed (unless you have flipped the switch while the power was out).

With smart bulbs some of them come at full brightness when power comes back (and white if they are color bulbs), regardless of what on/off, brightness, or color state that were in when power failed.

This can be irritating.

Some brands let you set the default power on action, ideally letting you chose between off, on in whatever state they were in before lower was lost, or on in a specific state you specify.


Hue is configurable, which is what you want if some of the bulbs are hard switched (so you want them on when power is restored) or soft switched (off or last state on power restored).

Power outages are rare, though; I've not bothered configuring the soft-switched bulbs.


I have a wireless battery-less doorbell. You push the button and the energy of your press is enough to make a short burst of RF enough for the receiver to "hear" it. Throw a few DIN switches on the back to customize the frequency and this simple system could work to trigger anything wirelessly.

Edit: They do exist: https://www.iebelong.com/u2-us-standard-kinetic-light-switch...


Love anything that can accomplish the same goal with less material. I originally thought he was going to store the mechanical energy from turning it on and off, similar to those cranked flashlight. Rf makes way more sense!


This is nothing new, this one has been on the market since 2007 https://www.adhocelectronics.com/Remodel-Press-Release/ though the new one seems more unnecessarily complicated. Just get the energy from the switch motion...


This going to save maybe 10-15’ of 14/2 NM cable per light switch. Maybe 20-30’ of 14/3 if it’s a multi-way switch.

14/2 is about $0.45/ft and 14/3 is about $0.75/ft. A single pole switch is about $2.50-5.00.

This will cost more, assuming the wireless switch costs more than $15.


Would it be feasible to just use low-voltage wire given that everything is LED now?

Also, for places where I don't have a wire but would like to add a recessed light, I wish I could find a battery powered one. I wouldn't mind getting on a ladder and plugging in a USB-C charger once or twice a year if it means I don't need to hire an electrician and fish wires around


Commercial lighting controls do tend to use low-voltage cable for switching, dimming, and occupancy sensing power/signaling, I am much less familiar with the residential world.


If this becomes mainstream, then the economies of scale should bring down the cost significantly. And cost aside, it would be safer and more customisable- you can install the switch anywhere and even change the location easily, don’t need a conduit with wires behind it.


Sure, but it’s compatible with smart devices without any jerry-rigging for backward compatibility, not to mention labor savings. You can probably use the transmitter to power other low energy IOT devices too.


You need to also consider labor cost of installing that cable and switch. That's usually more than the cost of the materials.


Romex is around 1.5-2 hrs per hundred feet and a single pole switch takes around 5 minutes. Even at $120/hr it’s around $30-40 of labor.

These wireless switches need relays to work, so factor that in. And then factor in the time it takes your $120/hr electrician to figure out how the wireless switches work.

I would use a $30 Lutron Pico with a button cell and RF communication over this silly device that generates its own power.

But what do I know, I only sell electrical services for a living.


Residential retrofits: Older Northeast US houses have ungrounded old cloth wiring, no cable protection, and no overhead light switches. While an electrician might be willing to touch the wiring, the walls simply cannot be cut cleanly, demoing them is a nightmare, and repair needs a skilled plasterer or a full demo and rebuild which is not all that much more effort. Assuming the wireless solution can plug in to a bulb socket or is simple to retrofit to a fixture


Will you rewire my house? I'm pretty sure I can't get an electrician to show up for less than a few hundred! Plus touching the panel I think will force us to bring the entire thing up to code (60s construction).


For new construction. If it's a retrofit then the cost of fishing wire and cutting/repairing drywall has to be factored too.


Honestly an ideal application for wireless power transmission. The power usage of the device is low, and there is typically a substantial delay between uses, so a very low power RF source could still be sufficient to keep an internal battery charged. Position is static so the RF source could be highly directional for improved efficiency.

That being said, if you have an RF source, why not have it be the transmitter? Have a passive antenna in the switch that activates when the switch or sensor is active. The RF source detects the returned signal and then communicates with the powered device. Hell a clever engineer could probably get this to work with standard wifi.


I wish the article included a diagram of before and after wiring using this solution.


Makes a lot of sense to not waste heavy copper carrying one bit on information


Exactly, and the additional safety concern too.

As others have said, there are already wireless switches, some using battery-free ultrawideband.

But even if you prefer wires, it makes no sense to have the actual load-carrying wires routed to the switch - have a separate two-wire switch/control network.

Or have it all run on single-pair ethernet - which can carry enough power for lighting too, with PoE/PoDL.


RF harvesting seems like a terrible way of getting power into the switch. Photoelectric aside, why not just harvest the energy of someone's finger actuatinf the switch?


RunLessWire switches work this way https://runlesswire.com/


Shorter lifespan as your vital essence is siphoned away


Very cool. Ambient RF energy harvesting powers the switch, and looks like home sensors too.


Cool but... where is the relay/electronics/power supply that control the light itself?

Most systems like that are home run back at the house electrical panel. You don't save any copper at all compared to the traditional switch. In fact, you use more of it!

edit: In case I wasn't clear. The relay is back at the house electrical panel. So now every light requires a full copper run. This is because the relay has to be serviceable. An alternative would be to have special light fixtures with an accessible relay. This way the light electrical supply can hop from light to light.




Consider applying for YC's W25 batch! Applications are open till Nov 12.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: