I redesigned the bubble display to address some of the issues I was having with the previous version. Hit the jump to see what I changed to make it work.
One of the big problems I had in the previous version was that opening all eight air valves at once yielded dramatically smaller bubbles than opening just one. To combat this I added two large glass chemistry flasks to act as an air reservoir. I drilled a bunch of holes in to rubber stoppers so that I could run all the air lines into the glass. This worked out nicely because the rubber grips the air tubes securely without the need for hose barbs while creating an airtight seal. I used two flasks because I couldn’t put enough holes into a single rubber stopper to support all the solenoids. To help keep things even, I ran an air hose between the two. I am not sure if really makes much of a difference, but all the lines coming out of reservoirs running to the solenoids are all the same length. In the previous version I cut them to different lengths to make everything look a little bit more organized. I speculate that since air is compressible that this might have resulted in slight variations of air pressure running to the solenoids and consequently variations in bubble size.
The second big change is the addition of vinyl tubes to hold the liquid. The previous version had one large tank to hold the liquid with little dividers inserted to keep the bubbles on the right path. Now each solenoid gets its own length of clear vinyl tube. This makes a huge difference. Now when I force bubbles into the tubes they rise at a steady rate independent of the size of the bubble. The key seems to be making sure the bubble is large enough that it contacts the walls of the tube on all sides. Another thing is making sure that the bubbles are not formed too close together. There seems to a minimum gap that can exists between bubbles. Any closer and the trailing bubble will be sucked up into the leading one and collapse into one larger bubble. With mineral oil the minimum gap is acceptably small.
One thing to note about the new setup is that the amount of liquid displaced by the air is now much more noticeable. In fact, while I was initially calibrating the printing algorithm I increased the duration of time that the solenoids were opened for each bubble. That made for bigger bubbles, which made for more displaced liquid. At that point I had the tubes filled with blue colored water instead of mineral oil. Long story short, I now have blue stains on the ceiling in my study.
All in all I really like the new setup. It is much easier to work with since the check valves fit right into the vinyl tubes and make a wonderfully secure water tight seal. Plus with the tube there are no longer any worries about leaks along the seams (because there aren’t any). One problem I did have with the vinyl tube is that it comes wound up onto a drum and when you pull it off it wants to curl back up. To fix this I cut it into lengths, stretched them out on wooden dowels and then hit them with a heat gun until the curl relaxed. This is a functional solution, but it is really slow. If I do a bigger version, I will probably need to come up with a better solution.
In case you missed it here is the thing printing out the letters VML