In a week or so, my 2013 Tesla Model S, now fitted with an 85-kilowatt-hour battery,  will wind up its third year of residence in my driveway.

It just finished its third cross-country trip, from my home in New York’s Hudson Valley to a winter getaway in Carpinteria, California, and the odometer just ticked over 55,000 miles.

It’s been a great ride. Though my battery capacity has declined about 7 percent, reducing my range from 265 miles to 248, the car still looks and drives like new, and I still tingle every time I get in it.

Not once have I looked out through the windshield and said to myself, “I wish I were driving that car instead of this one.”

DON'T MISS: Tesla Model 3 Unveil On March 31, Deposits Accepted That Day

Until last week.

I’d dropped by the Tesla store in Santa Barbara to see about replacing my worn-out Michelin Primacy tires. A  sales guy asked about my car’s age and mileage, and suggested that now would be a great time to trade in my three-year-old car on a new Model S.

“We’re really looking for CPO cars,” he told me, referring to Tesla’s Certified Previously Owned program.

“We could give you a good trade-in allowance for yours right now," he said. "But 60,000 miles is the upper limit for a CPO car, so once you get past that number, the trade-in value would go way down.”

He pointed at a gorgeous 2016 metallic blue S90D parked in front of the showroom. “If you’d like to test-drive that one, I could let you have it for 24 hours.”

2016 Tesla Model S 90D during Southern California test drive [photo: David Noland]

2016 Tesla Model S 90D during Southern California test drive [photo: David Noland]

Deal! We traded keys, and I drove off in the 90D.

At the time, I figured there was zero chance of me doing a $40-50,000 trade-up just to get the latest model. (Sorry, sales guy.)

But I was eager to see how the Model S has evolved over the past three years. 

No more 85

As I write this, Tesla has just discontinued the 85-kWh Model S. Its replacement, the 90-kWh version, is now available only with the dual motors that give it all-wheel drive.

ALSO SEE: Tesla Model 3: What We Know About Its $35,000, 200-Mile Electric Car

So the 90D is in fact the closest current equivalent to my 2013 S85, and the appropriate model for a then-versus-now comparison.

My car, with air suspension, 19-inch wheels, and no sunroof, had a list price of about $81,000 in 2013.  

An equivalent 2016 90D, similarly equipped, would price out at $91,500 today. About half the difference is the cost of the dual-motor set-up;  the other half is a combination of a 2014 price increase and some minor changes to standard equipment.

2016 Tesla Model S 90D during Southern California test drive [photo: David Noland]

2016 Tesla Model S 90D during Southern California test drive [photo: David Noland]

Better acceleration

The first revelation came almost immediately, as I pulled onto the 101 freeway. The 90D definitely feels quicker off the line.

My 85 has a listed 0-to-60-mph time of 5.2 seconds. The 90D’s official number is 4.2 seconds.

That’s a big difference on paper, but I’d always wondered how the lower number translated into real-world use.

READ THIS: Life With Tesla Model S: Service Centers Get Crowded, Cost More

It translates very well indeed: the 90D is noticeably, grin-inducingly quicker than my 85. 

In fact, I’d call the 90D’s acceleration almost perfect: It's all you ever need for normal driving, plus a little bit extra just for the fun of it. 

Anything more—like the over-the-top acceleration of the Model S Performance versions—seems to me good only for bragging rights, street racing, and scaring the pants off passengers.

Not that there’s anything wrong with that, but those aren't my priorities.

 

Better handling

I’m not the type of driver who screeches around curves on mountain roads, so I’m not particularly qualified to comment about on-the-edge handling for either car.

2016 Tesla Model S 90D during Southern California test drive [photo: David Noland]

2016 Tesla Model S 90D during Southern California test drive [photo: David Noland]

The only deficiency I’ve noted in my rear-drive 85 is a tendency to lose traction briefly and fishtail a bit when I accelerate hard into a sharp turn: entering a busy major road from a side road or parking lot, for example. And it’s especially noticeable on wet pavement.

I tried this maneuver a couple of times with the 90D (on dry pavement), and it was clearly superior. No hint of a fishtail. 

Winter traction

I’ve found the winter traction of my 85 to be mediocre, at least with mid-life all-season tires. 

Yes, the vaunted traction control works fine, but all that means is that the car stops moving up the gentle grade in my driveway before it spins its tires. 

CHECK OUT: Is Chevy Bolt EV's Main Mission To Marginalize Tesla Electric Cars?

There was no snow available in Southern California for a winter traction comparison. (Damn!) 

But I think it’s safe to assume that the 90D would be vastly superior to my 85 in this regard.

Autopilot

Autopilot is all the rage in Teslaworld these days. (Head-snapping acceleration and that 100-MPGe efficiency are so 2014....)

My 90D tester was of course enabled with the Autopilot software, a $2,500 option.

After three years of perfectly satisfactory cruise control in my 85, I was eager to see what all the shouting was about. And as a former private pilot who used autopilots in small planes at every opportunity, I wanted to see how the autopilot concept translated to the ground. 

Tesla Autopilot suite of features - with version 7.0 update

Tesla Autopilot suite of features - with version 7.0 update

Unfortunately, Tesla’s Autopilot and I got off to a bad start. 

By coincidence, I first engaged it just before a section of Highway 101 where the lane-divider lines are poorly painted and the grooves between the slabs of concrete are particularly conspicuous.

I double-flipped the autopilot control to engage it and let go of the wheel. The car beeped softly, the lane lines depicted on the instrument panel turned blue, and the car began steering itself. 

But within a few hundred feet, the painted lane lines on the highway diverged from the concrete grooves. Apparently confusing the two, the car abruptly swerved halfway into the next lane before I could stop it.

Five seconds after I’d turned it on, Autopilot had scared the [obscenity] out of me.

Fortunately, there was no car in the next lane, or I might have learned about the outer limits of my insurance coverage. 

With better lane markings, Autopilot performed well during my subsequent tests. As advertised, it stayed in its assigned lane and followed the traffic ahead by whatever distance I commanded.

Tesla Autopilot sensor system

Tesla Autopilot sensor system

But I found that the car couldn’t figure out what to do when driving in a right-hand lane that ended, when traffic was supposed to merge left. After a couple of aimless darts back and forth, a quick driver takeover was again required.

If I left my hands off the wheel for too long, the car would remind me to put them back. But just the right  touch was required.

Too gently, and it didn’t feel the touch. Too firmly—when combined with a slight pressure right or left—and it would interpret the move as an attempt to overpower it, and shut off automatically.

As a result, there were a couple of times where I wasn’t quite sure which one of us was driving, me or the car.

My experience is apparently not unusual; Car and Driver magazine recently tested the Model S autopilot system (along with those of three other luxury cars), and noted 29 “lane control interruptions” along its 50-mile test route.

Amazingly, that was far better than the competing Mercedes, BMW, and Infiniti systems, which registered anywhere from 56 to 93 hiccups along the way.

Tesla Autopilot Test

Tesla Autopilot Test

In the end, I found that all the commanding and monitoring of Autopilot was actually more mentally taxing than driving. Ironically, for me the sigh of relief and feeling of relaxation came when I turned Autopilot off and took over the wheel myself.

That’s just the opposite of my experience in an airplane cockpit, where turning on the autopilot makes the pilot's workload and heart rate go way down.

Familiarity helps

I have no doubt that my comfort level with Autopilot would grow with experience.

And according to Tesla, the car actually learns from its mistakes—and the mistakes of other Teslas as well—and improves over time.

2016 Tesla Model S 90D during Southern California test drive [photo: David Noland]

2016 Tesla Model S 90D during Southern California test drive [photo: David Noland]

And I can see how the system could be an absolute Godsend in stop-and-go traffic.

But for me, overall, Autopilot proved to be more trouble than it was worth. Perhaps this reflects more on me and my driving philosophy than on the car itself.

My 85, with its instant acceleration, instant deceleration, and quick, smooth steering, is actually fun to maneuver in traffic on busy freeways and Interstates. It will go exactly where I want to put it, instantly and effortlessly.

Yes, it takes some engagement and attention. But so does Autopilot.

I discovered that I prefer to use my engagement and attention to drive the car rather than just monitor it as it drives itself.

2013 Tesla Model S owned by David Noland, Catskill Mountains, NY, Oct 2015

2013 Tesla Model S owned by David Noland, Catskill Mountains, NY, Oct 2015

Battery range

It’s said that lithium-ion battery cells are getting cheaper and more energy-dense by 6 or 7 percent a year. If that’s true, then the latest Model S should have 20 percent more range than my old 85.

That hasn’t happened. The 90D I tested had an EPA range of 288 miles, only about 9 percent better than my 85’s EPA range of 265 miles when new. And the extra range comes at a premium of $3,000. Measured in cost per kWh,  that’s really no improvement at all.

But in the real world, the new 90D I tested has about 18 percent more range than my three-year-old 85.

2016 Tesla Model S 90D during Southern California test drive [photo: David Noland]

2016 Tesla Model S 90D during Southern California test drive [photo: David Noland]

As I mentioned, my usable battery capacity has declined from 75 kWh to 70 kWh, about 7 percent. When fully charged to 100 percent, my car now gives a rated range of 248 miles.

Just to see, I charged the 90D to 100 percent on the Oxnard Supercharger. The readout was an impressive 294 miles of rated range—46 more miles than my car.

That’s a significant chunk of miles—enough to eliminate the occasional mild range anxiety I feel on long legs between Superchargers in cold weather, uphill, and/or against strong headwinds. 

Efficiency

According to the EPA, the 90D is 12 percent more efficient than my 85. 

(The numbers: A 90D gets 100 MPGe and uses 34 kWh per 100 miles, while the 85 gets 89 MPGe and uses 38 kWh per 100 miles.)

But I didn’t see it. In fact, if anything, the 90D I tested seemed to use a bit more juice than my 85 typically does.

Over the 60 miles of driving I did in the 90D, I averaged 326 Watt-hours per mile. Under recent similar conditions, over roughly the same trip distances, my 85 used 305 Wh/mi—about 7 percent better  than the 90D.

In warm weather, my 85 has consistently averaged 290-300 Wh/mi in a variety of conditions. By the 12-percent  yardstick, the 90D should have been showing me 250-260 Wh/mi. Even in the Southern California sunshine, I never saw anything even approaching that.

Unfortunately, I didn't have the opportunity to drive both cars back-to-back, at the same speed, over a substantial distance. That’s the only way to compare them accurately.

But until then, I am skeptical about the 90D’s alleged efficiency.

And look at it this way: The 90D’s supposed 12-percent higher efficiency, combined with its 6-percent larger battery, should give it 19 percent more EPA range than the 85. 

2016 Tesla Model S 90D during Southern California test drive [photo: David Noland]

2016 Tesla Model S 90D during Southern California test drive [photo: David Noland]

The 85’s EPA range is 265 miles; thus the 90 should theoretically come in at a whopping 315 miles. Yet its official number is only 288 miles.

Something is fishy here. The numbers simply don’t add up.

70D efficiency: also doubtful?

This is the second time I’ve encountered lower-than-claimed efficiency in a dual-motor Tesla. 

In my test drive of a 70D model a year ago, I found it used virtually the same amount of energy as my 85, even though it was supposedly 13 percent more efficient.

Compared to the 60-kWh model it replaced, the 70D, with its 17 percent bigger battery and supposed 6-percent better efficiency, should theoretically have had 24 percent more range.

2015 Tesla Model S 70D, Apr 2015 [photo: David Noland]

2015 Tesla Model S 70D, Apr 2015 [photo: David Noland]

But in fact the range was only 15 percent better: 240 miles vs 208. If the 70D were as efficient as  the EPA claims, it would have a range of 258 miles. 

Is there something about the Tesla dual-motor setup that somehow “games” the EPA's efficiency rating, but not its range rating?

Little stuff

I noticed a number of other differences between my old 85 and the new 90D. Among them:

  • Motor Noise: While my 85 is virtually silent, one or both of the 90D’s dual motors gave off a faint but constant whine—more like a whistle—that increased with speed.
  • Windshield Washers: Instead of a narrow jet of fluid, the 90D sprayed a broad frothy plume. Much better.
  • Proximity Warnings: During close-quarter maneuvering, the 90D was constantly beeping at me and displaying direction and distance of potential obstacles. I found it mostly redundant and annoying. But a ding on my 85’s front bumper is evidence that perhaps I could have used the help on occasion.

2016 Tesla Model S 90D during Southern California test drive [photo: David Noland]

2016 Tesla Model S 90D during Southern California test drive [photo: David Noland]

  • Smaller Power/Regen Meter: This central feature of my 85’s instrument panel has been relegated to a tiny gauge off to the side, replaced by the autopilot visuals. I pay a lot of attention to the power/regen meter, and would dearly miss it.
  • Lane Departure Warning: Even with Autopilot off, changing lanes without signaling triggers an annoying vibration in the steering wheel. This drove me nuts almost immediately--if there's no traffic around, I see no need to signal for every  lane change.

Bottom line

Clearly, the 90D is a much better car than the 85: quicker, better handling, far better winter traction, and with 23 miles more EPA range.

And for most drivers, if not me, the Autopilot feature would be considered a major step forward. 

Compared to my specific car after three years, the real-world range advantage of the 90D climbs to 46 miles. And since mine doesn’t have the optional tech package, the 90D would include as standard equipment a couple of features missing from my car: turn-by-turn navigation and auto locking.

Yes, I’d rather be driving a 90D.

The trade-In math

Tesla hasn't yet gotten back to me with a trade-in value for my car, but I'd expect it to be right around $40,000.

A 90D equipped the way I’d want it—metallic paint and uprated seats, but no sunroof, autopilot, or air suspension—would list at $91,500. Counting the $7,500 tax credit, call it $84,000.

So we’re talking a cool $44,000 to upgrade. Sorry, not a chance.

2016 Tesla Model S 90D during Southern California test drive [photo: David Noland]

2016 Tesla Model S 90D during Southern California test drive [photo: David Noland]

How about a 70D, which would offer essentially the same acceleration and range as my 85, plus superior handling, far better winter traction, navigation and auto-locking?

With tax credit, a new 70D would run $72,000, or $32,000 to trade up. Closer, but still no cigar. 

My good old 85 still looks and runs like new, has fantastic acceleration, and plenty of range. It still turns heads everywhere I go, and it makes me feel good every time I climb into it. 

Sure, there are three or four days a year when I’d like better traction in snow. A few more miles range would be nice on my annual winter cross-country round trip.

And having to lock and unlock my car by hand suddenly became an annoying pain in the backside after my recent 24 hours with auto-lock.

But, hey, I’ll suck it up.  I fully expect to see 100,000 miles in my original car.  

Maybe by then Tesla will have launched its Model Y crossover, with all the Tesla range, acceleration, and style the brand stands for. 

In my mind, it's priced around $50,000, and it's built higher off the ground--with a wider door to accommodate my ever-creakier old bones.

Meanwhile, for the moment, I'm good with my old car. For now.

_______________________________________

Follow GreenCarReports on Facebook and Twitter.