You’ve probably seen multicolored flashing LEDs embedded into clear plastic cups or coasters before, they’re quite commonly used in fancy restaurants that also feature animatronic characters and a gift shop on the way out. But have you ever wondered about the logistics of maintaining such devices? When the anthropomorphic rodent shuts down for the night, you’re going to want to clean all those blinking doodads; but any opening to connect a charger or insert a battery is just a leak waiting to happen.
[Scott Clandinin] has come up with a solution to this problem that’s equal parts brilliant simplicity and unabashed overengineering. Using wireless charging and supercapacitors, he’s developing an LED coaster that can be hermetically sealed in clear resin.
With no plugs to connect or batteries to change, these coasters can be permanently encapsulated with no ill effects. Granted the supercapacitors will degrade with time and eventually won’t hold a charge for as long, but even the most conservative estimates would have these coasters still partying in a decade.
For his prototype version [Scott] has put together a simple charging base, but we imagine in a full deployment such devices could be charged with induction coils built into a bar or table. While the energy consumption could potentially be a showstopper, we’d love to see a future version that integrates a radio receiver. Then the coasters could double as pagers to let diners know their table is ready.
While this device is obviously much thicker than a traditional coaster, it looks fairly reasonable even at this early stage. We like the concentric design that puts the coil inside the PCB, and wonder if similar cutouts couldn’t be used to get the twin 15F supercapacitors and charging module hunkered down just a few millimeters more. The 2019 Hackaday Prize is all about evolving an idea into a design suitable for production, and those are the sort of incremental improvements that the judges will certainly be keeping an eye out for.
How long will it be able to operate inbetween charges?
The write up in his article says 4 hours.
That’s really optimistic. An average current of 2,5mA with 100% conversion efficiency from the caps.
I wrote up 4 hours without realizing it didn’t have a full charge actually. Letting it charge for a few hours will get me between 5-6 hours, though after the 5th hour it begins to get noticeably dimmer before dying off.
You’re getting 4 hours with 30F? I’ve got the calculations for 7F on one led at 9 minutes? This article has one standard 20mV 3.3 LED on 40F at 1 hour. The write up said you got 4 hours on what looks like 30F? Am I missing something. Proof is in the execution, and it worked, but I’m confused and doing a pre-build now. I’ll have my caps and charger in a week so I can report back. https://www.quora.com/How-long-would-a-50-farad-capacitor-light-up-an-LED
Hey, this is actually 60F to get 5-6 hours. The article is a bit incorrect, and my documentation probably a bit lacking. I found the length of time I got with 30F was pretty low so I doubled up on the caps which helped a lot.
What you linked doesn’t totally apply in this situation because I have no boost converter that keeps a constant voltage and current draw. Without this it actually helps the lifespan I think. And note that at full charge the caps hold 2.8V, and the LED forward voltage is 2V. With 100ohm resistor this puts my peak current draw at 8mA (ignoring other minor current drawing devices). And this 8mA is always dropping because my source voltage is slowly decreasing. By the time the cap voltage is at 2.4V, it’s only drawing 4mA. And the less current is flowing the slower it drains itself, which drags out the time it’s on for until the caps reach 2V and the LEDS won’t turn on.
It’s very hard (for me at least) to put an equation to this with the basic information I know about capacitors. And there is absolutely a lot of optimizing that could be done to improve the lifecycle. Let me know how it goes, I’d be curious how your results turn out.
Thanks Man. 60f makes way more sense. I ordered way too low (1F) having only done cursory research on this. I saw a video about a joule thief extending the life of a 22f led circuit to 20 minutes. It was a large Led, so I’d like to imagine maybe I could get a 40f cap up to 40 minutes or an hour. I’ll do a post and link you to it after I build it.
Very cool project.
A big hurdle will be bringing down costs. Those supercaps are spendy and as the project page mentions, more power brings more cost as well.
I’m sure the wheels are already churning on this, but: A way to mass-charge these would likely be needed in a business setting. A business probably would not mind leaving something on the charger overnight if that’s what it takes, but they would definitely mind using an employees time to run these to the charger one at a time.
Since it’s coil based I’d think a ferrite rod on a stand would be perfect, just have a hole in the coasters and slip them onto the rod each evening.
Would be great, but people steal these a lot.. they would have to be suuuper cheap
Mount everything in holes in the PCB. Use more capacitors of a thinner style. Make the charging coil part of the PCB.
“With no plugs to connect or batteries to change, these coasters can be permanently encapsulated with no ill effects.”
Well, except for the primary purpose of the coaster:
Absorbs the condensation from a cold beverage.
I’ve seen plenty of hard plastic coasters. I think the whole absorption thing is kind of secondary at this point to looking cool and perhaps protecting the surface/glass from smacking into each other too hard.
For me a coaster is not for absorbing anything, but to protect the surface from the heat of a hot beverage.
Ask any woman and she’ll tell you they are to avoid rings on her table
It basically has to use BLE to prevent theft, and I’d personally halve the number of LEDs and required capacitance to keep it as cheap as possible since you can just build in wireless chargers for every table and in a mass production sense will be cheaper overhead instead of overbuilding the sensors
So you’re going to run a power cable to each table?
Or with a simple RFID. Just need a reader by the exit(s).
RFID range is worthless (1foot?), and then the charge coil would have to be dual purposed as an antenna.
There are RFID systems which operate over entire warehouses.
The charged coaster can be dispensed with the drinks. You just need the charger by the bar where the coaster and used glasses are collected and cleaned.
For those worried, theft is already considered in the price of food/drinks. Same as broken glasses, etc.
Additional thefts means additional price increases to cover the losses.
Water down the drinks with lot of ice.
More ice isn’t watered down. A drink consists of a few ingredients, zero sum to the volume of the container: spirit, mixer, ice. Spirit is the constant, since it is the highest cost and measured pour. The remainder is mixer or ice. Add more ice, less room for mixer.
Asking for a Long Island Ice Tea with less ice means the same alcohol, but more sours and cola to fill the glass: a much weaker drink.
Wouldn’t it be nice to integrate a temp sensor so you know that is is still cool. Can also add rfid to the glass buttons and the house can track the movement and other logistics of the drinks.
“fancy restaurants that also feature animatronic characters and a gift shop on the way out”
In my opinion a fancy restaurant has a Michelin star, not a carnival… ;-)
I would have used conductive pads, on the top and bottom so they can be Daisy chained on one charger