Got some new data with the newly adjusted shell, so I decided to try to run some new numbers for percent improvement. I am only going to use the most recent figures between the last time I removed the shell to now to try to reduce variability. Most of my trips are to the same few places, and most of the time I know any outliers that would skew the data and can throw them out. I know this isn’t the ABA quality everyone would like, but it is the best I can do right now given time constraints I am under.
For my “shell off” baseline, I am going to use the 5/8 and 5/19 tanks. These are both very typical routine tanks. The 5/17 shell off tank was way off as I was running different tires and making short trips so it is out. These two tanks sum 480.168 miles using 17.52 gallons so 27.407 MPG.
For my “shell on” comparison, I am going to use the 5/20 and 5/22 tanks. The 5/20 is a very typical routine tank (first tank at over 30MPG!) but the 5/22 tank was slightly a-typical and in worse conditions. These two tanks sum to 404.56 Miles using 14.123 gallons so 28.6455 MPG or 4.52% improvement.
I know one of the trips from the 5/22 tank was way off my usual driving routine, if I factor out that one trip by removing 28.4 miles at 25 MPG per the Ultra Gauge so 1.136 gallons, that would make the figures 376.16 miles at 12.987 gallons or 28.964 MPG or 5.68% improvement. This would make this tank more typical in terms of destinations of the other three tanks. Note I am not trying to “cherry pick” figures, just trying to get an apples to apples comparison and minimize variability.
The first data set showed 5.45%, so any way you slice it, it looks like the half shell gave around a 5.5% improvement to which I would assign a +/- 1% tolerance given the inaccuracies of my measurement procedure, weather, and variations in even my “routine” trips. So between 4.5% and 6.5% FE improvement would imply a roughly 9%-13% drag improvement.
Utilizing Phil’s drag as a function of aft body truncation table and the AST-II overlay, I was predicting a 11.2% reduction in drag. So it looks like the theoretical landed right in the middle of the measured results range. So once again thank you Phil and everyone else on this site who have shared such a wealth of knowledge.
|