Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Instantaneous Velocity of a car after letting off the throttle

  1. Jul 27, 2010 #1
    I have an interesting scenario that a co-worker and I have been arguing over.
    I noticed that when I accelerate at full throttle to 80km/h then quickly take my foot off the throttle and simutaneously push in the clutch (as to remove the accelerated force) the velocity remains increasing to around 87km/h after a couple seconds where it then begins to decelerate.
    Why does this occur??? Basic physics dictates that this should not occur.

    Is there some kind of rotational intertial force of the wheels/ tires/ and drivetrain being greater then the opposing wind resistance forces and the bearing frictional forces???

    Or is it something to do with the rate of acceleration change???

    All jokes aside I do not own a Prius!! :) This is something I've always wondered. How can the instantaneous velociy remain increasing after letting off the throttle???

    Thanks in advance!!!
    Last edited: Jul 27, 2010
  2. jcsd
  3. Jul 27, 2010 #2


    User Avatar
    Homework Helper

    There's no such thing as "rotational inertial force."

    I don't know much about the internal workings of your car, but my guess would be that the connection between the engine and the wheels is not being completely broken, at least not right away.
  4. Jul 27, 2010 #3
    Not to point out the obvious but you were probably going down a slight gradient. If you did this going up hill you'd certainly start decelerating straight away.

    If you let off the throttle it's possible you may get a slight delay in decelerating as the engine runs down (I seriously doubt you'd notice this though as the effect is pretty instant) and the drive train still has inertia, but if you pull in the clutch then all drive is removed and your acceleration becomes 0.
  5. Jul 27, 2010 #4
    It's nothing to do with "my car" all cars do it. Even if you don't press the clutch and allow the compression of the engine to slow the car when the throttle is closed there is a second or two where the speed still increases. I understand how basic physics states that this should not occur, but it does. There is something happening that is beyond my understanding of physics for this to occur. If you don't believe me try it for yourself.

    To better duplicate this scenario a car with a manual transmission should be used so you can press in the clutch quickly and put the car in neutral, but you must be accelerating quickly to make the change more obvious. An automatic transmission will upshift to the next gear and jerk the car forward when you release the throttle so it won't make a valid analogy.
  6. Jul 27, 2010 #5
    This is on a complete level surface. It will even do it on a slight uphill. I'm really looking for someone to shed some light on this issue. Unless you've actually tried the scenario I'm explaining you can argue it on paper all day. And I will agree with you, to my knowlege this should not occur, but it does.
  7. Jul 27, 2010 #6
    I'd also like to point out that speedo's in a car are not precise instruments. Back when I had my mk2 Golf the best it could tell you was the speed within 5 mph up to 60 and then it settled down. I loved that wobbly needle.

    To summarise: your car isn't breaking the laws of physics it's likely inaccurate data gathering and observation that is leading to incorrect conclusions.

    and EDIT:

    Yes I have tried this, and my car along with all normal cars will decelerate when you lift and deceletates quickly when you pull in the clutch. Do it in a lower gear and the effect will be more pronounced, if what you are saying is true you should see the same effect there. I've even done this with an accelerometer (to solve an argument about peak power peak torque acceleration), the instant you lift you decelerate.

    But you dont, if you do this in 1st. Take it to redline and lift you'll get a hell of a jerk. You wont notice the effect in 5th although it is still there. The change is deceleration is less harsh to the point where you don't notice it.
    Last edited: Jul 27, 2010
  8. Jul 27, 2010 #7
    No, the rate in which the speedo moves is very slow. It is accurate enough to see the rate of change that occurs. Sure the speedo could be off by 5mph but the change of speed can still be monitored.
    I even tested to see if the speedo was too slow to react by slamming on the brakes, but the needle went down immediately.
  9. Jul 27, 2010 #8
    An analogue speedo is a device that reads averages, it typically works by having a rotating magnet drag a needle up to read a particular speed and a spring may be used to force the needle down. Consequently the rate at whihc the needle moves up and down can be different. You say you were accelerating flat out in the car, I say the needle had not caught up to read your maximum velocity and although you have disconnected the engine to the gear box you have not discnnected the speedo to the gearbox, therefore the speedo needle will still be trying to catch up to read the vehicles actual velocity after you have disconnected the drive to the vehicle. Try the same experiment on a car with a digital (no magnet or spring involved) speedo and advise the outcome.
  10. Jul 27, 2010 #9
    Burden of proof is on you. You are claiming a scenario which doesn't occur, so post proof.

    I'll have a rummage for the acceleometer spreadsheet and values, it's not 100% relevent but does show that as the input torque is reduced acceleration reduces accordingly.

    So by extransion, if you remove the input torque (by disconnecting drive to the wheels), there is nothing pushing your forward your forward acceleration due to ethe engine is zero. There is no trick to this, where your observation differs from this your observation is wrong i'm afraid.

    As I said before, inertia will keep the wheels spinning and may even cause an acceleration as you may get an extra fully fueled combustion cycle after you close the throttle plate, but this would be virtually unnoticable.
  11. Jul 27, 2010 #10
    That is a very good explanation that could be it, but why would the speedo then go down right away if I press the brakes?? I have an chassis dynamometer at my work that runs a digital speedometer. I'll check it out.
  12. Jul 27, 2010 #11
    I did this quite a while ago so I've got got all the data anymore unfortuately, but here is accelorometer data for MK4 golf Gti 20vT. The car was a little sick at the time and not putting out the torque it should have been so the G figures are a bit low. It was putting out about 140BHP@5500 RPM at the time and should have been putting out closer to 180BHP. (turbo remaps are win!)

    I can't find the dyno plot from around that time either so I can't show that acceleration almost directly follows the torque output line. It was only a cheap crappy acceleromoter to the accuracy isn't fantastic, but was good enough to show the trend the accn directly followed torque.

    Like I said it's not 100% relevent as I didn't record the acceleration on rundown or with the clutch in as it wasn't relevent to why I collected the data. But I can assure you that when the clutch was in there was no acceleration on flat ground.


    MODS I have no idea how to resize the images.

    Attached Files:

    Last edited: Jul 27, 2010
  13. Jul 27, 2010 #12


    User Avatar
    Science Advisor

    Most modern speedometers have a lag to them. I'd blame it on that.

    Anybody has a speed radar to test this out?
  14. Jul 27, 2010 #13
    You could also use one of those GPS based anti speed camera gadgets. Analogue speed indicators also tend to have a linear error and tend to overestiamte speed.

    So @30 they will be fairly close, but at say 140 the indicatied speed is likely to be higher than actual roadspeed.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook