How to avoid a runaway condition

[Dr. Smoke]

Presumably the OP fears an electronic failure. Older DC-powered electric cars could fail with a short in the MOSFET(s) and hit full throttle. Maybe the OP is aware of this. This is why if you were smart, you added a kill switch to your conversion. (Lots of ppl didn't, and never had a failure)

But the Tesla is a 3-phase AC induction motor. If the IGBTs in any phase fail the computer will notice and shut everything down. Anyway, a failure on any phase of AC would violate the finely coordinated waveforms necessary for efficient drive, which means the motor would slow or stop rather than run fast. It would want to overheat but would get shut off. The challenge wasn't to make the car go, it was to make the car not stop, with all the coordination that's necessary...

What if the control circuitry fails? You think they haven't thought of this? The accelerator pedal actually feeds two signals direct to the rear inverter. These are compared and had better match or a serious fault is thrown. And the pedal signals had better be proportional to the current going to the motors or a serious fault is thrown. Everything is intensely monitored by the computers, including things you can't imagine.

Rather than use a vacuum brake booster, newer Teslas (post-Sep 2014) use an electromechanical brake system, and the feeling under your foot comes from the resistance of a spring and an electric motor -- it's like a steering rack on its side. The Porsche 918 is the only other production car using this system, and the arrangement gives Tesla great flexibility with the automatic brakes in autopilot mode. No electricity? It defaults to unassisted hydraulic of course. You think they didn't worry about that? Hydraulic failure? Same or more likelihood for any other brand of car.

 

[Saghost]

Presumably the OP fears an electronic failure. Older DC-powered electric cars could fail with a short in the MOSFET(s) and hit full throttle. Maybe the OP is aware of this. This is why if you were smart, you added a kill switch to your conversion. (Lots of ppl didn't, and never had a failure)

But the Tesla is a 3-phase AC induction motor. If the IGBTs in any phase fail the computer will notice and shut everything down. Anyway, a failure on any phase of AC would violate the finely coordinated waveforms necessary for efficient drive, which means the motor would slow or stop rather than run fast. It would want to overheat but would get shut off. The challenge wasn't to make the car go, it was to make the car not stop, with all the coordination that's necessary...

What if the control circuitry fails? You think they haven't thought of this? The accelerator pedal actually feeds two signals direct to the rear inverter. These are compared and had better match or a serious fault is thrown. And the pedal signals had better be proportional to the current going to the motors or a serious fault is thrown. Everything is intensely monitored by the computers, including things you can't imagine.

Rather than use a vacuum brake booster, newer Teslas (post-Sep 2014) use an electromechanical brake system, and the feeling under your foot comes from the resistance of a spring and an electric motor -- it's like a steering rack on its side. The Porsche 918 is the only other production car using this system, and the arrangement gives Tesla great flexibility with the automatic brakes in autopilot mode. No electricity? It defaults to unassisted hydraulic of course. You think they didn't worry about that? Hydraulic failure? Same or more likelihood for any other brand of car.

I was with you until you said that what you're feeling in the brake pedal is a spring and electric motor.

As I understand it, on modern iBooster cars the pedal is connected directly to the master cylinder piston just like an unboosted brake system - and an electric motor is also geared to the piston, providing the assistance and hold and autonomous braking functions.

One of the interesting unique outcomes of this is you can feel the car pull the brake pedal down under TACC/AP or right at the end of a stop sometimes as vehicle hold engages.

 

[jaguar36]

Sorry if my post gave the impression everything occuring at the same time, -just the opposite was meant: either/or aspects. To put it as simple as possible:
What do you do when driving away with a cold battery downhill (which is normal and frequent) and as a single fault hydraulics fail?

Exactly the same thing you would do in any other ICE.

I would try press the parking brake stick continuously and only hope it is a very slight downhill as these brakes were not built to slow the car as means of deceleration.

Pretty sure that's exactly what they are designed for.

Seriously, is this just the op on a new user name? Most ridiculous thread I've seen so far on TMC.

 

[Doug_G]

The OP is on temporary probation for trolling. He can post but not without moderator approval.

This whole thread is about grasping at straws. As has been pointed out repeatedly: the car has multiple fail-safes to prevent these problems. Even if somehow everything failed simultaneously in such a way to cause a runaway (or rolling down a steep hill with no regen), the brakes will still stop the car. The brakes will work without any electrical power, and have the standard dual redundant hydraulic system that every other production car uses. Even if you lose a brake line you still have fully functional brakes on two wheels.

This isn't rocket science; this problem was solved decades ago and just because the car has a different power plant doesn't mean the rest of it is built any differently.

 

[radio]

Exactly the same thing you would do in any other ICE.

You would use motor braking and/or turn off the engine on an ICE. Try this with a cold battery as stated and hence no regen available. Not a good idea to rely on.

Pretty sure that's exactly what they are designed for.

Not exactly. Pretty sure Parking brakes were designed for parking. From Wikipedia: Although sometimes known as an emergency brake, using it in any emergency where the footbrake is still operational is likely to badly upset the brake balance of the car and vastly increase the likelihood of loss of control of the vehicle, for example by initiating a rear-wheel skid. Additionally, the stopping force provided by using the handbrake is small and would not significantly aid in stopping the vehicle.

 

[radio]

Presumably the OP fears an electronic failure. Older DC-powered electric cars could fail with a short in the MOSFET(s) and hit full throttle. Maybe the OP is aware of this. This is why if you were smart, you added a kill switch to your conversion. (Lots of ppl didn't, and never had a failure)

But the Tesla is a 3-phase AC induction motor. If the IGBTs in any phase fail the computer will notice and shut everything down. Anyway, a failure on any phase of AC would violate the finely coordinated waveforms necessary for efficient drive, which means the motor would slow or stop rather than run fast. It would want to overheat but would get shut off. The challenge wasn't to make the car go, it was to make the car not stop, with all the coordination that's necessary...

What if the control circuitry fails? You think they haven't thought of this? The accelerator pedal actually feeds two signals direct to the rear inverter. These are compared and had better match or a serious fault is thrown. And the pedal signals had better be proportional to the current going to the motors or a serious fault is thrown. Everything is intensely monitored by the computers, including things you can't imagine.

Rather than use a vacuum brake booster, newer Teslas (post-Sep 2014) use an electromechanical brake system, and the feeling under your foot comes from the resistance of a spring and an electric motor -- it's like a steering rack on its side. The Porsche 918 is the only other production car using this system, and the arrangement gives Tesla great flexibility with the automatic brakes in autopilot mode. No electricity? It defaults to unassisted hydraulic of course. You think they didn't worry about that? Hydraulic failure? Same or more likelihood for any other brand of car.

Thanks a lot Dr. Smoke, as this was exactly the helpful information and reassurance I was looking for in this thread. To sum up what I learned:
- an AC motor in a runaway mode is quite near to impossible due to a very demanding electric management and double signal feed from the pedal. Hence my comment to the OP wether or not brakes would be able to stop the car has little or no relevance.
- driving without regen should be avoided as it temporarily removes one of the few effective safety features when it comes to braking safely.
- next time I'll better check wether I forgot to restrict the recharge limit to 90% or if it was a frosty night before I leave my steep driveway onto the busy crossroad...

I find this to be a rather tiny and acceptable downside compared to the very long list of advantages of owning the best EV in the world.

 

[Dr. Smoke]

I was with you until you said that what you're feeling in the brake pedal is a spring and electric motor.

Fascinating.

As I understand it, on modern iBooster cars the pedal is connected directly to the master cylinder piston just like an unboosted brake system - and an electric motor is also geared to the piston, providing the assistance and hold and autonomous braking functions.

What did I say? What were the words what come out my mouf? It's electric motor assist, a very similar mechanism to the electric rack and pinion steering. Buy one and take it apart.


And Mods, I've been banned several times. Free them all.

 

[Saghost]

Fascinating.


What did I say? What were the words what come out my mouf? It's electric motor assist, a very similar mechanism to the electric rack and pinion steering. Buy one and take it apart.


And Mods, I've been banned several times. Free them all.

An electric assist, yes. But a direct mechanical connection to the master cylinder piston, not a force simulator as you seemed to be saying.

 

[Doug_G]

- driving without regen should be avoided as it temporarily removes one of the few effective safety features when it comes to braking safely.

That is a gross exaggeration of the situation. Brakes are the primary mechanism for stopping cars. Tesla's braking systems have the exact same safety features as every other car on the road, i.e. purely mechanical dual-redundant hydraulic circuits.

 

[stopcrazypp]

That is a gross exaggeration of the situation. Brakes are the primary mechanism for stopping cars. Tesla's braking systems have the exact same safety features as every other car on the road, i.e. purely mechanical dual-redundant hydraulic circuits.

Yeah, I don't get his point about regen either. ICE cars don't have regen either and can brake perfectly fine. And the amount of engine braking available from ICE cars is insignificant compared to regen.

 

[oktane]

My sons has a Tesla S and I ask him if something happened that the car took a mind of its own went to accelerating and wouldn't stop, how do you shut it off to avoid a collision? He wasn't sure. We worked together and found one thing he could do is put it in neutral. There is no way to our knowledge to shut it off while driving. If there is other alternatives please reply. Every owner should be prepared for an emergency such as this. Randy

Do not let the car slow to below 55 MPH...

 

[radio]

Yeah, I don't get his point about regen either. ICE cars don't have regen either and can brake perfectly fine. And the amount of engine braking available from ICE cars is insignificant compared to regen.

Simplified answer:
If hydraulic brakes fail in an ICE you have a single effective backup (=engine braking) -ALWAYS.
In an EV there is a comparable single effective backup (=regen braking) -MAYBE

In detail:
Engine braking (ICE) or regen braking (EV) is the first means of deceleration, by lifting the gas pedal compared to applying the hydraulic brake (guessing: 70% vs 30% of the time) and both (engine and regen braking) are able to stop a car by itself alone (regen earlier than engine b.). More importantly engine braking is -always- available no matter what, while regen braking in EVs unfortunately is not and depends on circumstances (full/hot/cold battery).
Losing regen braking, as it sometimes happens (and is no failure) forces you to fall back to the only remaining effective braking method beeing electro-hydraulic braking. Thankfully, as "Doug_G" noted, this last resort -hydraulic braking- in a Tesla is quite fail safe as it is a "dual-redundant" system and hence reassuring for me.

Not everything we know from ICE cars can be translated 1:1 to EV's and Tesla is no exception here and hence it is worth it to learn more detail on probably the most crucial systems of this car.
Just today in the german Tesla-Forum an owner experienced a loss of servo boost (= partial hydraulic failure) and described it as a scary moment especially if he drove 200 km/h (GoogleTranslate):
Google Übersetzer

The misunderstanding is maybe due to the fact that in Europe we use engine braking by adjusting the -usually non-automatic- gearbox to effectively slow the car, especially on longer downhill slopes (for trucks even required by law). This may be different in the US using automatic gearboxes.

 

[Ulmo]

Simplified answer:
If hydraulic brakes fail in an ICE you have a single effective backup (=engine braking) -ALWAYS.
In an EV there is a comparable single effective backup (=regen braking) -MAYBE

I've experienced transmission failure in 4 of my ICE cars that caused engine to not slow down the car. In every case, the transmission became disconnected from the engine, due to breakage, with no possibility of repair while driving; this happened on two stick shifts and two automatic transmissions; they were made in Sweden, Japan, Mexico, and Germany; in every car except perhaps the Japanese one, at least one example was on the freeway for each car; the only common denominator was that they were all old cars and all happened in San Jose (perhaps exacerbated by Road Diet).

In trucks (40 ton, fully loaded), if you are going down a hill (such as grapevine or any mountain or big hill) and attempt to downshift, the truck will accelerate faster than the gear you are trying to shift into can be shifted into, and the brakes will heat up beyond their ability to stop the truck, and the truck will become a runaway truck and you will have to use a runaway truck ramp to stop your truck from killing people, and you will likely not survive. That's why trucks have to get into low gear at slow speed before falling down a hill on wheels, and must stay in that gear the whole way down, and explains why many trucks go different speeds downhill, since every driver picks a different safe speed (and thus gear (that they must stay in all the way down the hill (left foot solidly off clutch, firmly planted in footwell on floor so you don't touch clutch))) for their truck's load angle, weight, and truck ability to handle all of that, as well as driver guesses and mistakes.

So, no, ICE cars (and trucks) do not have the engine available as a brake every time. FAR FROM IT.

Press and hold the Park button on the "gear" stalk. It will take the car out of drive and also apply the emergency break.

Every time I tested it, I chickened out because it took too long. Doesn't it take a second or two?

Okay, I see now you're not serious. Sorry I took the time to respond.

Indeed. Seeing that post explained to me why TMC'ers sometimes jump down my throat for information I collected about Tesla that isn't always 100% awesomely great.

 

[Saghost]

Simplified answer:
If hydraulic brakes fail in an ICE you have a single effective backup (=engine braking) -ALWAYS.
In an EV there is a comparable single effective backup (=regen braking) -MAYBE

In detail:
Engine braking (ICE) or regen braking (EV) is the first means of deceleration, by lifting the gas pedal compared to applying the hydraulic brake (guessing: 70% vs 30% of the time) and both (engine and regen braking) are able to stop a car by itself alone (regen earlier than engine b.). More importantly engine braking is -always- available no matter what, while regen braking in EVs unfortunately is not and depends on circumstances (full/hot/cold battery).
Losing regen braking, as it sometimes happens (and is no failure) forces you to fall back to the only remaining effective braking method beeing electro-hydraulic braking. Thankfully, as "Doug_G" noted, this last resort -hydraulic braking- in a Tesla is quite fail safe as it is a "dual-redundant" system and hence reassuring for me.

Not everything we know from ICE cars can be translated 1:1 to EV's and Tesla is no exception here and hence it is worth it to learn more detail on probably the most crucial systems of this car.
Just today in the german Tesla-Forum an owner experienced a loss of servo boost (= partial hydraulic failure) and described it as a scary moment especially if he drove 200 km/h (GoogleTranslate):
Google Übersetzer

The misunderstanding is maybe due to the fact that in Europe we use engine braking by adjusting the -usually non-automatic- gearbox to effectively slow the car, especially on longer downhill slopes (for trucks even required by law). This may be different in the US using automatic gearboxes.

In addition to @Ulmo 's disturbing experiences, there are a lot of engine failure modes that make engine braking impossible - thrown rods, seized engines, even a runaway or stuck throttle.

All the conditions that cause no regen happen at the start of a trip; within five or ten miles you should have significant regen no matter what conditions are affecting you - more than engine braking anyway (though full regen can take quite a bit longer.)

 

[Ulmo]

In addition to @Ulmo 's disturbing experiences, there are a lot of engine failure modes that make engine braking impossible - thrown rods, seized engines, even a runaway or stuck throttle.

Thanks, but luckily they weren't that disturbing. I had plenty of clearance and reaction time to take other countermeasures, and in all but one case, the brakes were working fine. (In the brakes not working fine situation, I was on US-101 North during rush hour about to go under the train tracks just north of SJC, but I came to a safe stop; I believe the emergency brakes worked adequately that time, and all other conditions were decent: dry roads, mostly benign commuters*. It did put enough fear in me that I junked the car after that, though, despite it probably being unwise not to repair it (financially).)

* In my experience, commuters are better drivers (during commute), compared to their non-commute counterparts, since they have a lot more practice and are intentionally more prepared for the task at hand, across all social, economic, cultural, language, intelligence, genetic, gender, and other diversities. Practice really does help. I am much more comfortable being around my fellow commuters than other drivers, in general. Today I was absolutely amazed (for about the 2,415th time) how well all the other drivers were driving on my regular commute home. No way could that commute have gone that well if even 0.03% of the drivers had mucked up more than they did. (And, this helps explain why I think Elon's tunnel sled idea will work really really well in tunnel (less relative speed variances as well as opportunities to coordinate speeds), but may get people out of practice once they get back to surface, but if that's all they need to know how to do, they might do fine.)

 

[radio]

I've experienced transmission failure in 4 of my ICE cars that caused engine to not slow down the car. In every case, the transmission became disconnected from the engine, due to breakage, with no possibility of repair while driving; this happened on two stick shifts and two automatic transmissions; they were made in Sweden, Japan, Mexico, and Germany; in every car except perhaps the Japanese one, at least one example was on the freeway for each car; the only common denominator was that they were all old cars and all happened in San Jose (perhaps exacerbated by Road Diet).

In trucks (40 ton, fully loaded), if you are going down a hill (such as grapevine or any mountain or big hill) and attempt to downshift, the truck will accelerate faster than the gear you are trying to shift into can be shifted into, and the brakes will heat up beyond their ability to stop the truck, and the truck will become a runaway truck and you will have to use a runaway truck ramp to stop your truck from killing people, and you will likely not survive. That's why trucks have to get into low gear at slow speed before falling down a hill on wheels, and must stay in that gear the whole way down, and explains why many trucks go different speeds downhill, since every driver picks a different safe speed (and thus gear (that they must stay in all the way down the hill (left foot solidly off clutch, firmly planted in footwell on floor so you don't touch clutch))) for their truck's load angle, weight, and truck ability to handle all of that, as well as driver guesses and mistakes.

So, no, ICE cars (and trucks) do not have the engine available as a brake every time. FAR FROM IT.

Sorry, I probably simplified too much and should have written "-normally ALWAYS" and "-normally MAYBE" to clarify my assumption was not about a double failure. Of course any failure of the transmission system can occur anytime and you having 4 of them is something really special. But if I understand your other following post you also had brake failure at the same time on US-101 which makes it an extremely unfortunate event/experience (transmission+brakes). Did the car have a double hydraulic brake like in a Tesla or was it only a simultaneous failure of the brake booster?

I agree very much with your good description of the truck going downhill less the point "must stay in that gear the whole way down". There is no apparent reason why a truck wouldn't be able to stop to select a better gear if slopes change, curves get tighter or obstacles occur.

In addition to @Ulmo 's disturbing experiences, there are a lot of engine failure modes that make engine braking impossible - thrown rods, seized engines, even a runaway or stuck throttle.

All the conditions that cause no regen happen at the start of a trip; within five or ten miles you should have significant regen no matter what conditions are affecting you - more than engine braking anyway (though full regen can take quite a bit longer.)

Right, very good point and helpful. That was also the time I was mainly worrying about -to start from my steep driveway and hit somebody on the sidewalk or being hit when freewheeling into the busy crossstreet (before I learned in this thread about details and specifics of the brakes in my Tesla). I must note that I maxed out the speed of my car once (reaching 253 km/h) and noted a reduced regen availability for a short period afterwards (not completely) and related that to the hot battery. I was't able to test that any further, though.

 

[kort677]

 

[DarkMatter]

Sorry, I probably simplified too much and should have written "-normally ALWAYS" and "-normally MAYBE" to clarify my assumption was not about a double failure. Of course any failure of the transmission system can occur anytime and you having 4 of them is something really special. But if I understand your other following post you also had brake failure at the same time on US-101 which makes it an extremely unfortunate event/experience (transmission+brakes). Did the car have a double hydraulic brake like in a Tesla or was it only a simultaneous failure of the brake booster?

Some newer automatic transmissions are designed to freewheel, as it's a slight efficiency benefit. No engine braking in that scenario.

And yes, all cars made since 1976 must have a split circuit braking system if they are to be sold in the US. This does not prevent failures from happening. The typical failure is not noticing fluid loss, as when enough fluid is lost both circuits will fail. This is why there is a low fluid warning light in most any car.

 

[Yinn]

And yes, all cars made since 1976 must have a split circuit braking system if they are to be sold in the US. This does not prevent failures from happening. The typical failure is not noticing fluid loss, as when enough fluid is lost both circuits will fail. This is why there is a low fluid warning light in most any car.

The only time I've lost brakes was in the late 90's, my foot sunk right to the floor when I hit the pedal. Looked down and my indicated speed was roughly 130mph (was on a track) and I needed to get down to about 70mph for a walled turn. I'm pretty sure I lost some fluids right there and then. Turns out the car did too...

 

[JHuberman]

I had my accelerator stick when I took my foot off the accelerator at the top of a hill. My reaction (at 17 years old) was to put the car in neutral (which blew the engine) . Then thought to turn off the key. I was leaving for college in a few weeks so we gave the car away.