Routing risk management

The routing of the title is not the network routing that LJ does for a living, but the routing that we needed to complete our cross-country adventure.  As anyone who knows both LJ and me might guess, I’ve been taking a more conservative approach to Tesla energy usage than LJ.  Many of the western SuperCharger-to-SuperCharger legs are laid out such that the first part of the leg is energy-intensive as we climb a mountain and then the second part uses little or no energy or even returns energy to the Tesla.  An extreme example from Colorado is shown below.

To ensure we use our time efficiently while avoiding getting stranded takes two steps: (1) use the evtripplanner calculations and overall elevation change to determine the average energy for the overall segment and (2) constantly monitor the energy usage to keep the segment average energy under that predicted number.  An example energy calculation from the software is shown below.

This screenshot is for the last two days of our trip.  Each of the SuperCharger stops is shown in green (yay for conditional formatting in Excel) and the overnight stop is shown in yellow.  From left to right, the columns are for: segment time, miles per hour, miles to drive, Rated Range miles, average energy in Wh/mi, the elevation gain and the elevation loss.  For each of the green lines, the data is a summary of the lines above it.  

Evtripplanner export with Excel conditional formatting

Taking the last SuperCharger leg of the trip as an example, we needed to use an average of 303 Wh/mi or less to get from Atascadero to Gilroy.  This Gilroy leg was an easy one to manage, as the high energy usage came at the end.  Additionally, for the Atascadero stop, we took a long lunch break, which means we were well above the 187 Rated Range miles that were needed to get us from there to Gilroy.  Had we just put in the 187 Rated Range miles we needed and caught a bad headwind, we would just need to drive more slowly than noted in the spreadsheet.

The most challenging legs for me were the ones where we were intended to use a high average energy rate early in the journey due to huge elevation changes.  On those legs, it was not unheard of for the average energy to stay well above 500 Wh/mi for miles on end or even 600-700 Wh/mi on really steep climbs.  While intellectually I understood that the calculations were quite accurate at sea level and skewed in our favor at high altitude, I still got very uncomfortable with that high of an energy usage.

LJ, on the other hand, was much more willing to follow the numbers.  The loose aviation analogy would be that LJ can handle a Tesla instrument rating and that I would be much more comfortable with VFR (visual flight rating).  [Note to the pilots and aviation nerds out there: I realize that this is not an absolutely correct analogy.]

*******

As we got closer to a destination, the absolute Rated Range value became less important in favor of the relative percentage difference between the Rated Range and the projected range.  If we were traveling at a reasonable 300 Wh/mi with 20 miles to go and 40 miles of Rated Range, we would have to instantaneously more than double the energy usage to be in trouble.  The three most critical variables for energy usage are elevation change, altitude density, and speed.  Elevation change is accurately predicted by evtripplanner and altitude density is dependent on the elevation, so the only user-controllable parameter is the speed.  As long as we don’t double our travel speed—which is difficult when traveling at or near the speed limit—we would not get stranded.

The closest we got to needing a tow truck was back on Day 2, when we rolled into the Angola, IN, SuperCharger with only 5 miles of Rated Range.  A few minutes prior to that, LJ got distracted by a hawk at the edge of the highway and almost missed the exit.  In the Midwest, exits are 5 or 10 miles apart, which means that we would not have made it if we had missed that turn.  Luckily, we caught the exit in time and got to the Angola SuperCharger.

Close to empty

*******

Speaking of luck... we knew we were pushing our luck by doing this #TeslaElectricStartupSuperTrip across the upper part of the US in early November.  It wasn’t until a couple days after arriving in CA on Sunday that we realized how close we had cut it.  Murdo, SD, and other parts Midwest and West caught a typical late fall snowstorm that would have made travel difficult or impossible for us.  The Denver area had snow and low 20s temperatures today.  Given that the Tesla is much less efficient in cold weather, we would have been booking multiple nights in a hotel and I would have been taking last-minute vacation as we spent a few extra days on the trip.