P mode vs Park mode

[Mario Veras]

What’s the difference between putting the Model 3 in P versus Park mode? I’ve seen some YouTubers talk about that it makes no difference, so why is there a logo displayed on the screen if it makes no difference?
I can’t seem to find this in the manual.

 

[DownshiftDre]

Superbowl weekend. Any updates? Feels like almost there with data backed conclusion. This thread always pops up in my mind whenever I push P. Today parked car in my driveway with slight incline and held P just in case there is additional clamping force lol

 

[gearchruncher]

Had a race last weekend that I had to put the car back together for, and out of town all week. Bought some new cables that can handle more current. Still on my list to get to the bottom of this with data. (or someone else can do it with a "simple voltmeter" )

 

[Haggy]

That noise does not mean its applying any new process or additional force. Much like electronic locks on a car.

Lock the doors (car other than Tesla) once and they make that classic "cla-chunk" sound. Lock them a second time and you hear a soft "click". That doesnt mean you car is extra locked or any more safe. It just means an event was triggered a second time and because the locks were locked, you hear relays and/or the attempt to lock them again. It's trying to move the lock mechanism but cannot.

The noise you hear is likely just the process attempting to move the brake actuator but not actually doing so.

The problem with that notion is that it's not analogous. If I lock the doors on another car, and then push the lock button, I will hear a sound as you described. If I do it again, I'll hear the same sound again. That would keep happening. With a Tesla, if I push the end of the stalk, I hear a screwing sound. The manual even said that tapping the end of the stalk applies the parking brake. But if I press and hold, I hear another sound, and the icon goes on. And if I press and hold for a second time, I hear nothing. If it were merely trying to activate something, and clicking because it was already applied, it would do that if I repeated pressing and holding. Also, the car knows that it was already in Park, so there would have been no reason to do anything and even less reason not to do something if it's repeated, if it accomplished nothing.

The obvious problems are that the manual said explicitly that the parking brakes are applied automatically when the car is in park. But the latest online manual also says explicitly that you can apply the parking brakes by pressing and holding the button while in park. They also changed the section in the user manual about the parking icon to include its illumination when the press and hold is done.

So that leaves me with the impression that it does something, especially since I can hear it doing something, it says it does something, and the manual says that it does something. But it also doesn't answer four fundamental questions. Why would it need to be done? What does it do that's different? Why doesn't the car do the same thing automatically whenever the end of the stalk is tapped while the car is stopped? If there's a reason not to do it each time, what's the disadvantage of doing so?


Tesla did something for a reason. They wouldn't pay people to change the function and update the manual unless it did something. But it's also absurd to tell people that the parking brake is applied automatically when the car is put in park, and that a person can apply the parking brake while the car is in park.

 

[gearchruncher]

Tesla did something for a reason. They wouldn't pay people to change the function and update the manual unless it did something. But it's also absurd to tell people that the parking brake is applied automatically when the car is put in park, and that a person can apply the parking brake while the car is in park.

It's not absurd if there is a FVMSS that requires a way for the driver to manually engage the parking brake at all times.
Most new cars have electric parking brakes. They mostly apply them when you go into park. They all also have a manual button to do it, and the button lights up when you manually press it, even if you're in park.

The Tesla system isn't any different. Sometimes stuff is just there to comply with a rule that may not cover a modern system well.

 

[Haggy]

It's not absurd if there is a FVMSS that requires a way for the driver to manually engage the parking brake at all times.
Most new cars have electric parking brakes. They mostly apply them when you go into park. They all also have a manual button to do it, and the button lights up when you manually press it, even if you're in park.

The Tesla system isn't any different. Sometimes stuff is just there to comply with a rule that may not cover a modern system well.

In that case, I would think that it would meet requirements if it checked to see if the parking brake was on, put it on if it isn't, and then put on the icon if it's a requirement. According to the documentation, the parking brake is already on when it's in park, and it had a function for emergency if the driver holds it in.

I'm not saying that there's no reason, but Tesla could tell customers when and why they might use it. Also, I never said that it was absurd. The overall discussion, here and elsewhere, revolves around whether it does anything at all. I no longer see a compelling argument that it does nothing but put the light on, and the discussion is more academic than practical unless someone can figure out a benefit.

 

[DownshiftDre]

Life happened and no additional data was collected. There were some very interesting traces provided in the page prior. I think we left this inconclusive but it always crosses my mind when I park on slight downhill slope driveway.

 

[LiamPope]

In that case, I would think that it would meet requirements if it checked to see if the parking brake was on, put it on if it isn't

And what more simple and reliable way to do that than to initiate the command to drive the motor to stall stop one more time? Which is exactly what it does.

As an electrical engineer I had a blast reading this thread. But what a rollercoaster of emotions when scope traces were posted, but the ones we really wanted to see never materialised!

I still wonder what was behind the talk of the differing sensor signals? It did suggest that maybe some clamping force feedback signal is derived within the controller and potentially used to stop the motor at a certain force. Normal park mode triggering at 1.7 V, but 'superpark' mode triggering at 4.5 V, i.e. higher clamping force.

But I have no idea. Seems more likely they would be the same and the push and hold 'superpark' is simply the manual input complying with e-brake/handbrake system regs. The only mode of operation I'd expect to have some intelligence and variable behavior is operation while the car is moving.

That said, I'm also very much a mechanical engineer into mechanical testing and this would be my preferred way to determine for sure! All we need is a friction modifier to bring the force required to defeat the brakes into easily measurable range! Is anyone volunteering to oil their brakes? 3D print some grooved nylon brake pads? If only we had a drum-based handbrake mechanism like many of my older cars - it'd be so weak we could do this test easily without such modifications!

 

[Haggy]

And what more simple and reliable way to do that than to initiate the command to drive the motor to stall stop one more time? Which is exactly what it does.

As an electrical engineer I had a blast reading this thread. But what a rollercoaster of emotions when scope traces were posted, but the ones we really wanted to see never materialised!

I still wonder what was behind the talk of the differing sensor signals? It did suggest that maybe some clamping force feedback signal is derived within the controller and potentially used to stop the motor at a certain force. Normal park mode triggering at 1.7 V, but 'superpark' mode triggering at 4.5 V, i.e. higher clamping force.

But I have no idea. Seems more likely they would be the same and the push and hold 'superpark' is simply the manual input complying with e-brake/handbrake system regs. The only mode of operation I'd expect to have some intelligence and variable behavior is operation while the car is moving.

That said, I'm also very much a mechanical engineer into mechanical testing and this would be my preferred way to determine for sure! All we need is a friction modifier to bring the force required to defeat the brakes into easily measurable range! Is anyone volunteering to oil their brakes? 3D print some grooved nylon brake pads? If only we had a drum-based handbrake mechanism like many of my older cars - it'd be so weak we could do this test easily without such modifications!

Then explain why pressing and holding the stalk would cause it to do that to the drive motor one more time, merely tapping the end of the stalk would not, and if a person does press and hold it, a subsequent press and hold would do nothing? If it wants to double check, why not do it with a simple tap? And why would I hear the drive motor actually turn as if it's tightening? I know that some people reported merely hearing a click, but I heard more than that, and it was more pronounced on my Model S than on my Model 3. It was unquestionably moving something.

Of course this has nothing at all to do with electrical engineering. It's a question of why Tesla would go out of their way to design something in a way that made no sense, given that they had a working design, and given that the car would already know that it's in park. But the cars are acting as if they know that they are in park by doing nothing if the stalk is tapped again, and know that they are in whatever state you want to call it if it's held in for a second time. In both instances, the car won't do anything if it's already been done and the car knows that it's been done.

If the theory that "it's clicking because it's merely checking and there's nothing to move" were true, then I'd expect a click from pressing and holding for a second time, which does not happen. The bottom line is that in both cases, it does something only the first time, and in neither case does it "try again."

 

[Gauss Guzzler]

As @gearchruncher noted, the normal operation reduces the effective voltage by using pulsed DC at around 80% duty cycle in the example on page 6. The system might vary this duty cycle for consistent engagement/retraction speed, or it might just use the same duty cycle for everyone and simply trigger a warning if the motor speed ends up out of bounds. I'd guess the latter.

The "Parking brake" function likely does a very quick check with pulsed DC to verify that the motor is already stalled, followed by a solid DC pulse that is long enough to deliver maximum current (~18A) for maximum force, but short enough to protect against overheating. This would explain the very loud and distinct "click" that is heard upon engagement.

 

[spad]

Instead of measuring input signal, get a load sensor to measure the “real” applied force would be easier

 

[gearchruncher]

Happy to do so if you can point me to a sensor that can measure applied pressure of the brake pads and costs under $50.
I'd expect finding a load cell that is just the right thickness would be hard and they are really expensive generally.

 

[Haggy]

As @gearchruncher noted, the normal operation reduces the effective voltage by using pulsed DC at around 80% duty cycle in the example on page 6. The system might vary this duty cycle for consistent engagement/retraction speed, or it might just use the same duty cycle for everyone and simply trigger a warning if the motor speed ends up out of bounds. I'd guess the latter.

The "Parking brake" function likely does a very quick check with pulsed DC to verify that the motor is already stalled, followed by a solid DC pulse that is long enough to deliver maximum current (~18A) for maximum force, but short enough to protect against overheating. This would explain the very loud and distinct "click" that is heard upon engagement.

If it did that, then doing it a second time would have the same result. It doesn't. There won't even be a subtle click the second time. And it's more than a click. It sounds like something engaging. If your theory were true, then it would make no sense for Tesla to document the function and tell you that you can do it when the vehicle is in park.

 

[evermore]

Happy to do so if you can point me to a sensor that can measure applied pressure of the brake pads and costs under $50.
I'd expect finding a load cell that is just the right thickness would be hard and they are really expensive generally.

Would a force sensor be good enough? A401-25 | Digi-Key Electronics

 

[spad]

Happy to do so if you can point me to a sensor that can measure applied pressure of the brake pads and costs under $50.
I'd expect finding a load cell that is just the right thickness would be hard and they are really expensive generally.

They are cheap

e.g.

https://www.aliexpress.com/i/32960696900.html

not sure if 150kg is enough though

 

[gearchruncher]

Yeah, nowhere close. Useful brake pad forces are in the 2500-5000 kg range.

@evermore has one that goes to a reasonable level, I'm looking into if you can just shove one between the pads and the disc and if that would work or not.

 

[spad]

Yeah, nowhere close. Useful brake pad forces are in the 2500-5000 kg range.

@evermore has one that goes to a reasonable level, I'm looking into if you can just shove one between the pads and the disc and if that would work or not.

That high? The traditional wire based handbrake isn’t that hard to pull <10kg I guess.
I know there is leverage in the system but don’t think ratio is so high.

 

[gearchruncher]

I know there is leverage in the system but don’t think ratio is so high.

You move the handbrake lever a foot or 250cm.
The pads move a 1/32nd of an inch or a mm.

Plus, the car weighs 4000lbs and can lock up the brakes with a 12" diameter rotor and brake pads that have a mu of well under 0.5. This is simple math.

Seriously, a quick search on the internet will tell you I'm right.