Model 3 as a Student Driver car

[AquaMan]

Small update: Although I don't have a model 3 wiring diagram, I've learned that most cars with a 6-wire connector and drive-by-wire are internally, just two potentiometers. (Redundancy seems smart since potentiometers are unreliable.) So it seems very doable with 1-2 external relays. I'll know for sure when I get my Ebay accelerator pedal/sensor assembly.

Alan, I hear ya about the "weirdness" of the car. My intention is that until I have several cars, students can only book it for their final lesson, and they can choose an ICE car if they wish. Remember we get only 6 hours with most students (in California) so they gain the vast majority of experience at home.

Also my memory is fuzzy on this from my test drive, but I think I remember the turn signal stalk has unique behavior -- It doesn't "click in place" and I think it's "flip to signal, then flip in the SAME direction to cancel?" If that's true, I wish it could be canceled by flipping in the opposite derection, then flipping again if you want the opposite side indicator. That would certainly be more intuitive... Or I wish it was an on-screen setting.

I'm only a "potential future owner". Any current owners confirm about the turn indicator weirdness?

My experience differs from yours regarding the mirrors and I'm 6'1". What we teach is that a teensy bit sliver of the car should be visible in the side mirrors. I know there is a school of thought that the blind spot can be eliminated by moving them out. But trust this veteran driving instructor... it's just not true. Without a head turn, you're always missing something unless you have a convex mirror added, and those provide poor depth perception. Plus you can't pass a driving test without turning the head.

 

[AlanSubie4Life]

But trust this veteran driving instructor... it's just not true. Without a head turn, you're always missing something

I would agree with this. If you have them way out you can't see someone coming up fast who is not yet in the blind spot. Which may be worse than hitting someone in your blind spot (which is completely preventable in any case with a quick glance in that direction). In any case the range of adjustment is relatively limited compared to some cars.

It doesn't "click in place" and I think it's "flip to signal, then flip in the SAME direction to cancel?" If that's true, I wish it could be canceled by flipping in the opposite derection, then flipping again if you want the opposite side indicator. That would certainly be more intuitive... Or I wish it was an on-screen setting.

No. Two positions each way. Half way produces 3 flashes. Full pull leaves the signal on. You can cancel the signal with a half push in EITHER direction, regardless of which way you are currently signaling. Unless the "recent disengaged Autopilot while signaling" bug arises (it comes and goes), in which case the half push doesn't work to cancel the signal at all - you have to signal anew (I think in the opposite direction? Doesn't happen too often so hard to remember) and then cancel it.

 

[AquaMan]

No. Two positions each way. Half way produces 3 flashes. Full pull leaves the signal on. You can cancel the signal with a half push in EITHER direction, regardless of which way you are currently signaling.

Ah, thanks. It must have been the salesman who told me that, and I didn't move it far enough to observe the "more normal" behavior. I think autopilot was on when I used it. I always disable the "3-flash" feature whenever possible. It exists to sell cars, but it's never a safe way to signal and it confuses students.

 

[Big Earl]

I think driver’s ed in electric cars is great. I’d want to vary the vehicles a bit, as only learning in a Tesla will make switching to a conventional vehicle a little more difficult. Perhaps also add a Chevy Bolt to the fleet so they can get familiar with conventional controls.

One session in a fossil ought to be enough to impress upon them how bad they are.

 

[AquaMan]

Perhaps also add a Chevy Bolt to the fleet so they can get familiar with conventional controls.

Wish I could, but... range (among other concerns.) We need ~220 miles per workday, and even the LR RWD has barely enough if you consider that we'll charge to 85% for battery longevity and that it seems I should expect ~85% capacity after 400K miles. 325 x 85% x 85% = ~235 miles. Even the LR AWD would be right at the edge of what we need.

 

[Big Earl]

Wish I could, but... range (among other concerns.) We need ~220 miles per workday, and even the LR RWD has barely enough if you consider that we'll charge to 85% for battery longevity and that it seems I should expect ~85% capacity after 400K miles. 325 x 85% x 85% = ~235 miles. Even the LR AWD would be right at the edge of what we need.

Can you plug it in between driving sessions?

 

[AquaMan]

Can you plug it in between driving sessions?

Expensively, yes. Gotta pay employees for their time spent at a supercharger, and for the travel to get there. Plus, we like to have cars last 400K+ miles (Our ICE cars always have, believe it or not) and frequent use of superchargers isn't ideal for battery longevity.

 

[Big Earl]

Expensively, yes. Gotta pay employees for their time spent at a supercharger, and for the travel to get there. Plus, we like to have cars last 400K+ miles (Our ICE cars always have, believe it or not) and frequent use of superchargers isn't ideal for battery longevity.

I meant plug into level two chargers at your shop/office. Do you have a home base where you can charge? What’s the plan for overnight charging?

 

[AlanSubie4Life]

325 x 85% x 85% = ~235 miles.

You can charge to 90% probably. Tesla is totally fine with that (no one knows whether 80% vs. 90% will make any significant difference).

Forgetting about degradation for a moment...0.9*325rated miles = 292 rated miles. If you get lucky enough to get one of the ones with 325 rated miles of range (a lot of the LR RWD have 310 rated miles for whatever reason - though it's not clear that there is actually any difference in available energy between a 325 mile one and a 310 mile example, which is what matters).
Probably the LR RWD will be just enough. I realize it doesn't usually get too cold in Sacramento, but do keep in mind in winter it may be quite difficult to get even close to this. I imagine you spend a lot of time in driver training just sitting there (or driving slowly), so when heat is on that's going to have quite an impact.

I would assess what your average travel speed on a tank is for existing vehicles, in a day. Might be hard to get good data on this because cars probably only record the travel speed when in drive - not clear. Then we can tell you about how much range impact to expect (maybe). My concern is that your average speed may be quite slow and that heat may really add up. On the upside, you're in Sacramento so the occasions where heating is really significant will be quite rare. But it can get cold there in the winter occasionally.

If you have margin built into even that 220 mile number you'll probably be fine. And even having access to a few chargers (not Superchargers) during the day when transitioning between sessions (if you're going to people's homes that won't work of course) would get you there.

Basically I'm saying 220 miles in arbitrary conditions (rainy, cold, moving slowly, sitting idle a lot with heat on) actually could be a little tight even with LR RWD. Probably 9-10 months out of the year it would be no problem at all in Sacramento.

This problem is probably completely solved even with just a little bit of L2 charging during the workday if it's possible.

 

[67King]

5) I have already test driven, and I confirmed that while at freeway speeds if I start 2-foot driving with mild braking, it beeps but the brake and accelerator "fight" each other. (It slows, but not as much as brake alone.) This is what the ICE cars do also. What I didn't get to test is whether it will RAPIDLY stop when a student has fully depressed the accelerator.

I would suggest another test drive and pay a little bit more attention. When it beeps, it flashes up a message that reads "motor power removed." I tried this after reading your post. Basically, your concern about the torque of the motor overcoming the brakes is not warranted. The car will turn the power off to the motor when the brake is applied. Or, at least that is the case when sufficient brake force is applied when you would really need to be slowing the car

 

[Dana1]

Maybe r

Oops yes, that's 1.6 million miles. Still wish I had editing privileges.

Thanks for your input. I already have cheap cars of course. I've been doing this a while and there are many considerations when I choose a car, not all of which I've mentioned. I've considered other electric cars -- Not enough range for a full workday. I've considered Prius -- Poor driver visibility. I've considered just about every other hybrid -- Generally they add more to the price than they save in carbon-based fuel, they still require ICE maintenance, and they too have their share of blind spots and other issues. We're a high-end driving school, higher prices, retired law enforcement instructors, parents can ride along, etc. When I was a teen, I was trained by my P.E. teacher, who had no more driving knowledge than my parents, and there were 3 other kids in the back seat. That's not how we roll. I hope I've justified my reasoning for exploring this option, but if not I'm still looking for an answer.

So, to be clear: I have decided that this is the right car for us IF and only if the "simultaneous both pedals" issues can be effectively resolved by "electronically lifting the accelerator pedal" when brake lights are active. That's what I'm hoping this thread is about.

Maybe rent one on Turo to test your pedal bit and if you think it’s appropriate for your students.

 

[AquaMan]

I meant plug into level two chargers at your shop/office. Do you have a home base where you can charge? What’s the plan for overnight charging?

The plan is to charge overnight at the launch location. Mid-day charging just isn't cost effective even with level 2 chargers due to the need to pay employees for their time while charging, and to pay for their travel to charge. Our service area is 45 minutes of driving end-to-end, with the launchpoint in the center, so some days that could mean after an appointment at one end, they drive 20 min to a charger, plus 20 minutes charging, plus 20 minutes to the next appointment, which would adds an hour of payroll each day per employee. Plus every employee would get home much later, and they already get daily overtime.

Basically I'm saying 220 miles in arbitrary conditions (rainy, cold, moving slowly, sitting idle a lot with heat on) actually could be a little tight even with LR RWD.

I agree. Ultimately, this is the best car available for this purpose, and experimentation will answer all the questions. We MAY charge more on days we're expecting longer commutes, or students of various levels of skill (freeway vs. residential). We MAY find we can only use it for 3 lessons per day instead of 4, etc. We'll get a feel for it and find out what works.

I would suggest another test drive and pay a little bit more attention. When it beeps, it flashes up a message that reads "motor power removed." I tried this after reading your post.

I'll do that, and thanks for checking into it. I may have been wrong but I was specifically checking for this and it sure seemed that it was not slowing as fast as it would have without the accelerator pedal. Also since my original post I read somewhere that it only kills the motor after 3 seconds, which is much too long. (That's 264 feet at 60MPH.) Either way I'm now pretty confident I can "electronically lift the accelerator pedal" immediately when the brake lights come on, which I should probably do anyway.

 

[boriszima]

A very novel idea. Which cars have your company used that got to 400k miles? Tesla or any electric car will be cheap to maintain. Learning to drive in electric car would be cool, maybe will help with proliferation of BEV as most of us attest, once you drive it, you would not want to go to a smelly ICE.

 

[⚡️ELECTROMAN⚡️]

My daughter drove my S a lot when she was learning to drive, and I thought she did a great job. When she drove my wife’s icer, my wife was always yelling at her to slow down. I think being use to the regenerative breaking made her less likely to apply the brakes during normal driving, and she would not slow down soon enough while coasting the ice car. My wife eventually restricted her from driving my car to try to fix the problem, which I wasn’t going to argue with. I love driving my car and it’s definitely nerve racking having a rookie driver behind the wheel.

 

[AquaMan]

Which cars have your company used that got to 400k miles?

Off topic, but in our case, 2010 Nissan Versas. All four went over 400K. But that's not the complete answer. To make an ICE car last nearly half a million miles, here's what I personally think matters most:
1) Don't get a CVT transmission. Some of them are pretty good, but not "400K good". Unfortunately, with ICE cars, it's very hard to find a "traditional automatic" anymore.
2) Don't buy a car until its second or third year of production, or since its most recent redesign. They fix the design glitches during those first couple years.
3) Oil and fluid changes of course, full synthetic.
4) Beyond oil changes, with ICE cars the thing people often neglect is vibration. At every oil change, instead of JiffyLube, use a mechanic you know well, and task him or her to also "fully assess" the car, eliminating every shake and shimmy, which means replacing motor mounts, torsion mounts, and rubber suspension and steering "joints" long before it's absolutely necessary. A car which vibrates may not die due to the vibration itself, but the vibration will destroy every other part of the car, including wiring and hoses.

With EV's I'm excited about not only bypassing oil changes, belt maintenance and hose maintenance, but making it easier to avoid vibration. (But don't neglect vibration in the steering and suspension.)

 

[Darmie]

On one of the latest epsodes of the EV News Daily podcast, there was an interview with a UK company doing just that with an electric car. Can't remember the one being used. I do remember he ensured the training was geared towards learning to drive and not necessarily the difference of an electric vs fuel burning. He did mention that if the student wanted to they would pull up near a charger station just to get some experience but was not actually included in their normal training.

 

[AquaMan]

Update: Remember the goal is to "electronically lift" the accelerator pedal when the brake lights come on, regardless of the actual position of the pedal. This would be for safety in a driving school car where students often step on the "go pedal" while instructors are on the "stop pedal".

If you're not an electronics person, (and I'm no engineer myself) you may find this post boring.

So, to explore this option I received a used Model 3 accelerator pedal from Ebay. I put it on my multimeter, then when I found it doesn't behave as I expected, I took it apart. This isn't going to be as easy as I thought it would be, and it may be impossible.

I thought it would be two potentiometers, so I could just measure varying resistance among the 6 terminals as the pedal moved and duplicate the "fully lifted" state of the accelerator with one or two relays when the brake lights activates. Indeed, the terminals are labeled inside as Gnd1, Gnd2, Vref1, Vref2, Out1, and Out2, just as it would be for two potentiometers.

Well, that's not what's inside. There is a rotating triangular piece of metal which, through a layer of plastic, I believe interacts with the EM field on circular traces on a PC board. The PC board includes several very tiny surface-mount components including two chips. Like potentiometers, the PC board produces output voltages to be compared to a reference voltage and ground. Essentially it "looks" like two potentiometers to the car, but this is a more clever design. I don't know if it's a Tesla-exclusive, but it would definitely be more reliable than potentiometers as there are no "wipers" to wear out inside.

Although it still appears that the output is nothing more than a two voltages in reference to an input voltage (presumably 5 volts), just like potentiometers, the chips definitely do something. The pedal itself has QR codes on it, which implies the accelerator pedal has to be programmed for the car, like many Tesla components, although there is no terminal marked to produce a digital signal so I don't know how that would be so.

So, it seems it may still be possible to just treat it like two potentiometers and duplicate the appropriate output voltage on Out1 and Out2 for a "lifted state" when the brake lights come on, BUT... I don't know for sure.

Also, what is to be measured is voltage, not resistance... So it doesn't behave like potentiometers unless it has supplied power. Therefore without actually applying the appropriate voltage, I can't measure how much voltage should be output.

I could guess "zero volts" on one and "matching reference voltage" on the other, but that may not be a correct guess, and guessing wrong could cause the car to detect a fault condition.

To continue down this path, the next logical step is to measure the relative voltages on an installed pedal, using the car's own power. If the voltages turn out to be "100% and/or 0%" of the reference voltages when the pedal is fully lifted, this could still work. But if it's some other percentage of the reference voltage, I'd have to use external resistors or external potentiometers, either of which could be off by a bit and cause faults.

But, since this is all turning out to be a ton of work which may lead nowhere, the next goal is to appeal to Tesla. I NEED the "go pedal" to do absolutely nothing from the moment the brake lights come on, or this car may be unsafe. Unfortunately, that's not how it works from the factory... I observed there's a 3 second delay before the motors are shut off. (Why????) This could certainly be done with software, and I have no idea why they haven't done it that way originally. Or, it could be an optional setting.

 

[JasonR67]

A quick comment. You mentioned several times "When the brake lights come on" you want the accelerator to do nothing..

On Tesla's, and most / all EVs with regenerative braking the brake lights can come on even when the accelerator is pressed. Disabling the accelerator would make it a dead pedal until regenerative braking turned off ~8 mph on level ground. This would NOT be ideal.

 

[AquaMan]

Disabling the accelerator would make it a dead pedal until regenerative braking turned off ~8 mph on level ground. This would NOT be ideal.

I'm aware the brake lights can come on when the accelerator is depressed, even if the brake pedal is not depressed. Sorry, I'm failing to understand what you're saying would occur as it decelerates past ~8mph? I THINK you're saying the following sequence could occur...

Accelerator pedal is down say 40%, cruising along.
Accelerator pedal is released to ~10% (or whatever) which turns on brake lights.
Immediately the accelerator pedal would be "electronically released" to 0%.
This would induce abruptly harsher regenerative braking.

If that's what you're saying, I get it, and agree. In truth what I need is to disable the accelerator instantaneously when the passenger-side (instructor) brake is depressed, which would be configured to pull the driver-side pedal down just as it does in all existing student driver vehicles. Admittedly, I referred to using the brake lights as "shorthand", (and it was my initial thought) but in truth and after some thought I realize a switch would have to exist on the passenger-side pedal itself, so this would occur only when an instructor wishes to brake. In such a case, it should ALWAYS be appropriate to fully release the accelerator pedal.

Still, it seems to me Tesla should always treat the accelerator pedal as fully lifted if the brake pedal switch is activated. (For all drivers.) I still don't know why they've added a delay before that happens. 3 seconds is easily enough time to crash.

 

[JasonR67]

I THINK you're saying the following sequence could occur...

Accelerator pedal is down say 40%, cruising along.
Accelerator pedal is released to ~10% (or whatever) which turns on brake lights.
Immediately the accelerator pedal would be "electronically released" to 0%.
This would induce abruptly harsher regenerative braking.

If that's what you're saying, I get it, and agree.

yes, that was my concern. Sounds like you have taken that into consideration.