To know how big is a drop of shaving oil is crucial from several aspects. First, how much should someone use when shaving a particular area? Second, how big should the bottle be that it is distributed in? (i.e. you don’t want someone going to the store every week or two, nor do you want the the shaving oil to go bad before it is consumed) So it is important to know how big is a drop.
When I was first experimenting with shaving oil many years ago, I wanted to know how much I was using. I was using a dropper bottle from a previous project and had no idea how big each drop was.
To figure that out I used my kitchen scale which is a great tool to figure out how big the drops were. Knowing that the density of water is 1 gram per cubic centimeter, I filled the bottle with water and went to work. I collected data by counting the number of drops and how much they weighed.
The thing that made this a little tricky is that the scale only has a resolution of 0.1 grams and the drops are much smaller than the resolution of the scale.
Additionally, the scale has a “de-noising” circuit to help provide stable, reliable readings and so it was having a hard time deciding if I had added a drop or two or if it was experiencing noise. This required a little “trickery” on my part. The de-noising circuit, or logic, could only be relied upon to give a new value if there was a sufficiently large change in the mass that it was outside what the circuit considered as noise. What I did was to add a drop to the collection container and then I would depress the scale with my finger so that it would see a large change. When it stabilized after that much larger excursion of releasing my finger, I would write that number down. I recorded the data and put it into a spreadsheet. Using the X-Y graphing feature, I had the program generate a best fit line with equation through my data.
In the graph (click on it for full size view) you can see the equation. The value of the X coefficient (0.0423) represents the drop size. So how big is the drop? My data shows it to be 42.3µL (micro-liters). The R-squared value of 0.999 gives me great confidence in the quality of my data collection.
As an aside, you can see in the graph that the data comes in groups of 3 to 6. Notice that there are groups of data with lesser slopes sitting on the longer, overall average line. (Look in the vicinity where the data crosses the 2g line -5 points and also just before it crosses the 3g line -4 points) That shows the effect of the scales digitizing circuit and how it groups or interprets the data which does not come in quantities that exactly match the digitizer. As I was collecting the data, I would see this manifest as the number of drops required to increment the reading change. For example: 3, 3, 3, 2, 2.
Now that I knew how big is a drop, I could plan rough bottle sizes. When I went into production and purchased bottles, the vendor I chose provided specific details on drop size. They answered the “how big is a drop” question for all of their bottles. It turns out that for their numerous bottle sizes, the size of the drop was either 40 or 42µL.