Get Amped Tour: Chicago, 7/26 - 7/28

[Larry Chanin]

Does this help? "In order to simulate stopping the car with a stuck accelerator, my usual test procedure is to start by flooring the accelerator with my right foot, accelerate to 100 km/h and then apply the brakes firmly with my left foot while holding the accelerator down. "

Perfect. Great work.

Larry

 

[Norbert]

Agreed, that's easily understandable (though that's always a subjective thing... )

Thanks!

 

[favo]

In order to simulate stopping the car with a stuck accelerator, my usual test procedure is to start by flooring the accelerator with my right foot, accelerate to 100 km/h and then apply the brakes firmly with my left foot while holding the accelerator down.

Judging from Tesla's letter to NHSTA, the car should immediately start slowing.

[T]he Model S is equipped with brake override where the controller gives priority to brake pedal input over accelerator input. If the brake pedal is pressed after the accelerator, the controller will ignore the accelerator inputs and respond only to the brake signal.

I assume when they say "brake pedal is pressed after the accelerator," that covers your test of holding the accelerator down while subsequently and simultaneously pressing the brake with your other foot. Seems like it is not working the way the said it would. Is it possible you let up on the accelerator, pressed the brake, and then re-pressed down on the accelerator?

I'm guessing the following

Thank you, you helped me find it...
"The customer claimed that the 'throttle misbehaved'. Tesla looked up the logs and told the customer that 'you had pressed the throttle and the brake at the SAME time for longer than 3 seconds, so we automatically disengaged the throttle. That is why your speed was only 15 mph when the incident took place and not 55 mph'. "

interesting story from test drive at factory

So I retract my earlier comment about 3 seconds... This doesn't explicitly say you need to press both pedals for 3 seconds for the accel/brake override to engage.

describes a kind of hill-hold-style functionality that only applies when you have the brake pressed down first, and, while it's still pressed, you then also press the accelerator. This allows a couple of seconds for you to remove your foot from the brake pedal and not slide back. (Seems like 1 or 2 seconds would be enough.)

Tesla's letter goes on to give two other options to deal with a runaway vehicle. Might be interesting to see if these work as advertised on a test drive. Check with your Tesla minder first, though.

In addition, the Model S transmission is designed to allow shifting from drive or reverse into neutral while the vehicle is in motion. Shifting from drive or reverse into neutral will remove torque from the motor. Finally, all Model S vehicles will be equipped with a park brake capable of dynamic application. When the park brake button on the center screen is depressed, regardless of vehicle speed, the brake is applied slowing the vehicle.

I presume that in the latter case (park brake pressed), it also removes torque from the motor.

Mods: maybe this should get its own thread, since it's a pretty serious safety issue.

 

[RichSh]

Conclusion
I went to the drive expecting and hoping to be wowed, but was not. Maybe this is because the Volt is already so much better than most other cars that the Model S cannot be that much better. Or maybe the Volt is just a better fit to my needs and expectations. The Model S is a much bigger car than I need for daily single occupant commuting. I will keep my reservation and re-evaluate when the time to configure comes in the hope that the failed items are corrected and the untested items pass. Other items that could affect my decision include adding creep, adding a poorly implemented hill hold, interior lighting changes, lack of front proximity sensors and collision avoidance.

As others have said, the Model S is simply too expensive a car to use for one, at most two hours a day and not meet all my expectations.

J, thanks for the very detailed review. Amazing the the number of things you were able to think of and measure for the brief amount of time we had in the cars.

Curious if you already own a Volt, or are simply considering it as an alternative to the Model S? We brought my wife's Volt to my test drive of Model S on Saturday, and see the 2 cars as very much complementary of one another for our needs. Will be nice to have 2 cars capable of short range electric driving, Model S for mid-range electric, and the Volt's ICE generator for very long trips. My wife tends to drive more conservatively, and does not require the added performance of Model S. Still, both of us were wowed (even giddy) from test driving Model S in comparison to the Volt--and by things you noticed--differences in acceleration, smoothness with the air suspension, the lower center of gravity on curves all lead to a far superior driving experience when compared against what is an already a superior experience in the Volt. I commented that Model S was like the Volt on steroids. My wife also found the roller-coaster-like experience of Model S to be far superior. Running the course in both the Volt and the Model S only helped to cement those views. On top of the performance differences, the opportunity to have a pure electric car, with the huge screen, panoramic roof, and much greater electric range easily make the value proposition of Model S understandable in my eyes. The ability of Model S to use the HD backup camera when moving forward helps address concerns with rear visibility. Many of the minor flaws that people see in Model S are far too small to keep me from wanting the car when the total package is considered. Tesla could still address many of those issues before we take ownership (or later via firmware updates, upgrades at annual service, etc.). My main remaining concern is the potential cost of service rangers for annual mantenance living in a smaller market and/or the logistics of making the long drive to the nearest service location. If you prefer a smaller car and really like the Volt, then perhaps the Volt is right for you, or perhaps a Volt lease followed by a Tesla Gen III purchase. It's great to see US automakers providing us with these superior models at this time.

 

[qwk]

So here is my review of the Model S silver non-performance car I drove in Chicago. This review is presented in terms of whether it passes or fails my expectations, particularly with respect to the Volt and in areas that haven’t been extensively covered by others. To restate the obvious, the amount of time allowed was sufficient for a drive, but not for a test drive, yet, I tried my best.

Pass
1. Accelerator pedal mapping: Accelerator pedal mapping is done well and allows for good control at low speed.

2. Turn radius and steering feel: My perception of the turn radius agrees with the published 37 ft. While this is more than the Volt’s 36 ft, I give it a pass considering its monster size. Steering feedback was good enough for me and I can see myself using both the comfort and sport power assist settings at different times.

3. Turn signal blip: The turn signal flashes three times after a blip of the stalk.

4. Max acceleration into a right turn from a full stop: Chassis stiffness, suspension, tyres and traction control all need to work together to make this successful. The result was on par with the Volt. The Model S has less chirp from the inside front tyre, but probably only because it is rear wheel drive, while the Volt is front wheel drive. Considering that the Model S is much bigger and heavier than the Volt, I give this test a pass.

5. Cruise control: The cruise control is really well done. When on, it displays the speed set point as a bug on the outside of the speedometer circle, just like the bugs on the IAS or HSI on an aeroplane. Resume and increment/decrement reacts immediately and adjusts the speed assertively without any undershoot or overshoot. The increments are 1 mph (see more below).

6. Acceleration: 0 to 60 mph came in 6.5 seconds and 0 to 100 km/h in 6.8 s. However, as can been seen from the Graph 1 below, the motor overpowered the brakes when applying power against the brakes, causing the timer to trigger prematurely by not more than 0.3 s. So, 0-60 mph in 6.2 s and 0-100 km/h in 6.5 s are probably more accurate. There were four people on board with an estimated mass of 290 kg. With only an 80 kg driver on board, we can expect 0-60 mph in 5.6 to 5.7 s which is close enough to the claimed 5.6 s. Acceleration in the peak torque band was 0.47 g compared to 0.39 g (two occupants 140 kg) or 0.40 g (single occupant 85 kg) in the Volt. While this difference is noticeable, it is not dramatically better, and neither my wife nor I said “wow” or “oh my god”. The important difference, though, is that the Volt starts losing torque at 50 km/h, while the Model S keeps going until 76 km/h (according to the specs page, and confirmed here). At 100 km/h, the Volt has only 0.15 g acceleration, while the Model S still has more than 0.3 g. This will allow highway manoeuvres with much more confidence.

7. Air suspension: The air suspension really does make a difference to the ride quality. I also suspect that it contributes greatly to the quiet interior.

Undecided
1. Regeneration: I was hoping that the Model S would have stronger regen braking than the Volt. What I felt during the drive was that it was initially weaker than the Volt, but then became stronger as the car slowed. Looking at the graph below, this is confirmed. From 65 to 45 mph it was 0.13 g and limited by the 60 kW charge limit. Below that speed it is a constant 0.15 g. The Volt starts at 0.15 g and ramps down as it slows. I prefer the constant deceleration of the Model S, but would’ve preferred it to be at least 0.20 g.

2. Immediate access to air recirculation: This is a big irritation in the Volt. It has 37 buttons on the centre panel for everything under the sun, except for recirculate, which is the one thing that you need immediate access to when driving through a cloud of dust or behind a smoke billowing truck. The Prius has a recirculate button right on the steering wheel, which is perfect. The Model S does not have a recirculate touchon in the A/C controls area, but it does have an A/C on/off touchon, which does turn off the fan. I suppose this is better than nothing.

Failed
1. Door close action: I have written before (Nissan Leaf) that the Volt has the best door close action of any car I have interacted with. The doors will close themselves after just a slight nudge out of the hold detent. The model S has an extra hold detent close to the closed position that effectively “catches” the door before it closes. And there isn’t enough spring action left to close it after moving it out of this detent. After all the fuss and bother from Franz about the door handles, I am quite upset that they couldn’t get the door close right. Big Fail.

2. Displays in SI units: The firmware in the beta unit in the store has an option to display distance in miles or km. However, nothing visible happens on the screen. On the test car the option was not in the same place, and the co-pilot did not know and could not find a place to change either distance or speed.

3. Tyre pressure display: The co-pilot did not know and could not find how to display the tyre pressures on the centre screen.

4. Regeneration cut-out speed: In the Volt regen stops at 5 km/h. On the Model S, it was between 7 and 6 mph (see above) so, probably at 10 km/h.

5. Rear visibility: The hatch window looks really small in the rear view mirror. The Volt’s is already small compared to the Prius, but the Model S is almost not worth the effort. I also found that the driver side mirror does not adjust far enough out to cover the 7 to 9 o’clock area.

6. Brake override: See why (Toyota Settles Over Death of Family in High-Speed Crash - NYTimes.com), and how (About Toyota | Our News | Toyota Evaluates Unintended Acceleration Complaints in Remedied Vehicles). GM also (GM to Expand Brake Override Software Globally by 2012). In order to simulate stopping the car with a stuck accelerator, my usual test procedure is to start by flooring the accelerator with my right foot, accelerate to 100 km/h and then apply the brakes firmly with my left foot while holding the accelerator down. Even in cars without brake override, the brakes are strong enough to bring the car to a controlled stop pretty quickly. The Model S failed this test in the most unexpected and unusual way. The test route didn’t have enough space to perform this at 100 km/h, so I started at about 70 km/h / 40 mph. The brakes prevented further acceleration, but did not slow the car. I was about to abort the test, when it slowly started to decelerate, so I kept going. This was probably due to better friction from the pads as they warmed up. Then, at somewhere between 30 and 20 mph, and 3 to 5 seconds after braking, the car suddenly came to a violent stop. This startled me and upset the co-pilot tremendously! It is entirely possible that I didn’t push the brakes hard enough, but there wasn’t enough space or time to try again and the co-pilot certainly wasn’t in the mood. However, I would then argue that the force required for maximum braking is too high, and that the abrupt stop at the end indicates that I did push hard enough. So, it appears that the brakes are not powerful enough, and if there is a motor power reduction, it takes far, far too long to trigger (should be < 0.5 s). Huge Fail. I’ve been encouraged to, and will try to get an explanation of this behaviour from HQ. (Update: With Tesla Engineering as of Aug 1)

Untested
1. Sun visor extension: I wanted to check that the sun visor swivel arm can extend so that the visor can reach the B-pillar, but fussing over the speed display distracted me and took too much time.

2. Interior noise level: The intention was to measure A-weighted and C-weighted noise at 80 and 100 km/h, but the practice run revealed that there would be time for only one measurement. Unfortunately operator error resulted in no measurement taken. For reference, the Volt noise levels are: 59 dB-A and 81 dB-C at 80 km/h and 61 dB-A and 82 dB-C at 100 km/h on new asphalt. My perception was that the Model S is slightly quieter in the C-weighted range and significantly quieter in the A-weighted range. I suspect the air suspension has a lot to do with that.

3. Cruise control increment in km/h: This test was not performed because we could not get the speed display in km/h. Pass would be if the cruise control increments in 1 km/h steps if the speed display is in km/h and fail would be if it increments in 1.6 km/h steps.

4. Ease of parking: I am concerned about the width of the car when parking (particularly in narrow California parking spaces). Unfortunately all the spots on the approach to the staging area were taken.

Conclusion
I went to the drive expecting and hoping to be wowed, but was not. Maybe this is because the Volt is already so much better than most other cars that the Model S cannot be that much better. Or maybe the Volt is just a better fit to my needs and expectations. The Model S is a much bigger car than I need for daily single occupant commuting. I will keep my reservation and re-evaluate when the time to configure comes in the hope that the failed items are corrected and the untested items pass. Other items that could affect my decision include adding creep, adding a poorly implemented hill hold, interior lighting changes, lack of front proximity sensors and collision avoidance.

As others have said, the Model S is simply too expensive a car to use for one, at most two hours a day and not meet all my expectations.

View attachment 8315
Graph 1: Model S Acceleration. Non-performance. Standard regen. 4 occupants.

View attachment 8316
Graph 2: Volt Acceleration. Normal mode, L regen. 2 occupants.

View attachment 8317
Graph 3: Volt Acceleration. Normal mode, L regen. 1 occupant.

I simply don't understand how you can say that the volt is on par with the model S, especially on things like acceleration. I have driven both cars and I would say that comparing the volt to the model S is like comparing a geo metro to a Bentley continental. There simply is no comparison. The Volt has horrible throttle response and is almost 4 seconds slower in the quarter mile. In drag racing that is a lifetime. I can appreciate different point of views, but this is as bias of a comparison as you can get, sorry.

 

[Zythryn]

I haven't had the pleasure of a test drive. I very much look forward to getting the chance to compare the two, as those will be our two cars.
I agree with the previous poster that mentions they compliment each other very nicely.

 

[brianman]

I simply don't understand how you can say that the volt is on par with the model S, especially on things like acceleration.

Perhaps he's going by the shape of the charts, not observing that the X axis (time) is different between the two.

 

[J in MN]

Is it possible you let up on the accelerator, pressed the brake, and then re-pressed down on the accelerator?

No.

Shifting from drive or reverse into neutral will remove torque from the motor.

Yep, tested that while in D. Any movement of the stalk makes it go to N, and it will not go to R while moving forward.

I presume that in the latter case (park brake pressed), it also removes torque from the motor.

I am not excited about using the parking brake as mitigation. If you are going 125 mph, are you really going to take your eyes off the road to search through the menus to find the parking brake touchon on the centre screen? And, if 4 large sets of brakes could not slow the car significantly, how will 2 small sets do any better? Unless, as you say, it also cuts motor torque.

Mods: maybe this should get its own thread, since it's a pretty serious safety issue.

I don't think we should make too much of it until after I hear back from Tesla Engineering. The way I see it, the behaviour can be changed easily through an OTA software update - it is not a difficult to fix, recall inducing design flaw.

Curious if you already own a Volt

Yes, we have since April 2011. It replaced a Corolla.

We … see the 2 cars as very much complementary of one another for our needs. ...and much greater electric range easily make the value proposition of Model S understandable in my eyes.

I absolutely agree. The Model S is intended to replace a Prius, and eventually a Bluestar/Gen III will replace the Volt. If family came to visit more often (they think the Twin Cities are too boring), or if we had children (unfortunately we don't), or if either of us had to make frequent trips longer than 50 miles, the Model S would've been a no-brainer. Absent that, I was hoping for something more spectacular about the driving experience to convince me to part with my money.

I can appreciate different point of views, but this is as bias of a comparison as you can get, sorry.

You quoted my entire posting, but did you read it all?

 

[favo]

Yep, tested that while in D. Any movement of the stalk makes it go to N, and it will not go to R while moving forward.

I guess you can then go back into D, but not R (unless you come to a stop first).

I am not excited about using the parking brake as mitigation. If you are going 125 mph, are you really going to take your eyes off the road to search through the menus to find the parking brake touchon on the centre screen? And, if 4 large sets of brakes could not slow the car significantly, how will 2 small sets do any better? Unless, as you say, it also cuts motor torque.

Yeah, clearly not a first option, but redundancy is good.

I don't think we should make too much of it until after I hear back from Tesla Engineering. The way I see it, the behaviour can be changed easily through an OTA software update - it is not a difficult to fix, recall inducing design flaw.

I'll be interested to hear the response. Thanks for following up.

 

[Brian H]

...
Now we are wrestling between Pearl White (her) and Brown (me). Unfortunately she has a negative association with brown due to her work in the medical field...

Don't win the argument. She'll probably dub it the Fecal Flivver!! :scared: :-D

 

[vfx]

Our copilot suggested a two foot start. Holding L foot on brake and stepping R foot on the accelerator to preload the motor. Then rapidly letting off the brake for a "launch" acceleration. In the Roadster I have never found that to be faster and we have been told many times it does not harm the car.

Question is, why would not the car shut down with both pedals depressed? Do you have to launch in under 3 seconds?
ETA: Answered above^^^^(which pedal is first counts)

BTW, your review is amazing detailed. Makes me feel like a troglodyte driver. And also that no car could ever meet all such metrics combined with styling and fun. Analyzing a door closure is so far out of my thought process that it might as well be counting grains of sand on a beach. I'm sure there is someone in a car's design team that looses sleep about the door closing as I obsess about my work (and play) on occasion but how many typical drivers think of such things? I would guess none until it was pointed out.

J, Tesla should hire you for a week to go over the car with the fine toothed comb you have in your back pocket.

 

[gg_got_a_tesla]

Analyzing a door closure is so far out of my thought process that it might as well be counting grains of sand on a beach. I'm sure there is someone in a car's design team that looses sleep about the door closing as I obsess about my work (and play) on occasion but how many typical drivers think of such things? I would guess none until it was pointed out.

Yeah, that's what I thought too when the DVD that came with my BMW 325i's manual showed a video of the time and effort BMW's sound engineers put in to get the various "bongs" in the car just right and how they get more 'playful' in a Mini vs in a BMW

 

[tnawara]

Photos of Chicago Cars' VINs

Sorry that these took so long to put up, but I went on vacation immediately following the Chicago Get Amped stop. Here are some images of the VIN plates on each of the four cars that were present in Oak Brook:

Silver Model S - VIN 5YJSA1AN1CFR00112

Pacific Blue Model S - VIN 5YJSA1AN9CFR00102

Dolphin Grey Model S Performance - VIN 5YJSA1AP6CFR00107

Black Model S Performance - VIN 5YJSA1AP0CFR00104

 

[dsm363]

So here is my review of the Model S silver non-performance car I drove in Chicago. This review is presented in terms of whether it passes or fails my expectations, particularly with respect to the Volt and in areas that haven’t been extensively covered by others. To restate the obvious, the amount of time allowed was sufficient for a drive, but not for a test drive, yet, I tried my best.

Failed
1. Door close action: I have written before (Nissan Leaf) that the Volt has the best door close action of any car I have interacted with. The doors will close themselves after just a slight nudge out of the hold detent. The model S has an extra hold detent close to the closed position that effectively “catches” the door before it closes. And there isn’t enough spring action left to close it after moving it out of this detent. After all the fuss and bother from Franz about the door handles, I am quite upset that they couldn’t get the door close right. Big Fail.

2. Displays in SI units: The firmware in the beta unit in the store has an option to display distance in miles or km. However, nothing visible happens on the screen. On the test car the option was not in the same place, and the co-pilot did not know and could not find a place to change either distance or speed.

3. Tyre pressure display: The co-pilot did not know and could not find how to display the tyre pressures on the centre screen.

4. Regeneration cut-out speed: In the Volt regen stops at 5 km/h. On the Model S, it was between 7 and 6 mph (see above) so, probably at 10 km/h.

5. Rear visibility: The hatch window looks really small in the rear view mirror. The Volt’s is already small compared to the Prius, but the Model S is almost not worth the effort. I also found that the driver side mirror does not adjust far enough out to cover the 7 to 9 o’clock area.

6. Brake override: See why (Toyota Settles Over Death of Family in High-Speed Crash - NYTimes.com), and how (About Toyota | Our News | Toyota Evaluates Unintended Acceleration Complaints in Remedied Vehicles). GM also (GM to Expand Brake Override Software Globally by 2012). In order to simulate stopping the car with a stuck accelerator, my usual test procedure is to start by flooring the accelerator with my right foot, accelerate to 100 km/h and then apply the brakes firmly with my left foot while holding the accelerator down. Even in cars without brake override, the brakes are strong enough to bring the car to a controlled stop pretty quickly. The Model S failed this test in the most unexpected and unusual way. The test route didn’t have enough space to perform this at 100 km/h, so I started at about 70 km/h / 40 mph. The brakes prevented further acceleration, but did not slow the car. I was about to abort the test, when it slowly started to decelerate, so I kept going. This was probably due to better friction from the pads as they warmed up. Then, at somewhere between 30 and 20 mph, and 3 to 5 seconds after braking, the car suddenly came to a violent stop. This startled me and upset the co-pilot tremendously! It is entirely possible that I didn’t push the brakes hard enough, but there wasn’t enough space or time to try again and the co-pilot certainly wasn’t in the mood. However, I would then argue that the force required for maximum braking is too high, and that the abrupt stop at the end indicates that I did push hard enough. So, it appears that the brakes are not powerful enough, and if there is a motor power reduction, it takes far, far too long to trigger (should be < 0.5 s). Huge Fail. I’ve been encouraged to, and will try to get an explanation of this behaviour from HQ. (Update: With Tesla Engineering as of Aug 1)

Untested
1. Sun visor extension: I wanted to check that the sun visor swivel arm can extend so that the visor can reach the B-pillar, but fussing over the speed display distracted me and took too much time.

2. Interior noise level: The intention was to measure A-weighted and C-weighted noise at 80 and 100 km/h, but the practice run revealed that there would be time for only one measurement. Unfortunately operator error resulted in no measurement taken. For reference, the Volt noise levels are: 59 dB-A and 81 dB-C at 80 km/h and 61 dB-A and 82 dB-C at 100 km/h on new asphalt. My perception was that the Model S is slightly quieter in the C-weighted range and significantly quieter in the A-weighted range. I suspect the air suspension has a lot to do with that.

3. Cruise control increment in km/h: This test was not performed because we could not get the speed display in km/h. Pass would be if the cruise control increments in 1 km/h steps if the speed display is in km/h and fail would be if it increments in 1.6 km/h steps.

4. Ease of parking: I am concerned about the width of the car when parking (particularly in narrow California parking spaces). Unfortunately all the spots on the approach to the staging area were taken.

Conclusion
I went to the drive expecting and hoping to be wowed, but was not. Maybe this is because the Volt is already so much better than most other cars that the Model S cannot be that much better. Or maybe the Volt is just a better fit to my needs and expectations. The Model S is a much bigger car than I need for daily single occupant commuting. I will keep my reservation and re-evaluate when the time to configure comes in the hope that the failed items are corrected and the untested items pass. Other items that could affect my decision include adding creep, adding a poorly implemented hill hold, interior lighting changes, lack of front proximity sensors and collision avoidance.

As others have said, the Model S is simply too expensive a car to use for one, at most two hours a day and not meet all my expectations.

I have to say for me at least the things you consider not up to par are very minute and probably fixable by Tesla fairly easily. I'm not sure how the door close action would negate getting this car but that's a personal opinion. #2-4 seem minor or fixable by software. The rear visibility does seem limited but think driving with the rear view camera on may help. Unintended acceleration seems so far down on the list of possible things that could happen in a car but I've never even thought to test this but if the difference between 0.5 seconds and 3 seconds is a deal breaker, maybe Tesla could revise this in the future with software.

You have to be happy with the car you are buying of course though.

 

[brianman]

Unattended acceleration

I did a double take on this one. I had flashes of (Arnold's) Total Recall.