So far these are the components we have:
Ten metres of 10mm silicon piping (food grade1) and fifteen solenoid valves.
The current plan is to have all the spirits and mixers in upside down bottles above a bench, tubes will come down from these bottles, pass through the valves then all come together in some form of tap above where you place your glass. The valves will be controlled by a Raspberry Pi with an Android tablet hooked up to allow selection of which drink you want.
It will of course look much fancier than this sketch, personally I’m thinking some nice dark stained wood with brass fixtures, some form of laser cut drip tray grate with an icon representing the limitless potential of this device, and of course only top shelf liquors allowed.
Designing it this way means there is very little mechanical design that needs doing, really the most complicated part will be ensuring the bottles are held securely while being easy to replace as they get empty.2 Which is all for the best, while we have 3 mechatronics engineers on the project only one has done any hardware work since leaving uni.
The biggest disadvantage of this is the limitation of only working with liquids. It would be amazing if we could have ice and garnishes added automatically, but (at least for the first iteration) we are going to require the user to manually add these.
Since we won’t have much mechanical design to do we can instead spend longer on the software side. If we get a really nice client designed for the Android tablet, complete with all the background info about cocktails and the ability for the user to adjust the strength on the fly, then this will be re-usable if (once) we build a version 2 with more capabilities.