"Accelerate evenly through the ball"

bhupaes

Professional
First and foremost, all this physics stuff should be purged out of one's mind while playing - let's be clear about that! :)

That said, I make sense out of all of this as follows. If you recall how I defined extension (in the extension discussion a few days ago), I said extension reaches a maximum just before impact. At that point, be it a forehand or serve, the hand can go no further, and reaches a maximal point where it slows down a lot. This sudden slowing down of the hand releases the wrist, and the racquet head accelerates to impact. At some point after impact, the wrist can release no further and starts dragging the hand/forearm, which causes the hand/forearm to pick up speed after impact into the follow through.

Well, what does this imply for coaches... that's a good question! I would say a really relaxed wrist, and a good runway into extension by laying the wrist back so there's enough room to accelerate the racquet head. I know I'm not being precise here, but you get the idea.
 
C

chico9166

Guest
And this is why I focus on the players intent.
We can find data that shows lots of bits and pieces.

Maybe the intent to accel the hand only results in less Decel, so
that is fine as well.
If i understand you correctly, then I'd hate to see you use a hammer, or axe, or any other tool. Does the handle move at the same speed as the end point?
 

5263

G.O.A.T.
Does the handle move at the same speed as the end point?

This is exactly what we are trying to avoid from traditional tennis.

Not sure how you got that from me saying the hand is what we directly control, but the control of the hand and the technique is what brings the racket face around with acceleration beyond the hand.
We don't have direct control of the racket face, but control thru technique and
via the handle thru the shaft.
 

TennisCJC

Legend
Man, this has gone off the deep end.

If you want to hit a controlled rally shot, accelerate thru the contact zone with a smooth stroke. If you want to hit a harder shot, accelerate thru the the contact zone at a faster pace. Neither is slow and neither is out of control. The concept of accelerating thru eliminates deceleration which is the kiss of death on groundstrokes and serves.

I think the charts and graphs probably say the same thing but they hurt my head. Swing thru the damn ball and finish strong.
 

5263

G.O.A.T.
Man, this has gone off the deep end.

If you want to hit a controlled rally shot, accelerate thru the contact zone with a smooth stroke. If you want to hit a harder shot, accelerate thru the the contact zone at a faster pace. Neither is slow and neither is out of control. The concept of accelerating thru eliminates deceleration which is the kiss of death on groundstrokes and serves.

I think the charts and graphs probably say the same thing but they hurt my head. Swing thru the damn ball and finish strong.

Pretty good way to say it.
 

sureshs

Bionic Poster
Double pendulum seems to be the wrong model. Look at the clip - the bar representing the lower arm almost stops while the racket continues moving up into the ball. It happens because the racket has energy from its swing on the other side.

Now, try to hit a top spin forehand by decelerating and then stopping the lower arm when it is about vertical and tell me whether the racket moved up by itself and hit the ball!

That is why you get the wrong plot that hand velocity is 0 when racket velocity is maximum.

I have objected to the pendulum analysis when it was put forward in the polarization threads.
 

5263

G.O.A.T.
Double pendulum seems to be the wrong model. Look at the clip - the bar representing the lower arm almost stops while the racket continues moving up into the ball. It happens because the racket has energy from its swing on the other side.

Now, try to hit a top spin forehand by decelerating and then stopping the lower arm when it is about vertical and tell me whether the racket moved up by itself and hit the ball!

.

I was noticing this too.
You can do it and many beginners start something like this with very poor results, lol.
 

sureshs

Bionic Poster
I was noticing this too.
You can do it and many beginners start something like this with very poor results, lol.

In the light racket clip, the racket does not move up much. I was kind of OK with that, except I felt what purpose does it serve. Then the heavy racket clip showed the racket moving up much more, and again I was like, OK, but who hits the ball like that?

I can imagine that this model MIGHT describe a golf put, but not a tennis shot. Even feeding the ball requires more effort than that.

The hand and the racket have to move together - one cannot stop and hope that inertia will swing the racket up into the ball just by keeping a loose wrist.
 

5263

G.O.A.T.
Double pendulum seems to be the wrong model. Look at the clip - the bar representing the lower arm almost stops while the racket continues moving up into the ball. It happens because the racket has energy from its swing on the other side..

THis is also why I did not feel I could depend on the data, especially
in reference to the hand slowing in the graph.
I think with a true modern swing, you will see the speed stay up there better
with the hands and the racket, also
for a slightly longer time than Cross' graph shows.
 

sureshs

Bionic Poster
THis is also why I did not feel I could depend on the data, especially
in reference to the hand slowing in the graph.
I think with a true modern swing, you will see the speed stay up there better
with the hands and the racket, also
for a slightly longer time than Cross' graph shows.

I think it was a case of picking the only studied mathematical problem which even remotely resembles the real case. You see this in textbooks, where the purpose is to give the student a very broad idea of the actual thing to help them visualize. But in this case, the analysis does not seem to be useful to anyone.
 

julian

Hall of Fame
Clarification

Yes, I can appreciate your comment about the OP's intent.

Which charts, in particular, are your referring to? Fig 10 on this page shows that the angular velocity (magnitude) of the racket decreases after contact while the angular arm velocity increases again.

I'm a bit reluctant to use the term, acceleration, since it is a vector. Acceleration describes the rate of change of both the magnitude and the direction of velocity. The magnitude of the velocity can be constant while the direction is changing -- this would still be considered acceleration.

Cross uses the term "angular velocities of the forearm and the racket"
in the caption of Fig 4 of
http://www.physics.usyd.edu.au/~cros...ennisDPend.pdf
Both of them are obviously scalars,NOT VECTORS
These terms are used,I believe.first time at the caption of Fig 1a of the same reference
It comes without saying that the ratio of rotational energy of the racket/rotational energy of the arm
has good physical meaning
 
Last edited:

bhupaes

Professional
THis is also why I did not feel I could depend on the data, especially
in reference to the hand slowing in the graph.
I think with a true modern swing, you will see the speed stay up there better
with the hands and the racket, also
for a slightly longer time than Cross' graph shows.

The hand/forearm slowing down (not stopping!) before contact is very obvious in the serve.

For the forehand, what seems to be happening is that some time before contact there is a similar slowing down due to maximal extension. Also, the hand starts pulling the racquet in, which causes the forward component of the hand velocity to decrease, thus releasing the wrist and accelerating the racquet head - which is what matters. IMO this is what Rod's charts and graphs are showing, when you look at it qualitatively. The double pendulum theory and its associated equations would be useful for quantitative analysis, I suppose.
 

5263

G.O.A.T.

Now this could be an interesting perspective the subject.

Sure would like to get a look at their data with this!
 

5263

G.O.A.T.
Also, the hand starts pulling the racquet in, which causes the forward component of the hand velocity to decrease, thus releasing the wrist and accelerating the racquet head - which is what matters. IMO this is what Rod's charts and graphs are showing, when you look at it qualitatively. The double pendulum theory and its associated equations would be useful for quantitative analysis, I suppose.

Would be cool to see this data with the 1st pendulum being pulled back at the right time or better just continuing in a tighter radius which pulling right would give. By the way, this change of direction alone is accel in science terminology.
Right?
 

bhupaes

Professional
Would be cool to see this data with the 1st pendulum being pulled back at the right time or better just continuing in a tighter radius which pulling right would give. By the way, this change of direction alone is accel in science terminology.
Right?

Absolutely - changing the direction of a moving object requires the use of force, which by definition results in acceleration.
 

bhupaes

Professional
Toly, I stole your beautiful picture! Sorry!

zupov4.jpg


To make my point, you can see how the extended forearm has moved very little between frames 1 and 8, yet the wrist has come around completely.
 

sureshs

Bionic Poster
Toly, I stole your beautiful picture! Sorry!

zupov4.jpg


To make my point, you can see how the extended forearm has moved very little between frames 1 and 8, yet the wrist has come around completely.

If you are saying this is like a double pendulum, it cannot be. A pendulum uses gravity, and here the motion is mostly in the horizontal plane, not vertical. The action has to be forced by the forearm and the wrist, and is not passive.
 

5263

G.O.A.T.
Toly, I stole your beautiful picture! Sorry!

To make my point, you can see how the extended forearm has moved very little between frames 1 and 8, yet the wrist has come around completely.

I think it is just a bad camera angle mostly and is not supported by anything I've seen from better angles or from above. Evidence of the across aspect is in how the ball has moved from the throat area all the way
out to the outer part of the upper hoop in just these few fast frames.
 

user92626

G.O.A.T.
I'm with 5263. That has to be a bad cam angle or Fed was doing some sort of ultra speed slappy inside out fh to take away time from opponent. It cannot be representative of a regular, consistent and sound FH, ie hitting with a nonpassive forearm and wrist as suggested.
 

sureshs

Bionic Poster
I'm with 5263. That has to be a bad cam angle or Fed was doing some sort of ultra speed slappy inside out fh to take away time from opponent. It cannot be representative of a regular, consistent and sound FH, ie hitting with a nonpassive forearm and wrist as suggested.

Yes, he is improvising from outside the alley to angle the ball away to such an extent that the opponent cannot reach it, otherwise he will be toast in the open court.
 

user92626

G.O.A.T.
Yes, he is improvising from outside the alley to angle the ball away to such an extent that the opponent cannot reach it, otherwise he will be toast in the open court.

Good point/observation. No decent pro, let alone Fed, plays a usual shot and a usual FH from that position. That must be the last shot or he's screwed.
 

bhupaes

Professional
I think it is just a bad camera angle mostly and is not supported by anything I've seen from better angles or from above. Evidence of the across aspect is in how the ball has moved from the throat area all the way
out to the outer part of the upper hoop in just these few fast frames.

This is an extreme case, I agree. Fed's normal TS FH would show more across movement (you can see some pulling of the racquet inwards here also, as you have pointed out). But this illustrates the point I was trying to make very well - the slowing down of the arm, the across movement, and the accompanying acceleration of the racquet.
 

SystemicAnomaly

Bionic Poster
Two very different points:
Please specify whether
"not intend to accel" does refer to a RACKET or HAND or both?
Does it refer to FOREHAND or SERVE or BOTH?

Comment:
Once more Rod Cross says that his graphs are in agreement with the pro data
Additionally data are are the ROTATIONAL part ONLY
Do we doubt pro data?

I would think that an "intent to accelerate" would refer to increasing the speed of the stringbed/racket head rather than the hand. It's the collision point where we want to transfer energy/momentum to the ball. I would think that it could refer either to the serve or the FH. However despite our best efforts, the racket head loses speed at contact. By attempting to "accelerate thru contact", we ensure that we are still accelerating just prior to contact. Note: if we truncate our follow-thru, there is a good chance that we are slowing down just prior to contact.

As for the data and the graphs:

The data for the TWU graph that I show in post #41 is the very same data from the very same source that is presented in Fig 4 of your first links. The only difference is that y-axis (the angular velocities) is expressed in different units and the time axis is truncated at 0.3 seconds for Fig 4. It is my belief that the indication of the "impact time" is an overlay -- something that was tacked on to the graphs.

Figure 4 is from a longer document from 2010 while Rod Cross's TWU article was published in June of 2011. It is my belief that the impact time indication is a goof that was overlooked in the longer document and subsequently corrected in the later document.
.
 
Last edited:

SystemicAnomaly

Bionic Poster
If you are saying this is like a double pendulum, it cannot be. A pendulum uses gravity, and here the motion is mostly in the horizontal plane, not vertical. The action has to be forced by the forearm and the wrist, and is not passive.

But then gravity is just really just a force. In this case the force is provided by a different means. The double pendulum idea might be something of a first order approximation of what is happening. Would a trebuchet provide any further insight into what is happening?
 

sureshs

Bionic Poster
But then gravity is just really just a force. In this case the force is provided by a different means. The double pendulum idea might be something of a first order approximation of what is happening. Would a trebuchet provide any further insight into what is happening?

When the pendulum is moving up, gravity still acts downwards, pulling it back. It is not at all like the force supplied by the hand.
 

toly

Hall of Fame
Toly, I stole your beautiful picture! Sorry!

zttdg4.jpg


To make my point, you can see how the extended forearm has moved very little between frames 1 and 8, yet the wrist has come around completely.
I hope that in the end you realize that in case of Federer flat FH passive/active hand action can contribute more than 50% of the power. :)
 
Last edited:

bhupaes

Professional
I hope that in the end you realize that in case of Federer flat FH passive/active hand action can contribute more than 50% of the power. :)

Let's say the exact percentages of passive and active contribution are still under consideration - or should I say negotiation? :) There's no way to tell from pictures, as far as I know!
 

toly

Hall of Fame
Hi,
I believe there are two POSSIBLE models:
1.an elastic collision - the conservation of momentum is satisfied
2.an inelastic collision-a ball is deformed during the collision
the conservation of momentum is NOT satisfied
It is very well known fact; momentum has the special property that, in a closed system, it is always conserved. Kinetic energy, on the other hand, is not conserved in collisions if they are inelastic.:shock:
 

toly

Hall of Fame
Nice vid, but made it even more clear how strongly he is working across that ball. It probably had nearly pure side spin on it.
From physics and math point of view, there is practically no tangential component of the racquet point of contact velocity (vector) during impact. Thus, there is no spin at all. Do not argue with the lows of physic please. It doesn’t make any sense. I’ll try to provide picture which (maybe) support my explanation. :(
 
Last edited:

SystemicAnomaly

Bionic Poster
It is very well known fact; momentum has the special property that, in a closed system, it is always conserved. Kinetic energy, on the other hand, is not conserved in collisions if they are inelastic.

Already made that point a while back.

Cross uses the term "angular velocities of the forearm and the racket"
in the caption of Fig 4 of
http://www.physics.usyd.edu.au/~cros...ennisDPend.pdf

Both of them are obviously scalars,NOT VECTORS
These terms are used,I believe.first time at the caption of Fig 1a of the same reference
It comes without saying that the ratio of rotational energy of the racket/rotational energy of the arm
has good physical meaning

My comment was regarding the way that some other posters were using the term, acceleration. I don't believe that they were referring to a change in direction in their use.
.
 
Last edited:

SystemicAnomaly

Bionic Poster
Double pendulum seems to be the wrong model. Look at the clip - the bar representing the lower arm almost stops while the racket continues moving up into the ball. It happens because the racket has energy from its swing on the other side.

Now, try to hit a top spin forehand by decelerating and then stopping the lower arm when it is about vertical and tell me whether the racket moved up by itself and hit the ball!


That is why you get the wrong plot that hand velocity is 0 when racket velocity is maximum.


I have objected to the pendulum analysis when it was put forward in the polarization threads.

Have you discussed this with Rod Cross? I have several email addresses for him if you wish to try to contact him.
 

5263

G.O.A.T.
From physic and math point of view, there is practically no tangential component of the racquet point of contact velocity (vector) during impact. Thus, there is no spin at all. Do not argue with the lows of physic please. It doesn’t make any sense. I’ll try to provide picture which (maybe) support my explanation. :(

I try not argue with Physics, but when it is different than the event, it makes me question.

Are you suggesting he hit a knuckle ball here?

I'd be willing to bet a lot the he got some good spin here and with a good component of side to it.
The vid showed the racket face moving from outside the ball to inside pretty quickly.
Sorry, but I think you are trying to prove a faulty premise on how the wrist is used, and it is
clouding your objectivity.
 

5263

G.O.A.T.
My comment was regarding the way that some other posters were using the term, acceleration. I don't believe that they were referring to a change in direction in their use.
.

But you can be sure I was referring to change of direction, as that is the basis
of what we are saying about the modern Fh in MTM.
 

SystemicAnomaly

Bionic Poster
http://www.physics.usyd.edu.au/~cross/PUBLICATIONS/49.%20TennisDPend.pdf

Compare Fig 4 of the link above with Fig 10 of the TWU link below. Why do they differ? Is one of them incorrect? Am I overlooking something here?


http://twu.tennis-warehouse.com/learning_center/doublependulum.php


DPFig10.jpg
This graph from the TWU link makes more sense to me than Fig 4 from the 1st link. I would expect a reduction in the speed magnitude of the racquet (head) during contact. Fig 4 does not appear to indicate any consequence of the impact at contact time. The TWU graph seems to confirm what we see in the animation below. Can you guys pls enlighten me on why the other graph appears to contradict these other 2 sources?

http://www.science-animations.com/support-files/tennis.swf
.

Way cool! I contacted Rod Cross about the apparent discrepancy between his 2 graphs that show the angular velocities of the racquet and arm. It took him less than 45 mins to get back to me with a detailed explanation. Here it is:

Both graphs are actually the same graph, but I have labelled the impact time in different spots as you correctly point out. The main point that I wanted to show was that the forearm speed drops quite a lot before the player hits the ball. That is probably quite surprising but it makes sense since rotational energy is taken out of the arm and given to the racquet just before the racquet strikes the ball. It would also make sense if the ball is struck when the velocity of the racquet head is at a maximum. Whether that actually happens is a bit uncertain, but it probably does happen. It is uncertain for two reasons:

1. The velocity of the racquet head consists of two parts. One is the part due to pure rotation of the racquet. That is the part shown in both graphs. However, the racquet head is also moving forward because the whole arm is moving forward. In fact, it is possible to strike a ball without rotating the racquet at all, simply by pushing the racquet head forward, as players used to do in the old days with wood racquets. So, the exact time at which the racquet head has maximum speed could be before or after at the same time that the rotation speed (ie angular velocity) is a maximum.

2. In Fig. 4 in the AJP paper, I calculated the angular velocity a bit beyond the impact time, but I did not consider the effect of the impact itself. That is, I let the racquet swing as a double pendulum without actually hitting a ball. If the racquet did hit a ball then the angular velocity would drop suddenly. The sudden drop is not shown in Fig 4 because I was only interested in calculating the speed of the racquet resulting from the forces exerted by the arm. I did not include the effect on rotation speed due to the force of the ball on the racquet.

The impact time labelled in Fig 4 is the point where the racquet is lined up correctly to strike the ball as shown in Fig. 3.

The impact time labelled in the TWU article assumes that the racquet is lined up correctly when the angular velocity is a maximum, in which case the ball will come off the racquet at a different angle. In the TWU article, the angular velocity drops after the impact time, not because the racquet hit the ball but because that is what happens even if the racquet doesn't strike a ball. That is the expected behavior of a double pendulum that oscillates back and forth naturally without hitting anything. If the racquet hit a ball then the angular velocity would drop very rapidly, in 0.005 seconds, to about half its pre-impact value.

Rod Cross
.
 
Last edited:

5263

G.O.A.T.
very interesting peek inside there SA!

Also explains how we can't read too much into this data, as it was not clear there was not an actual impact (although I had suspected that given the graph), even though maybe that info was in there somewhere.

And thanks Rod, for sharing that with us.
 

julian

Hall of Fame
Reduction of the speed of arm

Way cool! I contacted Rod Cross about the apparent discrepancy between his 2 graphs that show the angular velocities of the racquet and arm. It took him less than 45 mins to get back to me with a detailed explanation. Here it is:

Both graphs are actually the same graph, but I have labelled the impact time in different spots as you correctly point out. The main point that I wanted to show was that the forearm speed drops quite a lot before the player hits the ball. That is probably quite surprising but it makes sense since rotational energy is taken out of the arm and given to the racquet just before the racquet strikes the ball. It would also make sense if the ball is struck when the velocity of the racquet head is at a maximum. Whether that actually happens is a bit uncertain, but it probably does happen. It is uncertain for two reasons:

1. The velocity of the racquet head consists of two parts. One is the part due to pure rotation of the racquet. That is the part shown in both graphs. However, the racquet head is also moving forward because the whole arm is moving forward. In fact, it is possible to strike a ball without rotating the racquet at all, simply by pushing the racquet head forward, as players used to do in the old days with wood racquets. So, the exact time at which the racquet head has maximum speed could be before or after at the same time that the rotation speed (ie angular velocity) is a maximum.

2. In Fig. 4 in the AJP paper, I calculated the angular velocity a bit beyond the impact time, but I did not consider the effect of the impact itself. That is, I let the racquet swing as a double pendulum without actually hitting a ball. If the racquet did hit a ball then the angular velocity would drop suddenly. The sudden drop is not shown in Fig 4 because I was only interested in calculating the speed of the racquet resulting from the forces exerted by the arm. I did not include the effect on rotation speed due to the force of the ball on the racquet.

The impact time labelled in Fig 4 is the point where the racquet is lined up correctly to strike the ball as shown in Fig. 3.

The impact time labelled in the TWU article assumes that the racquet is lined up correctly when the angular velocity is a maximum, in which case the ball will come off the racquet at a different angle. In the TWU article, the angular velocity drops after the impact time, not because the racquet hit the ball but because that is what happens even if the racquet doesn't strike a ball. That is the expected behavior of a double pendulum that oscillates back and forth naturally without hitting anything. If the racquet hit a ball then the angular velocity would drop very rapidly, in 0.005 seconds, to about half its pre-impact value.

Rod Cross
.
It looks like the forarm speed drop depends heavily on values of couples C1 and C2
 

toly

Hall of Fame
So is it true that the forearm slows down before impact? I am not just not able to digest this.
You cannot digest this because this is not true.

Rod Cross just calculated conversion of potential energy into kinetic energy for the simplest double pendulum model. It has absolutely nothing to do with real tennis strokes. There are practically no correlation between real tennis data and Rod Cross double pendulum model of tennis strokes.

IMO, it would be good idea to forget about this stuff completely.:)
 
Last edited:

jmjmkim

Semi-Pro
the racquet head should be accelerating as you make contact. that means the racquet is going faster at contact than it was 5 cm before contact. and rhs at 5cm before contact is faster than it was 15 cm before contact and so on and so on.

Same with a golf swing. That's why a pro swing looks effortless, and so easy. They start the swing with just the natural unwinding of the body coil, and then release the "wrist" just before it reaches the ball, thus producing tremendous amount of leverage and club head acceleration at impact. In tennis, it is this acceleration "thru" the ball that makes that "POP" sound when good players hit.

This is much, much harder to do on the move, or on the run. That's why without the footwork to get you in place ahead of time and initiating the swing process before the ball comes, you produce weaker and inconsistent shots.
 

sureshs

Bionic Poster
You cannot digest this because this is not true.

Rod Cross just calculated conversion of potential energy into kinetic energy for the simplest double pendulum model. It has absolutely nothing to do with real tennis strokes. There are practically no correlation between real tennis data and Rod Cross double pendulum model of tennis strokes.

IMO, it would be good idea to forget about this stuff completely.:)

I also think so
 

5263

G.O.A.T.
You cannot digest this because this is not true.

Rod Cross just calculated conversion of potential energy into kinetic energy for the simplest double pendulum model. It has absolutely nothing to do with real tennis strokes. There are practically no correlation between real tennis data and Rod Cross double pendulum model of tennis strokes.

IMO, it would be good idea to forget about this stuff completely.:)

And you started out so well here, lol.
Sure, it's far from perfect, but I think there are some reasonable correlations that can be observed.
You and Suresh are probably right that it is a good Idea not to be too concerned over it though.:)
 
Top