Barobot Serves Cocktails While Using Open Design The Right Way


Oh for the day when we can stop repeatedly looking up our favorite drink recipes on Wikipedia. Those may be just around the corner and you’ll have your choice of single-click delivery or toiling away in the workshop for a scratch build. That’s because Barobot is satisfying both the consumer market and our thirst for open hardware goodness. They’re running a Kickstarter but to our delight, the software and mechanical design files are already posted. Before you dig into the design files there’s a really good look at the constituent parts in the assembly manual (PDF) — that’s a lot of pieces! — and a tiny bit on the tech-stuff page.

This remind us of the Drinkmo we saw earlier in the year. That one cames complete with the high-pitched whine of stepper motors. We didn’t get to hear Barobot’s ambient noise in the promo vid after the break. But one place this desing really shines is a swiveling caddy that allows for a double-row of bottles in a similar footprint. One thing we’d be interesting in finding out is the cleaning procedure. If anyone know what goes into cleaning something like this let us know in the comments.

If you were wondering… that favorite drink recipe.

[via TechCrunch]

30 thoughts on “Barobot Serves Cocktails While Using Open Design The Right Way

      1. Rinse it out with soap and water and remount the bottles? It doesn’t seem to have an integrated cleaning feature but you only really need to clean it when you change over raw materials?

      2. Hi Mike, It’s awesome you wrote about Barobot. I never thought my project would hit Hackaday.
        Yes, rinsing the dispensers in warm water with soap is perfectly fine even for grenadine as long as you do it next day after usage. It’s just like dishwashing after the party. Of course dispensers with alcohol bottles can stay as long as you like.
        If you forget next day cleaning (which happened to us more than once during testing) you need to use more aggresive cleaning agents. We used dedicated dispenser cleaner and food safe chlorine cleaners with 100% success.

        1. I am a former bartender, barista, and current chemist. I see why people are wary of the cleaning aspect, but it does not look to be that big of a deal in reality. 99% of cocktail ingredients are VERY water soluble. A pass of hot water should do the trick even if left out for a while. Sanitizing can be done with ammonium quat salts (bar tablets) which are already rated as food safe without rinsing. I would recommend against the chlorine based stuff as bleach can embrittle some plastics in the long run and leech reactive compounds (IIRC from my research days).

          The one exception is milk and cream. Milk proteins can build up and also calcify on containers, leaving a grime similar to burnt egg. Milk fats and protien residue will also go rancid very fast unlike grenadine or Roses lime juice, which is both hygroscopic and acidic–inhibiting bacterial growth. I would dedicate a line to milk from how badly some of our stuff got gunked by cream and Baileys after the EtOH evaporated out.

          I also noticed terpenes from gin will leech into plastic a little. No big deal though as I don’t think it does anything besides smell funny.

          Also, just a suggestion, but have you thought about replacing juices with juice syrups? They are cheap, and an in-line mixer is 3-D printable and cheap–as long as you don’t want to carbonate the water. This would seriously reduce the footprint of the non-alcoholic stuff without raising the cost.

          Nice, elegant design btw. Much better than using a dozen solenoid valves/flow counters. I looked into building a bartender a long time ago and could not work around that problem.

  1. Are those dispensers COTS? I can find any details on them. I’ve looked at similar parts a fair bit, and I’ve found peristaltic pumps and similar to be pretty expensive compared to just doing pneumatic switching.

    1. The BOM only lists mechanical, electronics, wiring, bolts, nuts and washers. It does not appear to list any pumps. That said, it seems to not use pumps at all? It seems to use gravity and magnets to close tubes as best as I can gather.

        1. Ah, I see in the video. I guess they have some kind of actuator that moves with the glass to actuate the passive release valve. That’s an interesting approach, exchanging cost for design complexity some. I think I still prefer the approach previously shown here where you use cheap pneumatic valves to route pressurized air into bottles and then just run all the outlet tubes together.

          1. But then you need to have a source of oil-free pressurized air (and suitable for consumption in general), so not just any compressor won’t do?

  2. I believe those are peristaltic pumps. This would mean that the liquid doesn’t contact the components of the pump itself, just the tubing. So if flushing water through it doesn’t clean well enough you can pretty simply replace those tubes.

    1. Hi, i’m Hania from Barobot team. We don’t use peristaltic pumps at all (they can’t mix carbonated stuff). Instead we used upside-down noozes and we make use of gravity :) Magnets and weight sensors are in case Barobot won’t pour anything unless a glass is under dispenser. If you have any questions we’ll be happy to answer on our forum

  3. Every time i see one of these mixing robots its always shown with the machine pouring stuff into a glass and a happy person taking there “completed” drink off the end of the line, completely glossing over the fact that a massive number of cocktails require more than just dumping some ingredients into a serving glass.

    Maybe im a cocktail snob, but if I got “…that favorite recipe” given to me just dumped over ice, I would be pretty unhappy with the result. What you do with the ingredients once they are in the glass/shaker is almost as important as the proportions and unless the machine is gonna tell me how to shake/stir/strain/etc… whats the point?

    Something like this would actually be useful as a tool to pre-measure drinks (and take orders) for use by a person who is doing the required shaking,stirring, testing, pouring, garnishing, etc… but if the system requires a human at the end of the chain to do all that, the selfsame human could just measure their own ingredients faster than the bot could. Notice they almost completely cut around ever showing the bot move the glass from spout to spout… makes me wonder what the actual throughput on the thing would be.

    Maybe this is for a crowd that likes more novelty than quality… but i wouldn’t trust it with my drink.

    1. @dgently As the co-creator of “The Inebriator” which this seems similar to, I agree entirely (well i know what goes into the inebriator so i do trust it with my drink), its just a bit of fun to liven up a party, The barobot and the inebriator are a long way off replacing quality mixologists.

      1. @Jake
        I agree completly. Neither Barobot nor Inebriator or any other machine is going to replace human bartender any time soon. So it is definitley not dedicated for bars. Even on Enterprise where drinks were created by replicator there was a human (and kicking!) barwoman.
        But such machines could be quite useful in places where no mixologist pro is available like house parties, weddings or networking events. And as you know being served by a moving robot is fun by itself :-)

        Barobot takes about 40 seconds to pour a medium size, 4 ingredients drink. You can see this on the second movie on Kickstarter page. I can imagine that human barman can be a bit faster than that if he/she is in hurry. But as I wrote above it is not about efficiency or human-robot competition after all.

        1. Like I said, there is def the novelty appeal… and I am sure that these would beat the pants off of the usual “table of odd alcohols and various mixers” found at many parties, I just want my automated mixing robot to strive to a higher goal of making at least pretty decent attempts at a real cocktail, instead of just a mixed drink.

          maybe there is a barobot/inebriator 2.0 that has a big display that pours the drinks straight into a shaker and instructs the user as to the finishing steps for each drink. Could be like a little game… “bartender hero” put some sensors on the shaker or thermocouple in the stirring glass and give people points and timing or whatever. but at that point you could almost just skip the mixing robot part and just have little light up trays for all of the bottles and have people pour for themselves as directed by the program. That way you are even teaching people how to make the drinks w/o the bot.

          even just making sure the drink is prepped correctly with the ice is going to vastly increase the quality of the drinks from one of these things I bet.

          Of course you are still on your own for using quality ingredients with any form of home mixology… and I really hope nobody working on barbots actually ever uses “juice syrups” in place of real juices.

    2. It may be useful, but even a newbie bartender is still 5X faster than this. a Good bartender is 20X faster.

      Not going to be replacing the human bartender any time soon.

    3. I’m in my mid/late 30’s , so getting smashed really hasn’t been my thing for oh, about 16 years or so now and don’t really drink at all anymore. But you’re right, a mixed drink or cocktail is more than a bunch of booze dumped into a container. A lot of drinks require not only an exact measure of contents, but a certain way of mixing them to get the desired look and feel of the finished product. The kids that would be interested in using something like this would be better served just doing their usual keg stands, beer bongs and pongs… Or adapting this to a hose that one could chug from. Or incorporating a mixer of some sort?

  4. I became acquainted with this machine during Makerland in Warsaw two months ago and I must say I wasn’t sober when I went home. Of course this is more of a novelty or gadget (big one) than a barman replacement. Having something like that at a home party would be nice, useful and fun :)

    On the technical side, there were a few problems:
    – I’ve noticed the valves they use don’t work well with soda drinks and viscous/ fluids
    – hall effect sensors mark the positions of bottles but it still required manual calibration (minor bug?)
    – the carriage was stopping when reaching a bottle with no deceleration period, so the speed of the glass couldn’t be high or it would fall out – I guess that can be fixed with the code.

    The nice thing about this machine is that if everything goes well, the only thing you have to clean is that plastic valve. In all other drink mixing/serving machines I saw, I was always worried about the components that touch the liquid not being safe for food (like paintball gas tanks and the CO2/compressed air, valves, hoses…) – barobot is free of those worries.

    1. 1. If you use the same glass sizes, you shouldn’t need to stop – just switch off the fluid flow before the glass is no longer under the dispenser.

      2. My personal preference for a machine name is “BarBar Ella”. With a photo of [young] Jane Fonda.

    2. I’m Konrad from Barobot team

      A lot has changed for the last months. And we learned (again!) how long and bumpy is the road from a good prototype to a +/- mature product :-)

      The trick we found to pour carbonated drinks was really quite simple (and as such it took us a while to find it): once the bottle is installed you make a pinhole at the top of upside down coke bottle. It releases excess pressure and the liquid stays carbonated for several hours.
      Viscous syroups like grenadine take more time to fill the dispenser after it is emptied. We accounted for that in software.
      Using smaller magnets gave us sharper magnetic field spikes and better controll over positioning. No need for offsetting positions anymore.
      Deceleration in both X and Y axis is implemented. And it changes as liquid

      There are several smaller issues still waiting for fixing. That’s why we kickstart.

      1. > “The trick we found to pour carbonated drinks was really quite simple (and as such it took us a while to find it): once the bottle is installed you make a pinhole at the top of upside down coke bottle”

        So you found out that leaving the bottle open solved your problem?
        My Soda gets stall pretty quick when kept open. Even in the closed bottle when it is <50% full. I guess it has to do with surface area that is able to release carbon. What is you solution for this? A Pinhole might be small, but the surface area of your container is still broad enough that every soda will get stall <60 minutes instead of "several hours". I dont know if I can believe that part. Unless you want to rephrase your sentence from "The trick we found…" to "We just let the container open".

        1. Last time we were partying with Barobot a 2l Coke bottle was used up in less than 3 hours and was still somewhat bubbly at the end.
          At the same time I agree that pinhole is workaround and not perfect way. I am sure that more elegant and cost effective solution can be found here (maybe a punch through safety valve to keep pressure high, but not too high?). This is an open source project so sooner or later somebody will come up with upgrades.

  5. A fat out party novelty I guess. send your order remotely via your smart phone or tablet, why when have to walk over to the machine to get it? Just one more way to justify the total cost of the phone or tablet?. Don’t get me wrong I’d have an android phone if I could afford both the phone and required data plan, and yes I’d probably use it for something someone else would think silly, I’m sure you could bet on that.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.