all posts post new thread

Kettlebell How fast do we swing Kettlebells?

Status
Closed Thread. (Continue Discussion of This Topic by Starting a New Thread.)
MattM

MattM

SFG1
I've never really been into the science/physics of exercise but I was thinking about this recently.

Have any tests ever been done? Is it 2mph or 20mph? I realize that everyone will be different, just looking for ballpark numbers.

From my high school memory: force = mass x acceleration, right?
 
That force = mass * acceleration formula would apply if the kettlebell was just being swung once. It would become considerably more complicated when you factor in the rapid deceleration and change of direction at the bottom of each swing. The forces generated in that bottom phase of the swing would show some surprisingly high numbers.

It would be interesting to get a high speed camera and film it against scale in the background like they do on Myth busters velocity tests and crunch the numbers.
 
@B.Hetzler has done test with a force plate under the feet while swinging, I believe. If memory serves, some people generated 500 lbs. of force. Perhaps someone can post a link to the article in which this was mentioned.

-S-
 
Let's see. The OP mentions remembering high school physics, so should be able to follow this. Don't hesitate to ask questions if this is not clear. I will try to keep the response brief. First, average speed. As an approximation, a kettlebell travels along a curve that approximates a portion of a circle, an arc segment. Let's approximate this as one third of a circle on the way up, and the same on the way down. doing 10 swings thus has the kettlebell traveling along 10 X 2 / 3 = 6.6 complete circles. What is the radius of the circle? My arms are 28 inches long and the kettlebell handle and ball add about 6 inches for the distance between pivot point and center of mass, for a total of 34 inches. Since I am used to the metric system, I will approximate the radius of the circle along which the bell travels to .85 m. So, for a set of 10 swings, the bell travels 2 X pi X 6.6 X .85 = 35 m. A set of 10 swings takes between 15 and 20 secs, depending on many factors. Let's use 15 secs. In 1 min, the bell travels 4X35 = 140 m. In 1 hour, 60X60m = 8.4 km, so a kettlebell travels 8.4 km in one hour, or about 5.2 mph. Of course, there are a lot of approximations and assumptions and the real answer may be 4 or 6 and will depend on a lot of factors.

Now, that was probably not the question. The question seemed to be for maximum speed. Here again, we can get an approximation using physics. Once past vertical, the bell leaves the body and moves with no force except gravity doing work on it. This assumes that the force exerted by the arm is only radial, perpendicular to the movement and thus does no work. In this case, the potential energy gained by the kettlebell from bottom to top is equal to the kinetic energy of the bell at the bottom of the swing. Assuming that at the end the arm is parallel to the ground, and using the same approximation for radius of movement used above, .85m, we have g h = .5 v^2, with h = .85m, g = 10m/s/s, and solving for v: 4.12 m/s = about 9 mph.

Of course, this is all very approximative, but the order of magnitude is about 10 mph for the maximum speed. Also, this is good because it makes sense when compared to the first calculation regarding average speed.

I would be curious to see what video analysis would give. If someone has the right software, it should not be difficult to get a video from Youtube and measure the distance traveled by the bell between two successive frames. I also think that there exist devices you can attach to a kettlebell that include an accelerometer, and integrating this data would give speed.
 
For those more experimentally inclined, I found an hour today and downloaded a motion analysis software that I had wanted to take a look at for a while, Tracker Motion Analysis, which is completely free for those interested. I know, I have weird ways of spending my free time. That's probably why I like kettlebells ;). I used the software to look at a 2 hands swing performed by Geoff Neupert in his the Big 6 course, which I bought in April. ,Assuming Geoff is 6 feet tall, the maximum speed of the kettlebell center using the software was about 4.6 m/s which is (drum roll)... 10 mph. I did not use any sophistication in marking the center, which I did manually without zooming, so the real value may be off by several percent.

Interestingly, Geoff bends his knees a bit at the bottom of the swing, about 15 degree, just like you can see Pavel doing in The Perfect Kettlebell Swing: Is There Such a Thing?, and the arms leave the body a tenth of a second or so before the hips have finished snapping. This causes the bell to keep on accelerating for a tenth of a second or so as the shoulders keep on rising after the arms have passed vertical due to the hips and knees still opening, so that the maximum speed is not when the arms leave the body, but a bit further up. Also, Geoff swings quite high with the kettlebell reaching eye level, which requires a bit more speed than my theoretical example above.
 
Acceleration on the up part of the swing peaks at around 32 m/s2 in that video. That's around 3 times gravity. In term of force, the maximum force on the kettlebell, assuming a 24kg one would be F = m a = 768 N. 24 kg is simply because in these videos, 24kg is the maximal weight used through the series, and it doesn't make much sense for a big strong guy like Geoff to use a smaller weight. Even 24 kg looked very easy for him. Newtons don't mean much for most people. Converted in the informal unit of force of "pounds", that is 172 pounds. In other words, the force on the kettlebell is the same as the force required to hold against gravity a 172 pounds weight, without moving. These numbers are total acceleration and force, which has a component along the arms and a component along the motion of the kettlebell. I would need to learn more about the software to separate these two from each other. Maybe in a few days...
 
@Manuel Fortin if I upload a video, will you analyze the speed? I'm thinking 24kg, 2H swings, 1 set of 10.

This is great stuff! Really interesting.
 
@Anna C, sure. I must be able to download it on my computer. Simplest is to upload it to your Google Drive/Microsoft One Drive/... and send me a link. If you don't know how to do this, you can upload to YouTube and I will find a way to download it from there.

The video analysis software I use is powerful but basic. I may be wrong, but so far I thing it uses only 2D coordinates. Analysis will be much simpler and accurate if you film from the side, with a camera at about hip height. I could probably do some 3D stuff using Excel, but that would be much more time consuming. The disadvantage of this point of view is that we may lose the kettlebell for a frame when it goes between the legs. However, over 10 swings I am sure I can find a swing where we see bit of the kettlebell on each side of the legs between adjacent frames. For calibration, give me your height, or even better, hold an object of known length besides you, such as a yard stick, at some point in the video.

Other things that would help would be wearing tight fitting clothes, or shorts, of a color that contrasts with the color of the kettlebell so that we see the kettlebell as well and as much as possible as it goes between the legs. A good contrast between the kettlebell and the background would also be better. Thinking about it, if you can mark the center of the kettlebell on its side with a small piece of tape, for example white tape on a black or bell, that would be awesome. That way, tracking the center of the kettlebell becomes really easy.

While we are at it, I have been a lurker on the forum for a while and I remember that the question of the difference between 1H and 2H swings comes up pretty often. If you can film a set of 1H swings too, we could answer a lot of those questions. You could do one with the 24 and one with a 16.
 
Last edited:
Very interesting stuff.

So Manuel - and forgive me, it's been a while since I've used info gleaned from my physics classes - but I'm assuming the calculations you're referring to here are for the upswing, correct?

So, you could do the same calculations for the downswing to determine the force at the bottom of the swing. In other words, confirm the 10g approximation given in S&S that offwidth referred to above. If the downswing force (hardstyle) is similar to the upswing force, would that mean Geoff Neupert would be reversing force at 172 pounds? I would suspect it would be higher due to the reversal, but that's where my physics memory fails me.

Jason
 
Ooh, fantastic. Yes, I can do that camera angle, yardstick or something for measure (I'm 5'8"), and colored shorts. Some white tape on a black bell will be easy. I'll do 10 2H, 10 R, and 10L, will upload both to YouTube to put in this thread and to Google Drive so you can download. I should be able to get this in the next couple of days...

My son is a college student studying engineering and physics. He was able to follow your first post and liked it. :)
 
Yes, that is for the upswing. I did not look at the downswing as it is much more variable, depending on how much down force is exerted. The standard on the upswing is to apply no force with the arms, which is well accepted by people affiliated with Strongfirst. However, there are many ways of doing the downswing. We can propel the kettlebell down with little or no force, or use overspeed eccentrics and really push it hard.

Also, the mechanics of reversing the swing and of doing the upswing differ a bit. When absorbing the swing at the bottom, one can use tissue elasticity to absorb force, to reduce speed over a short distance, giving large forces, or use a large part of the downswing movement to reduce speed over a long distance, giving smaller forces. However, there is no such thing as a slow upswing. That will result in a kettlebell that does not go up.

I cannot find the article mentioning 10g originally, but I read it a long time ago and the force was measured using a force plate if I remember correctly, so that number is a "real" number. Overspeed eccentrics were probably used. Maybe the kettlebell was also heavier (32?).

Finally, to answer your question, Geoff was going really easy in the video and absorbed the downswing over a long distance. Acceleration on the down swing part was smaller (max of 26-27 m/s2) than on the upswing part and he was probably absorbing the swing over a long distance. Looking at vertical acceleration, at the beginning of the swing, it is about -10m/s2, which is acceleration due to gravity. He was not using overspeed eccentrics and was not pushing the kettlebell down, just letting is fall. This was the impression given by the float of the kettlebell in the video.
 
@Anna C If your son wants to give this a try, he can download the software for free at Tracker Video Analysis and Modeling Tool for Physics Education. To get examples of how this is used, he can look up Rhett Allain's columns in Wired at Dot Physics | WIRED. This software is often used in his blog posts. Your son may even have a class project analyzing kettlebell swings in the making. That would similar to what I am doing, except that he would need to learn about calibration, camera distortion, use statistics, .. . Add a force plate to measure force of feet on the ground, calculate also the tangential and radial accelerations in the frame of reference of the shoulder (that gives you for example the grip strength you need), compare the motion of the kettlebell during the top part of the swing and compare to free fall, and there you go, a good science project with experiments and theory that matches first/second year physics.
 
Amazing that his lower back doesn't bend that much for what he is doing. That looks like what would happen if you had not really decided whether you were doing swings or jerks, especially when looking at it in slow motion.
 
@offwidth I thought a bit about this. So we have 3 numbers: 500 lbs, 10g and 24kg (about 53 pounds). I don't think they mean that the kettlebell is accelerated at 10g and this number is bit misleading. I may be wrong here, but my understanding is that someone is swinging a kettlebell while standing on a force plate (just a sophisticated scale). Even without swinging, the force plate should read around 250 pounds (a 200 pounds guy plus a 50 pounds kettlebell). This is probably never removed as a baseline from following calculations. For example, if the kettlebell accelerates at around 3 g, this gives for a 24kg bell my 171 pounds number. Now, for technical reasons, you cannot just add 250 pounds to 171 pounds to get the number registered on the weight plate (for those who took and remember high school or above physics, this has to do with the fact the forces are vectors). Also, the kettlebell is not the only thing moving here. The subject is going from a bent position to a straight position, which will move his center of mass a bit, which also requires a force. Try doing this movement withouth a kettlebell in hands and you will see that you need to exert some force. That's how we get to 500 pounds, even if the kettlebell is not accelerating at 10g. With intent to maximize force production, I can easily see the force plate getting to 500 pounds, and maybe the kettlebell having a max acceleration of 4g or so.
 
Status
Closed Thread. (Continue Discussion of This Topic by Starting a New Thread.)
Back
Top Bottom