From what I can see, commercial high speed rail typically uses a limit of 1 m/s^2 acceleration and braking.

Important to passenger comfort is the third derivative of distance with respect to time, the so called “jerk”. From what I can see, the TGV and Japanese high speed rail keep this under 1.5 m/s^3. Put simply, a sharp increase in acceleration feels bad. 1 m/s^2 is a mild acceleration but if you go from 0 m/s^2 to 1 m/s^2 in a hundredth of a second, it will cause a queasy feeling.

Is there any sense in which the fourth derivative matters? ie if you go from 0 to 1.5 m/s^3 jerk in a hundredth of a second, will it feel any worse than if you do it over a second?

i dont know but i like the question.

Thanks, I’m glad someone around here appreciates a good question.

dv said:

Thanks, I’m glad someone around here appreciates a good question.

I think dv is just trying to prove something.

PermeateFree said:

dv said:

Thanks, I’m glad someone around here appreciates a good question.

I think dv is just trying to prove something.

fuck off dickwad…

The fourth derivative of displacement is called “snap” or “jounce” (the rate of change of jerk with respect to time) and as you suggest, it’s often used as a design criteria on transportation systems

Jounce eh. Cheers, d-s.

Dropbear said:

PermeateFree said:

dv said:

Thanks, I’m glad someone around here appreciates a good question.

I think dv is just trying to prove something.

fuck off dickwad…

Now that you have moved to Brisbane, how is your new found koala habitat fairing?

dv said:

Jounce eh. Cheers, d-s.

or snap…

interestingly enough, the 5th and 6th derivatives of displacement are ‘crackle’ and ‘pop’

totally cereal

I don’t know the answer to your question DV, but as an aside:

I’ve been reading a lot lately about the progress on the Oculus Rift head mounted “virtual reality” display which should be available to purchase soonish. One of the engineering issues they are struggling with is what they are calling “simulator sickness”. It’s basically motion sickness caused by the discrepancy between personal movement as seen in virtual reality and the fact that your physical body is not moving in the same way. The Oculus best practices guide suggests that software developers implement a system whereby if the person using the headset inputs a command to walk forward that the acceleration should be instant (ie zero jerk) as this provides a less nauseating experience for the user.

This is completely contrary to the information about trains in the OP.

I wonder if anyone has any idea why this might be.

Well that’s interesting. On a train, you could be looking out the window (in which case your visuals will match your relative motion) or looking at stuff on the train (in which case what you see will not give you any key to your motion).

I’ve felt disorientation when playing FPS games were you peer over a cliff to a valley below.

In a practical sense if the person feels alarmed / sick from being jolted around – its probably no good for the body, the body’s feed back instinctively understands what it likes and what it doesn’t .

For example some people won’t ride in cars with a hard suspension , it makes them feel ill.

Jolts could cause you injury , ever watch clarkson taken away by an ambulance because of a car’s jolts of acceleration ?

If a jolt is already making you feel ill it’s a matter of how much torture you are willing to submit to

You’d be better off looking at roller coaster rides , they seem to have a handle on the tolerances of the human body

esselte said:

I don’t know the answer to your question DV, but as an aside:

I’ve been reading a lot lately about the progress on the Oculus Rift head mounted “virtual reality” display which should be available to purchase soonish. One of the engineering issues they are struggling with is what they are calling “simulator sickness”. It’s basically motion sickness caused by the discrepancy between personal movement as seen in virtual reality and the fact that your physical body is not moving in the same way. The Oculus best practices guide suggests that software developers implement a system whereby if the person using the headset inputs a command to walk forward that the acceleration should be instant (ie zero jerk) as this provides a less nauseating experience for the user.

This is completely contrary to the information about trains in the OP.

I wonder if anyone has any idea why this might be.

Surely instant acceleration would be infinite jerk, rather than zero.

This is of course impossible in any real system.

The Rev Dodgson said:

esselte said:

I don’t know the answer to your question DV, but as an aside:

I’ve been reading a lot lately about the progress on the Oculus Rift head mounted “virtual reality” display which should be available to purchase soonish. One of the engineering issues they are struggling with is what they are calling “simulator sickness”. It’s basically motion sickness caused by the discrepancy between personal movement as seen in virtual reality and the fact that your physical body is not moving in the same way. The Oculus best practices guide suggests that software developers implement a system whereby if the person using the headset inputs a command to walk forward that the acceleration should be instant (ie zero jerk) as this provides a less nauseating experience for the user.

This is completely contrary to the information about trains in the OP.

I wonder if anyone has any idea why this might be.

Surely instant acceleration would be infinite jerk, rather than zero.

This is of course impossible in any real system.

I agree

The yanks studied all of this thing during the golden age if space travel

To get man to the moon they had to know and understand what the human body would tolerate , the jerks, rolls, pitches etc

There have been numerous studies on this subject

dv said:

From what I can see, commercial high speed rail typically uses a limit of 1 m/s^2 acceleration and braking.

Important to passenger comfort is the third derivative of distance with respect to time, the so called “jerk”. From what I can see, the TGV and Japanese high speed rail keep this under 1.5 m/s^3. Put simply, a sharp increase in acceleration feels bad. 1 m/s^2 is a mild acceleration but if you go from 0 m/s^2 to 1 m/s^2 in a hundredth of a second, it will cause a queasy feeling.

Is there any sense in which the fourth derivative matters? ie if you go from 0 to 1.5 m/s^3 jerk in a hundredth of a second, will it feel any worse than if you do it over a second?

I don’t think the fourth derivative matters at all. Or perhaps I should vary that statement. If the fourth derivative is kept low then that would allow the third derivative to exceed limits needed when the fourth derivative is not constrained.

The second derivative is a force, so people lean at an angle using static equilibrium to balance that force. The third derivative is what upsets that static equilibrium. People need time to adapt to the failure of static equilibrium, hence the limit on third derivative. So the third derivative corresponds to constant velocity.
The fourth derivative would only have an affect through “reaction time”, that short period of time before a person can react to a change in equilibrium.

I know about this through my Civil Engineering course as it related to road bends. You do not have a sudden change in road curvature from straight to curved. If so then you are asking people to suddenly swing the steering wheel from one position to another in zero time. Constant third derivative corresponds to constant rate of turning the steering wheel, which most people can manage without trouble. Constant fourth derivative corresponds to constant acceleration of steering wheel angle, which isn’t normally necessary – but – if the steering wheel is accelerating and decreasing smoothly then the top speed at which it is allowed to turn (the maximum third derivative in the initial post) can be allowed to increase.

Do a Google on “road transition curve”. Constant second derivative is constant curvature. Constant third derivative (or other gentle variation in second derivative) gives the spiral section of the curve. eg. http://www.mathalino.com/sites/default/files/reviewer-surveying/003-spiral-curve-transition-curve.gif

Admittedly the example I’ve given is for lateral motion (steering), but for longitudinal motion (braking and acceleration) exactly the same rules apply.

I don’t think the fourth derivative matters at all.

—-

That’s my gut feeling too.