Another treadmill thread: is everybody wrong?

[luigidorf]

Well, this can't be proven or disproven. I should not have introduced it. Frankly, I was hoping you'd simply see the error of your ways.


I simply do not know how you can possibly think this.

Again, if it were true, there would actually be no point in moving your legs in a circular motion, you could - as I said - simply push up and down like a pogo stick.

In fact...

in fact - if, as you claim - the motion of your legs imparts no horizontal force to help you walk/run, then it would not matter which way you rotated your legs. You could rotate your legs backwards (as if running backwards) and it would not slow your forward motion.


Alternately, if you were nudged up to speed so that you were set in motion going backwards at 5mph, you should be able to run forward full-tilt and yet it would not slow your backward motion.

Do you not see the absurdity of your claim?

If you moved your legs in a pogo-stick fashion, and didn't move them back relative to to your body, your foot would be moving relative to the ground as it touched down. Your foot would exert a breaking force on the ground, and the ground would exert breaking force on you. You would fall *** over tea kettle. The only way you can keep this from happening is having your feet (ideally) stationary relative to the ground as they touch down (i.e. moving back relative to your body). Then there's no horizontal force on your feet so you can continue at constant velocity. Do you now see the error of your ways?

That isn't possible: your foot is not directly below your cog, so it must apply substantial horizontal force as well.

This is a good point. If the force a runner exerted on the ground was only vertical, there would be a torque unless the foot was under his center of mass. Obviously people don't rotate relative to the ground as they run, so there must be an explanation. Either (a) the force actually goes through the runner's center of mass, meaning that there is a net breaking force as the foot lands and a propulsive force as the foot lifts off, or (b) the torque actually does begin rotating the person, but the motion of the runner's arms and other leg prevents the torso from rotating. I honestly think (a) is more likely (or a combination of both), which means I was wrong to say that a runner with good form can exert no breaking or propulsive force once he's in motion.

This would also explain why it's often more efficient to "pop" off the ground, and spend less time actually touching the ground. If the foot lands later and lifts off earlier, the force would be more vertical and less horizontal.

 

[luigidorf]

You're forgetting that acceleration includes change of direction. The leg that is supporting the runner's weight is moving backward. Then the weight shifts and the runner has to exert force to change the direction of motion of the leg that was just moving backward. He has to bring it forward relative to his center of mass. The horizontal motion of his legs happens independently of 1st law conditions that apply to his torso, etc. Look at it this way: his feet are always in a different inertial frame than the rest of him. Half the time they're at rest in the ground frame, the other half, they're in a third frame moving forward faster than his COM.

You have to remember that his legs are connected to the rest of his body (and to the other leg). We can agree that the runner's torso doesn't change horizontal velocity, so let's look at both legs combined. In the middle of the runner's landing, his legs are bend and the mass of the legs in concentrated under the torso. He then moves one leg forward and the other leg backward. There's no net change in the center of mass of his two legs. What you say would probably be true if the runner only had one leg, but each leg balances the other leg's motion.

 

[zoobyshoe]

You have to remember that his legs are connected to the rest of his body (and to the other leg). We can agree that the runner's torso doesn't change horizontal velocity, so let's look at both legs combined. In the middle of the runner's landing, his legs are bend and the mass of the legs in concentrated under the torso. He then moves one leg forward and the other leg backward. There's no net change in the center of mass of his two legs. What you say would probably be true if the runner only had one leg, but each leg balances the other leg's motion.

Oh, weren't we discussing a one legged runner?

I kid. I think you're right: the center of mass of the legs doesn't change and it is in uniform motion.

 

[A.T.]

I argue that when you're running at constant velocity with ideal form, you don't exert any horizontal force on the ground (except for a slight force to counteract air resistance).

Here you are wrong. See the ground reaction force vector here:

https://www.youtube.com/watch?v=USOYUMN5nwU

GRF is obviously not always vertical. Note that these are real world measured forces. Only the muscles are simulated.

But I agree with your general point that for running at constant speed there is no difference, between TM and ground (downwind at windspeed). Comparison of ground(dashed) and treadmill(solid) for running at 3 m/s:

From: http://jap.physiology.org/content/85/2/764.full

 

[luigidorf]

Here you are wrong. See the ground reaction force vector here:

GRF is obviously not always vertical. Note that these are real world measured forces. Only the muscles are simulated.

But I agree with your general point that for running at constant speed there is no difference, between TM and ground (downwind at windspeed). Comparison of ground(dashed) and treadmill(solid) for running at 3 m/s:

From: http://jap.physiology.org/content/85/2/764.full

You're right! I already acknowledged such things:

With walking it's pretty hard to walk without exerting a significant breaking force, which would mean a propulsive force would be required somewhere in the stride.

If the force a runner exerted on the ground was only vertical, there would be a torque unless the foot was under his center of mass. Obviously people don't rotate relative to the ground as they run, so there must be an explanation. Either (a) the force actually goes through the runner's center of mass, meaning that there is a net breaking force as the foot lands and a propulsive force as the foot lifts off, or (b) the torque actually does begin rotating the person, but the motion of the runner's arms and other leg prevents the torso from rotating. I honestly think (a) is more likely (or a combination of both), which means I was wrong to say that a runner with good form can exert no breaking or propulsive force once he's in motion.

I should point out that walking isn't the same as running. Also, the running graphs that you found (thank you by the way) are from a runner with poor form: the spike that occurs around 0.1 s on the vertical force graph is representative of a heel-strike, which results in a significant breaking force and thus requires more propulsive force. For a runner with good form, there would probably still have to be a slight horizontal force to prevent torque, but I think it would be significantly less than 200 N.
​

 

[A.T.]

Also, the running graphs that you found (thank you by the way) are from a runner with poor form: the spike that occurs around 0.1 s on the vertical force graph is representative of a heel-strike, which results in a significant breaking force and thus requires more propulsive force. For a runner with good form, there would probably still have to be a slight horizontal force to prevent torque, but I think it would be significantly less than 200 N.

200N seems consistent with this:

http://jeb.biologists.org/content/203/2/229.full.pdf
Where they tested 8 "experienced treadmill runners" (control without modifications).
Peak breaking force : 221±5N
Peak propelling force :169±4N

And this:
http://www.mlmixrunning.com.br/artigos/subida_descida.pdf
Level TM control:
Peak breaking force : 195±21N
Peak propelling force :169±12N

But maybe you mean forefoot strikes? They have no heel strike peak:

https://www.youtube.com/watch?v=TjrEyfQC5NQ

Compared to heel striker:

https://www.youtube.com/watch?v=wuBLkKnNKm4

I assume the horizontal peak is smaller as well for forefoot strikers. But the horizontal impulse might be the same, just distributed over a longer time span.

But regardless what the horizontal forces are, there is no mechanical reason for a difference between a treadmill at constant speed and running on ground with zero relative wind and constant net running speed. And the measurements with the instrumented TM show the same forces as on ground.

 

[DaveC426913]

But regardless what the horizontal forces are, there is no mechanical reason for a difference between a treadmill at constant speed and running on ground with zero relative wind and constant net running speed. And the measurements with the instrumented TM show the same forces as on ground.

He has acknowledged this (actually, it was his stance all along, it was his friend that thought there'd be a difference.)

He's since moved on to this idea of horizontal force.

See Post 6.

 

[sophiecentaur]

Note that once you are moving, Newton's 1st law demands that all horizontal forces sum to zero. So the only net/external forward force you provide is against wind resistance.

The vast majority of the energy expended while walking or running is in supporting (or bouncing!) yourself against gravity. And the longer your stride, the lower the angle and therefore the greater the force.

If you are not accelerating, the mean horizontal force is zero BUT - When you put your leg forward, when you are running at constant speed, and it hits the ground, I think there is an initial force / impulse 'against' your motion. You don't land on your front foot with the foot stationary relative to the ground immediately before contact. This will involve losing some energy as you absorb this impulse (a loss mechanism) and you need to make up for this by pushing forwards during the subsequent pace. In addition to this, there is energy lost as you sink down and lift up during each pace. There is a certain amount of 'energy return' in the resilience of the tendons but when muscles are under load they are actually expending energy ( they are not just like springs). This implies that running forward at constant speed takes more energy than running on the spot.

 

[A.T.]

But regardless what the horizontal forces are, there is no mechanical reason for a difference between a treadmill at constant speed and running on ground with zero relative wind and constant net running speed. And the measurements with the instrumented TM show the same forces as on ground.

it was his stance all along

I know. And I agreed with him on this. See post #34.

 

[luigidorf]

200N seems consistent with this:

http://jeb.biologists.org/content/203/2/229.full.pdf
Where they tested 8 "experienced treadmill runners" (control without modifications).
Peak breaking force : 221±5N
Peak propelling force :169±4N

And this:
http://www.mlmixrunning.com.br/artigos/subida_descida.pdf
Level TM control:
Peak breaking force : 195±21N
Peak propelling force :169±12N

But maybe you mean forefoot strikes? They have no heel strike peak:

I assume the horizontal peak is smaller as well for forefoot strikers. But the horizontal impulse might be the same, just distributed over a longer time span.

I was indeed referring to forefoot (and some mid foot) strikes. I was careful to refer to runners with "good form." Heel striking is bad form, and even if the runners in that study were "experienced," they likely still heel strike due to growing up wearing shoes. This discussion actually sheds quite a bit of light on why forefoot striking is more efficient: the spike in vertical force also suggests a spike in horizontal force, which means more propulsive force must be exerted each stride.

 

[A.T.]

But the horizontal impulse might be the same, just distributed over a longer time span.

This discussion actually sheds quite a bit of light on why forefoot striking is more efficient: the spike in vertical force also suggests a spike in horizontal force, which means more propulsive force must be exerted each stride.

See above. Smaller horizontal peak force doesn't imply "more efficient". The total horizontal momentum transfer (impulse) could still be the same.

You would have to compare the horizontal impulses for forefoot and heel strikers. Maybe these guys have already done so:
http://barefootrunning.fas.harvard.edu/