0
\$\begingroup\$

In this video, at 16:23, the presenter says that when we press the accelerator pedal in an electric car, we are "commanding torque and not speed". I'm confused here, aren't we making the car go faster (more speed) when we press the pedal? So aren't we in fact commanding speed?

Moreover, at 4:27 the animation is depicted in a way showing that higher torque means faster rotation (thus more speed). But, doesn't high torque not always equal high speed for a motor? (Because there exists high torque low speed motors, etc.)

Also, in situations where high torque does equal high speeds, then is there no need for speed control at all? Can we go faster or slower by just controlling the torque alone?

P.S. I apologize for asking 3 individual questions in a single post. They are all connected and hence I asked them together.

\$\endgroup\$
3
  • \$\begingroup\$ As many answers below point out, the pedal predominantly sets torque, but if you consider the user as part of the control loop, then "we" set the speed when kids are in the car, or the adrenaline when they are not. \$\endgroup\$
    – P2000
    Feb 2, 2021 at 17:09
  • \$\begingroup\$ Suppose the car is parked with it's nose right up against a brick wall. So it can't move no matter what you do. Then put the car in drive and slam the accelerator. Your speed will be zero, right? You are missing the concept of "load". If I floored the accelerator on your car on the freeway I would get to X MPH. If I remove the engine, and put it in a Semi truck, and again floor the accelerator, would you expect the same speed X??? How about if I strap it to a motorcycle??? \$\endgroup\$
    – Kyle B
    Feb 2, 2021 at 21:32
  • \$\begingroup\$ Driving on ice is another demonstration of how speed is only a predictable side effect of torque. \$\endgroup\$
    – K H
    Feb 3, 2021 at 1:06

6 Answers 6

4
\$\begingroup\$

I'm confused here, aren't we making the car go faster (more speed) when we press the pedal?

You press the pedal to go faster - that is your aim and your foot and brain are in control of what you want to happen i.e. go faster. But, you might also press the pedal a bit harder to maintain the same speed when going up a hill. So here, you are definitely not pressing the pedal to go faster. Do you see what I mean?

Either way, the real effect is to put more power into the motor to fight against the force that might be stopping you accelerating or, the force that is causing you to slow down on a hill. More pedal action means more power sent to the motor and, if that results in higher speed then, to reach that higher speed, more torque will be produced to fight against the force that was keeping you at a constant velocity.

But, doesn't high torque not always equal high speed for a motor?

The thing about motors is this: -

$$\text{Shaft power transmitted = } 2\cdot\pi\cdot n\cdot T$$

Where n = revolutions per second and T = torque.

So, some motors have high speed and low torque and, for the same output power, different motors might be the other way around. Feed a motor through a speed reduction gearbox and you'll have a high torque low speed drive output.

Can we go faster or slower by just controlling the torque alone?

Yes we can because the torque is countering the force that holds the motor at constant speed. But, more realistically you are asking the motor to deliver more power.

A final note - about cruise control - it is a speed control but uses the same mechanism (as the pedal) but within a control loop that seeks to maintain speed. Your brain and foot can do the same of course. You can also use an "overall control system" to make a real system appear to be something that it naturally isn't. For instance, a flyback converter power supply is inherently a power regulator in one mode but, with an outer supervisory control loop that makes it look like a voltage regulator. In other words, it has cruise control.

\$\endgroup\$
9
  • \$\begingroup\$ Yup got it, thanks for the explanation. Although, I have seen some videos where its said that a VFD drive present in between the battery and the motor is used to control the speed by altering frequency of current, and torque by altering the amplitude of current. But in the video that I linked, the motor torque is controlled simply by varying the current magnitude (given that the rotor and stator mmf are at 90 degrees - which is an FOC control). So does the alteration of frequency by a VFD to increase speed not come into play here? \$\endgroup\$
    – penguin99
    Feb 2, 2021 at 17:54
  • \$\begingroup\$ It depends entirely on the scenario you are trying to control @noorav. \$\endgroup\$
    – Andy aka
    Feb 2, 2021 at 18:31
  • \$\begingroup\$ Oh okay. But is there any reason as to why torque control is preferred over speed control? The "scenario" here being we choose to control torque rather than speed \$\endgroup\$
    – penguin99
    Feb 2, 2021 at 20:55
  • \$\begingroup\$ The simplest answer I can give is that it is smoother. But, it really does depend on the specific situation. Voltage control on a flyback controller is like speed control but, we are careful that when we make a control loop to regulate voltage we have, in effect, an inner loop that stops current at a limit. That’s equivalent to switching my cruise control on but not rapidly jerking to the set speed but doing so with a gradual torque (aka current limit for a flyback converter). \$\endgroup\$
    – Andy aka
    Feb 2, 2021 at 21:05
  • \$\begingroup\$ So to condense everything that's been said - We are commanding power to the wheels when we press the pedal, and since power is torque (going by answers and replies below), we are essentially controlling the torque of the wheels. This torque translates to higher or lower speeds based on the resistance faced. And the way the torque is controlled is by using FOC where we keep a 90 degree angle between stator and rotor MMF and varying the magnitude of current to increase or decrease torque. And current magnitude is varied by varying voltage supplied, which is essentially what the controller does. \$\endgroup\$
    – penguin99
    Feb 3, 2021 at 5:58
3
\$\begingroup\$

You are commanding more power to the wheels (more torque.)

On level ground, more torque translates to higher speed.

If you transition from level ground to a hill, you press on the accelerator pedal to put more power to the wheels to maintain your speed. The motor produces more power, but the speed doesn't go up.

It doesn't set the speed directly. It only controls the power produced.

Regulating the speed by the pedal directly would require a feedback loop so that a particular position of the accelerator always corresponds to a particular speed.

The accelerator could be made that way, but I doubt that anyone does it. More likely that they modeled the accelerator action of electric cars on the operation of the accelerator in cars with combustion motors. The accelerator in a car with a combustion motor controls how much fuel the engine gets, and thus how much power it produces. You've got millions (or billions) of users trained in that type of control, and no good need to change it.


Controlling the torque will always control the speed. There is always friction, rolling resistance, hills, and wind resistance to fight against.

With low torque, your vehicle can only overcome a certain amount of that resistance so your speed is low.

With high torque, your car can overcome more of the resistance and therefore moves faster.

Even on level ground, it takes more and more torque to go faster and faster because if nothing else the wind resistance goes up as you go faster.

\$\endgroup\$
5
  • \$\begingroup\$ So to clarify, I must view torque and speed together as one entity - power (rather than confusing myself by splitting it up). And this power can either translate to high speed or low speed based on the type of resistance the tires face. Am I right? \$\endgroup\$
    – penguin99
    Feb 2, 2021 at 17:52
  • 1
    \$\begingroup\$ Power is not speed. Power is what gives you torque. From torque and all the various resistances you get speed. \$\endgroup\$
    – JRE
    Feb 2, 2021 at 18:24
  • 1
    \$\begingroup\$ @noorav Power can be translated into higher speed with lower torque, or lower speed with higher torque, if you want to think about it that way. \$\endgroup\$
    – DKNguyen
    Feb 2, 2021 at 20:37
  • \$\begingroup\$ @JRE, DKNgyuen yup I get what you're trying to say \$\endgroup\$
    – penguin99
    Feb 2, 2021 at 20:56
  • \$\begingroup\$ @JRE, So to condense everything that's been said - We are commanding power to the wheels when we press the pedal, and since power gives us torque we are essentially controlling the torque of the wheels. This torque translates to higher or lower speeds based on the resistance faced. And the way the torque is controlled is by using FOC where we keep a 90 degree angle between stator and rotor MMF and varying the magnitude of current to increase or decrease torque. And current magnitude is varied by varying voltage supplied, which is essentially what the motor controller does. Am I right? \$\endgroup\$
    – penguin99
    Feb 3, 2021 at 6:02
2
\$\begingroup\$

Depends on how the control system is built. The control system can be made to correlate pedal position with torque (which is fairly synonmous with current for motors), speed, or voltage.

Voltage is simplest and corresponds to neither speed nor torque independently. Similar to a throttle on a gas engine. Does it control speed or torque? Neither. It controls how much gas is fed to the engine which controls power. That power gets allocated on its own to torque and speeds based on engine/motor/load characteristics.

Voltage is simplest because it is the most direct since batteries are considered voltage sources. Whether speed, torque/current, or voltage control, the control system in all cases will directly manipulate the motor voltage, but can be designed to to shield you from this so the pedal corresponds to speed or torque instead of voltage. Then the control system fiddles with motor voltage while monitoring speed or torque to achieve the torque or speed that the pedal is set to.

But increasing torque does increase speed because excess torque causes acceleration. The reason torque control isn't considered identical to speed control is that torque control results in different speeds at different loads. You can't set torque to be constant and expect a constant speed under varying load. Similarly, you cannot set speed to a constant level and expect torque to remain constant under varying load.

\$\endgroup\$
2
  • \$\begingroup\$ I'd guess that battery condition and current draw are conditions applied as well, and restrict the available torque/current, similar to combustion engines in "sport" vs "economy" mode. \$\endgroup\$
    – P2000
    Feb 2, 2021 at 17:05
  • \$\begingroup\$ @P2000 Yes. I would think in better systems a current limiting mechanism is in place. Whereas the constant torque/current I talk about in my post is a current regulation mechanism. \$\endgroup\$
    – DKNguyen
    Feb 2, 2021 at 17:07
1
\$\begingroup\$

They can make it whatever they want with digital control algorithms. They would likely choose to provide a similar UI to a gas pedal connected to an IC engine and traditional drive train.

Some full electric cars offer an optional braking mode that is quite different from the traditional.

I'm amused by the plug-in hybrid car I occasionally drive that mimics torque converter slip even when entirely in electric mode (you take your foot off the brake and it moves ahead slowly).

\$\endgroup\$
4
  • 1
    \$\begingroup\$ Like my old digital piano that had simulated pedaling noise. Like c'mon. \$\endgroup\$
    – DKNguyen
    Feb 2, 2021 at 15:39
  • \$\begingroup\$ Indeed, a similar user experience to gas/throttle control, even if they have access to all sensors and could control for a set acceleration, speed or torque. \$\endgroup\$
    – P2000
    Feb 2, 2021 at 17:03
  • \$\begingroup\$ @DKNguyen It would be nice to have one with a Glen Gould mode that would insert (human-sounding) humming. \$\endgroup\$ Feb 2, 2021 at 17:05
  • \$\begingroup\$ Haha, Rachmaninov did the same... \$\endgroup\$
    – P2000
    Feb 2, 2021 at 17:11
1
\$\begingroup\$

Short answer: when you press the pedal, you are commanding power, which is the product of torque and RPM (speed.)

From a cruise speed on level ground, pressing the pedal down commands an increase in power input to the motor. The motor, in turn, will speed up until the vehicle reaches a new equilibrium between friction + losses (drag) and commanded power level (thrust).

The torque will not be constant, but will suddenly increase initially as the vehicle accelerates, then taper off until reaching the new, higher level at the equilibrium point as the motor speed catches up.

Say you're cruising along at 50MPH, and using 20HP to do that. You command an increase to 30HP. The vehicle accelerates hard at first (high torque), then less and less until you reach the speed where rolling resistance is 30HP (say, 65MPH) at which point no further acceleration happens. With that RPM being reached, power demand is once more balanced with power command. That is, thrust = drag.

Now, that rate of acceleration? That's determined by both the increase in resistance, and by vehicle's mass.

"Simplify, then add lightness." - Colin Chapman, founder of Lotus

\$\endgroup\$
0
\$\begingroup\$

Usually we think of speed as a result of the torque applied and the load/friction effected. This is the common scenario for VFD controllers (V/f controllers)

I think you are asking not just whether speed OR torque are controlled, but also whether they can both be controlled simultaneously and independently.

Speed and torque can be independently set and controlled by using an Open-Loop Vector contoller (OLV). This involves keeping the magnetization current Id and the torque current Iq at 90deg.

OLV can provide higher torque at lower (starting) speeds, and provide 4-quadrant control (i.e. incl. breaking).

The PDF below by Yaskawa provides a nice overview, but there is tons of literature on this control technique (once you know what to look for...)

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.