We’ve seen BarBots that will automagically pour you a drink, but how about one with RFID? How about one with Facebook integration, so your friends know how much of a lush you are? Wait. Facebook already tells them that. Huh.
[Andy] and [Daniel]’s latest build follows on the heels of a lot of similar cocktail bots; an Arduino controls a few solenoid valves connected to a CO2 supply and a few bottles of liquor and mixers that allow drinks to be dispensed at the push of the button. Where this project gets interesting is its use of RFID and Facebook.
The user interface was coded for Windows 7, with an RFID tag (ostensibly issued to each guest) allowing a unique login that checks an SQL server to see what privileges the user has. The app pulls the user’s Facebook profile photo down and displays it in the corner of the screen, and with the server keeping track of how many drinks (and of what kind) they had, with the right permissions it should be possible to post that info to their wall. Because we all know what you did last night, even if you don’t.
perfect for a sausage fest.
Finally, because there was a desperate need for more drunk status messages in the world.
I like the CO2 idea instead of the moving glass type such as the Inebriator since it doesn’t need servos and as I see it is also easier to build, but all of the CO2 variants I’ve seen have used solenoid valves on the fluid side, so all the drinks flow through the valves. Instead I’d put airvalves on the CO2 feed to the bottles and run tubing right from the bottle to the glass, this way you wouldn’t have to clean the valves since only CO2 passes through them.
Thoughts? I’m planning on building my own version with a raspi and a coin slot so friends can buy drinks at my place (very cheaply of course, I dont want to make a profit, just cover the costs of the drinks)
I think it should work set up either way, and I definitely see the advantages you’re talking about. It seems to me that the current set up (constant CO2 pressure, valve on the liquid side) would probably result in more consistent flow rates though, since the contents of the bottle will be at a known pressure when you open the valve.
When the valve is on the liquid side, the only pressure drop you’re going to have in the system is the result of liquid flowing out. If the valve is on the CO2 side, you’re going to have a pressure drop from pressurizing the outflow side of the line from the valve, represurizing the bottle from whatever pressure it’s at (unknown), and then finally from the liquid flowing out. If dispensing is based on a timer, you’re going to get more consistent shots from the liquid side valve, but if dispensing is based on flow or weight, you should be good to go either way.
This is why we put the valves on the liquid side. We couldn’t maintain repeatable results using a time based approach. Definitely would be better if we didn’t have to clean it out after each use though!
A feature that could be added to this that might be interesting would be getting the RFID drink tag in exchange for your car keys that are only released when requested if your drink record indicates that your BAC is estimated to be below the legal limit to drive.
It might help prevent awkward moments or belligerence directed towards the host from a surly drunk insistent on driving home or prevent them from begging, intimidating, stealing or sweet-talking their keys back from the host (“Hey, it’s not me refusing to give you your keys, it’s the Barduino lock-box!”). I guess that means the key box needs to be sturdy, too.
To be reasonably accurate it would probably require additional data such as sex and body weight, but to be discreet and prevent people from lying about their weight it could be determined by issuing the tag while a guest is standing on a scale (whether they know it or not) and not listing it on the display (or a Facebook page!).
Conversely, an algorithm could be written that gradually alters the alcohol content of the drinks poured for each guest over time so that as the event progresses (and if a guest is drinking a lot as determined by the BAC-estimating algorithm) the drinks get progressively weaker since alcohol has a way of encouraging us to drink a lot more of it than we may have intended, quickly shooting past the “pleasant buzz” stage and ending up at “obnoxious/passed out” (no keys for you!). That would probably have to be an opt-in feature or else people would quit showing up for events and start bad-mouthing your “watered down” drinks.
In any case, while it’d be cool to have an automated drink dispenser for parties or events I personally would never utilize a device that links my recreational pursuits (no matter what they may entail) to a social network page (even though I don’t have one anywhere), but especially if it was FB.
I guess I’m the only one pondering CO2 chilled Everclear. You could expand out with microflavoring agents or mixing agents. MiO-ish stuff meets cold grain alcohol.
http://www.barduino.co.uk/