Question for Brian Gordon. Shoulder rotation interaction with elbow extension

[spacediver]

Hi Brian,

Very much enjoying your contributions to this site, and I have a question for you.

I'm having some difficulty understanding how the internal rotation of the shoulder can add power to the serve at the same time that elbow extension can.

The reason for this confusion stems from the observation that the rotational energy involved with shoulder rotation is around an axis that is orthogonal to the rotational energy involved with elbow extension.

My gut feeling is that the answer to this question is something like this:


When you have full shoulder abduction (so that your arm is pointing up to sky and shoulder is touching ear), internally rotating the shoulder will contribute virtually nothing to racquet head speed in the desired direction, but elbow extension will contribute a great deal.

(edit: actually, on second thought, internal rotation will contribute something, because it is powering a rotation of the racquet about its own axis - I think I may have answered my own question?)

On the other hand, if your upper arm is at a right angle to your torso, and your forearm is at a right angle to your upper arm (in this position, you'd look like you were waving), the situation is reversed: elbow extension contributes nothing, while internal rotation of the shoulder contributes a lot.

So perhaps when you have your upper arm at an angle intermediate between these two extremes, both motions can contribute something. Indeed, when I look at footage of people serving, they do appear to adopt an intermediate position.

I'm not convinced that this is the correct explanation, however, since in any possible position, whether it's extreme or intermediate, the rotational axes are always going to be orthogonal to one another.

How can there thus be any kinetic link between these two movements?

Thanks,

Marwan Daar

[spacediver]

Here's a pic that illustrates my question better:



After stepping through that animation from different angles, it seems as if the only contribution of internal shoulder rotation to racquet head speed is through pronation of the racquet during the final moments before contact.

Not only that, but the shoulder doesn't even seem to be rotating internally until the end (despite the high levels of the green bar during the firsts two thirds of the upward swing).

So in that particular animation, it seems as if elbow extension is contributing much more to the velocity of the racquet head. Internal rotation seems to provide only a basic twisting motion at the end (with not too much leverage).

[spacediver]

I may have figured out the answer to my own question. I had forgotton to factor in torso rotation and horizontal adduction.


So imagine you're in a hand waving position, with shoulders square to net.

At this moment in time, internal rotation of the shoulder will move the hand in the direction of the net (ignore racquet for this thought experiment).

But if we add in a counterclockwise rotation of the torso, then by the time the shoulders are perpendicular to the net (line between left and right shoulder extends towards net), then elbow extension will move the hand in the direction of the net.

So I guess that this is perhaps a major part of the answer to my question. The angle of the torso relative to the net greatly modulates the relative contributions of internal rotation and elbow extension to useful racquet head velocity.

In fact, you can achieve this full modulation without torso rotation, but solely with horizontal adduction.

[BrianGordon]

Hi Marwan-

This is a pretty lengthy question and covers a variety of aspects (and three posts) - it is clear you have spent some time thinking about it - glad I'm not the only sicko - anyway, I'm up to my ears in stuff right now - but pledge to (attempt to) answer your questions this weekend when I have the time your question deserves - Brian

[spacediver]

thanks Brian, I appreciate it. I think I resolved the apparent paradox that was haunting me, but would love to hear any insights you have to add.

[BrianGordon]

Marwan-

I'm glad you answered the question in your mind because I really am not exactly sure what you are asking - I guess the kinetic linkage part is what I don't get.

So let's try this - it is important to delineate kinematics and kinetics - the contribution parts of the article(s) are purely kinematic linkage findings - that is they describe linkage between the various joint rotations and racquet speed without attempting to address how the joint rotations are driven kinetically (force, torque, etc).

In that regard, it is purely mathematics that allows one to take the measured joint angular velocities, along with the location vector of the racquet face center relative to the respective joint to determine the contribution of each joint rotation - essentially a summation of: linear velocity [racquet] = angular velocity [joint] cross (times) relative location vector. The challenge ends up being measuring some joint angular velocities as I have discussed here and elsewhere.

Doing this allows one to see that the elbow extension contributes most in the mid portion of the upward swing, FOLLOWED by shoulder internal rotation in the final instants prior to contact - which I think you concluded. If the elbow is straight when internal rotation exerts its influence it causes the forearm and hand to rotate (because they are connected) which is linked to racquet motion. The details of the relationship between the shoulder internal rotation and forearm twist rotation were addressed in another thread.

Interesting to me is that a common problem arises (particularly among junior players) when these sources are combined - that is, when the elbow is still extending as shoulder internal rotation kicks in, the result is that the elbow extension axis is rotated to the extent that late elbow extension becomes a negative contributor (decreases) racquet speed. Correct sequencing in this regard is critical.

The kinetic linkage between the two is another and separate story. This is the story I tried to tell using the graphic for which you included a screen shot. Perhaps the confusion this presented is that there was a large internal rotation torque at the shoulder joint yet the upper arm was not internally rotating or doing so slowly - if so, it is important to understand that an internal torque does not imply actual internal rotation of joint/segment - it could be used to slow external rotation of the adjacent segment - and as related to the shoulder joint and upper arm, was identified as an important attribute of high level servers in upward swing p1 (I think).

So... hopefully this answers any lingering doubts about your solution - Brian

[spacediver]

Thanks a lot for the great reply Brian. Your explanation about how torque is independent of joint rotation was especially illuminating to me, and will clear the path for me to gain a deeper understanding of the biomechanics of the serve. Without this information I would have been struggling to reconcile verbal explanations about the order in which joints move, with the torque force across time.

My original question had to do with the transfer of force making an apparent "orthogonal transgression". In a kinematic linkage involving motion dependent torque, or passive transfer of momentum, I failed to understand how a joint rotating around one axis could transfer energy to a subsequent joint rotating around an orthogonal axis. The last paragraph and figures should explain this orthogonal transgression idea better.

There were two insights that resolved this confusion.

First insight: Torque generated along one axis can contribute to racquet head speed by causing the racquet to twist about the axis that defines the racquet's length, while the second orthogonal torque can simultaneously contribute to racquet head speed by rotating the racquet about its buttcap (or close to the buttcap). A case in point here, which we both have referenced, is the internal shoulder rotation providing the twist of the racquet, and the elbow extension providing the "rotation" of the racquet. This would occur when upper arm is pointing towards sky, so ignore the figure below for this instance. The leverage advantage afforded by the elbow extension is probably greater than the leverage advantage provided by the internal shoulder rotation (unless you had shoulders a metre thick!).

Second insight: you can actually have two orthogonal torques that contribute to racquet head speed, and have both torques contribute to the same racquet rotation! Look at the figure below, and imagine that there is a tennis ball being grasped in the hand of the guy on the farthest left. Suppose he internally rotates his shoulder, and releases the ball (second guy). The ball will travel at a certain velocity in the downwards direction. Suppose that instead of rotating the shoulder, he extended his elbow and released the ball (third guy). The ball will travel at a certain velocity in the leftward direction (our left). Now suppose you wanted to channel momentum from the shoulder rotation to the elbow extension. You're only allowed to generate torque through shoulder rotation, and any torque experienced at the elbow joint is motion dependent, or passive. You can see that this is impossible! UNLESS we simultaneously perform a transverse adduction of the shoulders (fourth guy).

[spacediver]

actually, now I'm not even sure if adducting the shoulders would make passive transfer of momentum possible. Either way, I'll figure this out eventually, even if I have to construct a working model with a two way hinge or something.

btw Brian, the interactive module is incredible, and you've done a great job at teaching us how to use it.

[bottle]

Very interesting. I just looked up "orthogonal"-- don't know how I lived without it for my first 70 years. Your pictures help with "adduction," another
word I've had trouble with even though I took four years of Latin.

In my last post I think I used "helicoptering" for "adduction" and even considered calling this "The Angie Effect" after a scene in an early Vic Braden film where Vic grabs the folded up elbow of a young girl named Angie and
whirls or adducts or helicopters it around in toward body median.

I remember being horrified. Here was VHS violation of everything I'd learned about keeping elbow BACK. But today I'm serving this way, i.e., with huge
adduction, and have obtained some good serves thus far.