I’m working on a project for an embedded system course, and I’ve decided on using the Arduino board. My project goal is to measure the speed and maybe angle of a golf club head at the point that it strikes the ball. To prototype, I will likely be attaching a mini breadboard to a golf-club-length shaft with an accelerometer chip, which I already have (ADXL-335 Sparkfun breakout), hardwired to the Arduino safely set aside. I’m hoping to use the measurements that it will print to the Serial Monitor to calculate speed at the time where it would be hitting the ball.
Now, for the actual golf ball, I won’t actually have anything to hit, maybe just a tee for reference. What I planned on doing in order to calculate the time where the club will be passing through where the ball would normally be is use an IR LED and a photosensor set up around the “ball”. When the golf club is swung through the two over the tee, the club head will block IR light from the LED, and the photosensor will record a drop in value. This will be used to calculate when the golf club leaves the “tee” area to start recording accelerometer values, and when it is swung through, in order to get a point-in-time of the “hit” to see where the speed and angle needs to be measured for.
I know that this won’t be easy, but I was wondering if anyone had any recommendations whatsoever. I feel like the IR LED and photosensor system is shaky. I’m afraid it might not be quick enough to detect a golf club head zipping by. If not, are there any other systems that I could consider? In terms of gathering information and calculating values, I had planned on taking the values from the Serial Monitor and using MatLab, either copying and pasting directly myself, or using some external software to do this. Is there anything recommended here?
I’m hoping to use the measurements that it will print to the Serial Monitor to calculate speed at the time where it would be hitting the ball.
Think about the acceleration of an object travelling at constant speed. It's 0.
Prior to hitting the golf club, on the downswing, you have a very complicated mathematical problem on your hands, further complicated by noisy accelerometer measurements. The acceleration the club head observes is a combination of gravitational acceleration and centripetal acceleration. Teasing a velocity out of that is not going to be easy.
Frankly I'd forget the accelerometer and elaborate your IR LED/photosensor to construct a "speed trap". Basically, use two of them. Measure the time between when one photosensor trips and the next one trips and calculate velocity from that. The farther apart they are, the better your resolution, but the less of the speed right before impact you'll be getting.
IR LED's/photosensors are plenty fast for what you're doing. Assuming a temporal resolution of 1 microsecond (reasonable I think) and photosensors spaced 1" apart, a 120mph golf swing will take about 473 microseconds, or 473 units of resolution.
--
The Gadget Shield: accelerometer, RGB LED, IR transmit/receive, speaker, microphone, light sensor, potentiometer, pushbuttons
The system you're suggesting seems like it would be a good option with the two photosensors. That would definitely work, with some tweaking, to gauge speed. The only problem is that the assignment description calls for interfacing 2 chips/systems, which is why I originally planned on using the accelerometer. I will, no doubt, be able to come up with something else to add on top of the photosensors, but I wont be able to do anything about club-head angle.
RuggedCircuits:
Think about the acceleration of an object travelling at constant speed. It's 0.
True, but the club head will be constantly accelerating given its circular movement.
Thanks for the help, I'll try to keep this post updated with code and updates.
Can the arduino collect data fast enough? From doing similar things I am not sure...
How many readings Where will you collect them? Time between readings?
Thanks for the help WillR. The info on the speed of the ball as calculated from the velocity and mass of the club will be really useful information for the project.
As of right now, I think I will be going with the suggested system using two photosensors and measuring the time delay between the two. I'm have actually stumbled across two 3V powered Red-Lasers, which I believe I can just focus on the photosensors for a more precise detection system. I'll post the code up here for anyone interested when I move further with it.
I'm being stubborn though, and I still want to find some use for the accelerometer. Partly because I think the professor would prefer it, and partly because I think if I can get it to work (not easy, to be sure) it could add a nice touch.