“Tell Us The Condition Of Our Craft” — The Case For Pilot Mode: Improving Energy Feedback In The Tesla Model 3

Sign up for daily news updates from CleanTechnica on email. Or follow us on Google News!

This is part one of a four-part series. Enjoy, and stay tuned for more soon!

There is an epic moment in the Philip Kaufman adaptation of Tom’s Wolfe’s The Right Stuff that teaches us about a vital improvement electric car manufacturers should make for the public.

It is February 20, 1962, and Astronaut John Glenn is hurtling above the earth at 17,500 miles per hour over Australia in his third and final orbit. He is the first American in orbit — a daring flight aboard a failure-prone rocket. Meanwhile, as Glenn zips along, a warning light is indicating that his capsule’s heat shield might be loose — a fact that, if true, would surely be fatal on reentry. What to do?

Mercury spacecraft cockpit. (Courtesy Brett Williams and Motion Arts)

On the ground, engineers nervously discuss leaving the retro-pack on through the fiery return — a potentially dangerous, untested, and completely unplanned sequence in the flight. Meanwhile, flight directors at Mission Control don’t want to alarm Glenn or anyone else. Actor Scott Glenn, who plays Astronaut Alan Shepard, is the capsule communicator at Mission Control in Cape Canaveral, relaying information to the orbiting astronaut while engineers discuss the harrowing prospects.

Finally, in exasperation, Shepard turns to the Flight Director, “How long are you going to keep him in the dark?” Flight Director: “What do we tell him?”

“He’s a pilot,” Shepard declares. “Tell him the condition of his craft.”

“Tell him the condition of his craft.” Alan Shepard as Capsule Communicator. (Source: NASA)

Informed as to what to do with the information (manual everything on reentry), John Glenn hums his way through a dangerous flaming return to earth and an eventual hero’s welcome.

Astronaut John Glenn, Mercury Atlas 6, February 20, 1962. (Source: NASA)

So, what does all this have to do with electric vehicles? This:

Right now, when we drive an EV, we’re being unintentionally kept in the dark on how energy is being used. When you are not moving in the car, you are completely unaware of the energy use of the automobile’s non-propulsive systems. The theory might be that we wouldn’t know what to do if we had better information. But is this really true? And why do I think it matters?

People driving cars with internal combustion engines don’t seem to need such information. Why should it be any different for EV drivers?

I’m making the case that we need a special mode: Pilot Mode. And I’ll explain why.

What happens when EV drivers are kept in the dark?

They estimate and speculate. Theories and assumptions emerge — some informed; some not.

One thing that is quickly realized when studying Tesla owners is that a surprising number are engineering savvy. Say what you may, but they are a data-hungry group. (Look at Tesla forums if you need convincing.) They know they are driving a technologically advanced vehicle and want to understand its inner workings and nuances. Many want to optimize its performance. Indeed, that was a critical warning for me in even writing this article. Lots will be looking over my shoulder to see if the numbers are right.

Range is critical for EV drivers, even those who only faintly pay attention to such things. This is particularly true on long trips or under winter conditions. When EV drivers don’t have specific information on how energy is being used in their car, they evaluate or sometimes even invent stories (usually with some supporting data) to explain what they see with regard to changing range when driving. What do I mean specifically?

A little background

In October 2018, after a long wait along with many others, I took ownership of a white Tesla Model 3 Long Range (LR) with rear wheel drive (RWD) and 19 inch wheels. I was immediately smitten with the automobile, which seemed to be something of an engineering marvel.

Tesla Model 3 LR on October 6, 2018. Photo by Danny Parker.

I was intrigued by the energy displays and also the changing range estimate for the battery pack. However, driving it over the next month left me puzzled — several key items clearly influenced range, but information regarding their energy use seemed to be unintentionally hidden from drivers. I live in Central Florida, with little need for heating.

Electricity use for all systems in an EV are so critical that drivers not wishing to be surprised by adverse changes to range and performance need to have that information — and not only while moving.

Would engineers not want you to see what’s going on?

The Model 3 has a lot going on under the frunk. An obvious intention with Tesla’s electric cars is to make the transition for typical drivers as undramatic as possible. One line of thinking: the more prominently one shows numbers on energy demand from the battery, the more people may worry about them even if they should not.

Yet, that may not be a good reason to shield drivers from how the systems are operating. For sure, if most users do not want detailed information on energy demand, instantaneous and over time, they can easily have a default user interface (UI) that doesn’t show that information. However, for the others, having such information available is really important to understanding what is going on with the energy systems in the car. And it is vital if one wishes to maximize range.

For instance, in winter, the Model 3 uses waste heat from the inverter to condition the high-voltage battery when you are approaching a navigation-planned Supercharger stop in order to get the battery to a temperature that will more quickly accept energy when you get to the charger. This will help maintain a quicker rate of travel and help your car heat the battery more efficiently. The car will tell you when it is doing this, but with many of the heating, ventilation, and air conditioning (HVAC) operations, arguably more important to drivers, the car tells you nothing about your less-than-optimal choices.

A really terrible problem is that when running the cooling and heating system in the stationary car (not driving), the energy consumption from the battery pack is not shown in any fashion. The displays show watt-hours per mile (Wh/mi) when the car is moving, which was fine, but it really does not teach the driver what is causing its values to vary so much. The aggravating clue is that one sees range fall while sitting in the car. The key thing missing is effective feedback on power (kW) coming out of or going into the battery pack regardless of whether the car is moving or not.

Figure 5: With the car in park, not the slightest indication on the left-hand status panel that space conditioning system is running and set in least efficient mode (AUTO). Note potential fix: insert for status panel. (Author)

Early on, I also noted a strange inconsistency in the energy display for the 15 and 30 mile driving history against the 5 mile display, which seemed a bit misleading — the amplitude of the displays for 15 and 30 miles should either be the same as that for 5 miles or it should be noted that the data for the other two are being averaged over distance for those displays. But that was a small issue compared with the missing indicator for kW to let drivers know what power drain is like from the battery, whether moving or not.

I tend to be a person who acts on engineering deficiencies encountered — at least, the ones that appear easy to address. What to do?

I posted my frustrations on Twitter that autumn and was soon contacted by my good friend Jon Koomey, then at Stanford, who then put me in touch with JB Straubel at Tesla. I was surprised when Straubel quickly responded.

“Indeed that display in your twitter post is a direct descendent from a display that I built far back in the Roadster,” he wrote, “that evolved into the S/X and now was mostly carried over into the 3.  We haven’t done a big overhaul of it in many years.”

I wrote back: 

Dear JB and others at Tesla,

Thanks to my old friend Jonathan for the introduction.

I have become aware of the key information that a driver does not have available and that would provide valuable insight into how to improve performance.  It would also end the confusion I’ve seen out there about what energy is used for accessories such as air conditioning.

And yes, I get it, that nerdy stuff about consumption and metrics is not high on your list. Tesla Forums show there are plenty interested customers out there and that niche group wishes to get the most out of the product. 

We Need a Special Mode

1) The instantaneous kW draw from the battery or power supply.  This can be noted on an upper banner on the Wh/Mi display.  This needs to be given to at least two  significant digits.  Examples:  with acceleration: 19.44 kW.  Vehicle parked and headlights on: 0.26 kW; Vehicle parked and headlights on and air conditioning on: 1.86 kW.

2) Regen kW: The display should ideally also show Regen in kW to the battery when in that mode (a quantitative description of what is currently shown in the green bar on the lefthand display panel.)

Why is this important relative beyond the Wh/mi display?

Because the Wh/mi display is linked to the car’s velocity which does not help drivers know how much accessories draw electrically.  It can even mislead users since the impact of various choices can be hidden by speed. Also, there is no information at all when the car is stationary, other than  slowly falling  range.

For instance, when driving at freeway speed one might conclude that the energy use of the air conditioner is almost incidental to performance. For instance, if the air conditioner draws ~1.5 — 2.0 kW when thermostatically controlled (which it does as we’ll show later).  Let’s say the baseline energy use of the Model 3 without AC is 270 Wh/mi at 75 mph. Driven an hour on the freeway, it would use 20.6 kWh from the battery. Assuming the AC on AUTO draws 1.6 kW, this means that with the AC on at freeway speed, the Wh/mi only goes up 296 Wh/mi — a 10%  decrease in apparent efficiency.

One can see many drivers on the Tesla forums doing that arithmetic and deciding air conditioning doesn’t matter much.  Yet, that is not so — at least not in every application. For instance, when driving around town at 30 mph (without HVAC, efficiency is about 134 Wh/mi or 4 kW), the cooling power is increasing total vehicle energy use by 30–40% depending on the AC power draw.

 The EV as a mobile office 

As the car environment becomes nicer, the interior becomes more like a private drivable office and break room. People are spending more time in stationary cars. With the recent COVID pandemic, this is even more true. People are getting takeout coffee and then sitting in their cars to enjoy the drink and check on emails. Why not?

Increasingly in the COVID pandemic, automobiles are being used for office and entertainment. Photo by Danny Parker.

And in summer, sitting for 45 minutes with the cooling system on can mean losing 4 miles of range (1 kWh) before you leave. In Fremont, at Mission Coffee, finding a shady spot might allow one to avoid that altogether. In Florida, at 91°F, you know you’re going to be using the air conditioner. Winter is another matter. On a cold winter day, engaging maximum heat for the same period of time or engaging the defog mode could result in twice as many miles lost.

Having a display showing realtime kW would make the source of all that clear. It would also alert drivers that the heating or cooling system is on even though they don’t suspect it. And this is beyond Wh/mi, which confuses the issue for most people — and also provides no information when the car is stationary at a stop or when parked.

 What you don’t know can hurt you 

Various facets of the Model 3 heating systems are invisible — or at least not well documented. For instance, unless the cabin temperature is set to LO, drivers can unintentionally have the cabin heat come on in mild weather conditions and not even suspect it. Nothing in the car tells you that the heat is on other than maybe noticing your Wh/mile is higher. Not having a display of realtime power draw becomes a real liability. The same applies to defog mode, which can use both cooling and heating. Forget and leave it on by accident? Lose range needlessly.

All of these things mean that drivers don’t have needed feedback on how operation of these systems can influence range. And the lack of that information can create all kinds of supposition from “blower speed doesn’t matter” (it does) to “turn off the music to save power” (it hardly matters) to guessing as to whether Sentry Mode is increasing Phantom Drain while parked (it is — you can lose a mile of range each hour it is activated). Thus, it would be highly advantageous to have the instantaneous kW either a negative or positive value at the top of the display where there is currently a lot of white space, or else on the left panel. I have mocked up a display.

Yes, I know the Tesla displays are a crafted embodiment of Dieter Rams’ “less is more” aesthetic. Rams has said, “I always try to find things I can remove.” So, simple, clean, unobtrusive lines are important. But among Rams’ design principles is one that I’d like to call attention to relative to the current Model 3 User Interface (UI). Good design helps “understand” a product. He also says good design provides “a welcoming experience.”

I am also aware that Tesla has great experience with creating effective, clean UI displays. However, in its quest for beautiful simplicity, I’d like to make the point that not making Battery Power optionally available does not help drivers understand the car. It helps to mystify them. This is particularly so when parked, but even more true in winter when a number of factors make range preservation important. Also, the difficulty with the UI for the HVAC system is confusing enough for drivers to require all-too-frequent access to the control screen just to see what it is doing. That’s annoying, distracting, and not just for me. Rather than a quiet, intuitive, “welcoming experience,” it is difficult for drivers to figure out how to turn off heating or to run the fan without it.

My proposed fixes are not beautiful. I am not a visual designer. So, what I have mocked up is crude and can easily be improved upon. Would Battery Power be best shown in Pilot/Nerd mode on the Charge screen, or as another selectable plotted line over time on the Energy screen sub menu, or as a simple small black or green numeric value to the left or right below the MPH text on the status panel with the existing black and green lines showing battery discharge/charge? This is for Tesla to tackle. To address the confusion from the HVAC controls, it would be useful to simply have an unobtrusive Boolean display of HVAC system status on the left drive panel as shown in my mockup. Again, it could be easily improved upon by a UI designer. What I offer here is just a constructive starting point to show how this can be done.

Pilot Mode

Pilot mode would only be selectable and not a default option. Users who wouldn’t want to know about such things would never see them. However, the rest of us would be completely grateful to have access to that data. In particular, it would be great to see:

  • The instantaneous kW draw and regen from the battery on a display showing kW draw into or out from the battery to 1 or 2 decimal places with a 1 or 0.5 second update rate, which is optimal for cockpit rendering of non-critical outputs from a human-factors standpoint. That would clearly show the big advantage of the seat heaters (~300 Watts or 0.3 kW) versus the cabin heater (~3000–6000 W).

This is Battery Power in the Tesla logs, which comes in at 100 Hz on the CAN (Controller Area Network) Bus, but is then rendered at 60 frames per second on the ScanMyTesla apps used by the author to develop data for this series. For human attention in the cockpit, the optimal display rate is significantly lower: 1 or 2 Hz being a good compromise between rapid update and eye distraction. This would mean taking the integral of the high-frequency measurements to show a 1 second average. Faster than 2 Hz is not recommended. Indeed, numeric displays flashing at 4 Hz or greater can interfere with mental processes, particularly for those prone to epilepsy. Given the certain capability crossover between Tesla and SpaceX, human factors optimization within the Dragon cockpit and the Tesla cabin informational displays is available. Would it be useful to provide separate power data for the summed HVAC systems? There is plethora of data available to Tesla software developers: well over a thousand telemetry/logs on the multiple CAN buses.

Data can be displayed on the Energy Charge screen, which is currently fairly data sparse. It should be refreshed about each second. The display would be there whether the car is moving or not, so that power use when stationary can be tracked.

Below is an example during acceleration (red line on right panel). Note that arrow shows direction of power flow and quantity (10.11 kW at 54 mph)

Example of Real Time kW display available on charge screen (indicator vanishes when charge limit is being set). At 54 mph, the car is drawing 10.11 kW from the battery for propulsion and all systems. Numeric values only available by selecting Pilot mode in details. Image by Danny Parker.

A second example shows regenerative braking (green line on right panel). A green arrow shows direction of power into the battery.  It would also show charging. This variable is already readily available in the Tesla onboard logs and is referred to as Battery Power.

Example of Real Time energy display on charge screen showing regeneration of 3.41 kW on downhill stretch. Image by Danny Parker.

A key justification for disclosing this single metric is to provide information not only for what the drivetrain uses to move the car, but also for its various summed sub-systems, including heating, air conditioning, lights, and so forth that have a real impact on range and efficiency.

Right now it is not possible to see the status of the heating and cooling system without accessing that screen — very annoying if you are trying to be frugal with power. It would be best to have an optional display on the left-hand banner with speed and status that shows the current settings of the HVAC system (AC/HEAT/Setting/AUTO). Even better would be a display there that showed the kW of the HVAC system and heated seats at the moment.

Measured air conditioning energy use of the Model 3 

At the FSEC Energy Research Center where I work, we have detailed monitoring equipment hooked up to employee chargers that measure kW with good resolution. When the car is fully charged and is still plugged into the charge circuit, if accessories draw electric power, they are recorded by a dedicated logger on the Level 2 charger so long as it doesn’t exceed 35 amp. This can then be used to assess the power use of non-propulsive vehicle energy use. We found this to be an easy way to measure the air conditioning energy use of the Model 3: 

Measured Tesla Model 3 air conditioner power. Note initial 2 kW demand with fans on full (AUTO), that then throttles down to approximately 1.5 kW when thermostat setting is reached. Image by Danny Parker.

These graphic data showed that the compressor drew about 1.2 kW, the car drew about 0.2 kW while idling, and the high fan speed on AUTO when remotely cooling causes power to increase approximately 2 kW. Thus, high fan speeds increase power by at least 350 watts over low fan speed choices. Who would know without measuring?

I shared this and other collected data with Tesla and we had some meetings about why I thought better energy feedback was potentially important to some drivers.

I pitched my appeal to Tesla for a real-time energy display.

How did it work out?

It didn’t lead to anything, unfortunately.

I was passed on by David Lau to Scott Sims and then Adrienne Tran over the next few months. We had a few online meetings. My reception was kind and attentive and interactions were promising. Unfortunately, nothing came to pass.

Indeed, JB Straubel himself had warned me, before leaving Tesla as its Chief Technical Officer, that, company-wise, “[better] energy displays and giving people even more data about consumption metrics” were a low priority. He added that he personally thought them very important (Straubel is an accomplished pilot).

Although my efforts to suggest change to Tesla were without effect, these posts will serve as an appeal to Tesla to reconsider — at least with a special mode for those who want to know more about the condition of their craft.

Is this problem just one for data geeks and energy nerds? I don’t think so. Pilot Mode would help a lot of drivers.

Learning about how your machine works best is almost always an advantage. Ask any pilot.

Danny Parker. Photo by Lisa Shepperd.

Danny Parker is Research Scientist at the FSEC Energy Research Center, where he has worked in the energy efficiency field for the last 30 years*. Beyond better machines, he has a keen interest in low-energy cooling technologies, zero-energy homes, rockets, and good coffee. His sister lives in Fremont, where he became familiar with Tesla. Neighbors and other pals in Cocoa Beach work for SpaceX. Tesla investor.

*Disclaimer: The author’s information and opinions do not imply recommendation, endorsement or favor of specific products or services by the University of Central Florida or FSEC, its energy research institute.

Have a tip for CleanTechnica? Want to advertise? Want to suggest a guest for our CleanTech Talk podcast? Contact us here.

Latest CleanTechnica TV Video

I don't like paywalls. You don't like paywalls. Who likes paywalls? Here at CleanTechnica, we implemented a limited paywall for a while, but it always felt wrong — and it was always tough to decide what we should put behind there. In theory, your most exclusive and best content goes behind a paywall. But then fewer people read it!! So, we've decided to completely nix paywalls here at CleanTechnica. But...
Like other media companies, we need reader support! If you support us, please chip in a bit monthly to help our team write, edit, and publish 15 cleantech stories a day!
Thank you!

CleanTechnica uses affiliate links. See our policy here.

Danny Parker

Danny is principal research scientist at the Florida Solar Energy Center where he has worked for the last thirty years. His research for the U.S. Department of Energy has concentrated on advanced residential efficiency technologies and establishing the feasibility of Zero Energy homes (ZEH) — reducing the energy use in homes to the point where solar electric power can meet most annual needs. The opinions expressed in this article are his own and do not necessarily reflect those of the Florida Solar Energy Center, the University of Central Florida or the U.S. Department of Energy.

Danny Parker has 17 posts and counting. See all posts by Danny Parker