I have been trying to come up with a standard test to measure relative longevity of bubbles. To do this, I am trying to keep the bubble size constant and easy to reproduce. To do this, I cut off the bottom of a plastic water bottle. The bottle's neck is dipped in the bubble juice and then the bottomless bottle is lowered into a jar full of water. The timer is started and we measure the time until the bubble pops.
All was well and I tested quite a few solutions (5 dips per solution -- although it should really be more like 20 dips to get reliable stats but that would be too time prohibitive).
The next night, I went to test the next batch of solutions plus one solution tested the night before so that it could be used as a standard against with they others were measured. The standard solution's was creating bubbles whose longevity was very different (much shorter) from the previous night's tests. The conditions were almost identical (70 degrees and humidity of 50%) but I was in a different room from the previous night. I went back to the original room and the results were different from both the room I just came from AND the previous night. I then put the apparatus in the sink -- and lo-and-behold the times corresponded to the previous night.
The realization came that even very small air currents (small enough that they don't cause much/any wobble in the bubble) are effecting the bubble life. AHA! I can protect the bubbles from the elements by putting a container over the bubble to protect it. PROBLEM! Even the solutions that produced bubbles with poor open-air lifetimes were lasting 5 minutes or more (versus an average of 12 seconds for the same solution. And the solutions with good lasting power were lasting more than 15 minutes. With these longevities, it would be impractical to apply the test to a lot of solutions.
One benefit of not overprotecting the bubble is that the test result is an indicator of longevity with some durability built-in which may be a measure of greater interest IF it maps to something in the real world.
So, I need to find a way to protect the bubbles similar to the protection offered by the sink (which yields averages of 10 seconds for poor solutions and several minutes for the best solutions) without having to actually use the sink (since our kitchen is only quiet enough for me to test in during the wee hours of the night.
I have a few ideas but won't get to try them for a few days.
MORE INFO (Added evening of 6/11/2010). This morning, I realized that reducing the size of the test bubble makes them less prone to accidental break since they wobble less. Unfortunately, I also got a lesson in how radically changes of humidity affect bubble longevity. At noon, I was able to do a battery of tests while the humidity was 51%. As the day progressed, the humidity dropped to 30%. The same solution that averaged a minute per bubble (over 10 bubbles) is averaging barely 20 seconds! It isn't a surprise but it drives home what every bubbler knows. So, for these tests to be meaningful, I have to hold the humidity as close to constant as possible.