Countdown to Launch — How I Made My Own DIY Ping Pong Ball Launcher
“The strain (deformation) of an elastic object or material is proportional to the stress applied to it”
Above is a law of elasticity discovered by the English scientist Robert Hooke all the way back in 1660 but one which we still use every day and, conveniently, that I have also based my ping pong launcher on!
So once again, welcome back to my blog! Or, if you’ve just stumbled across this post in your quest to make the best DIY Ping-pong Propellor known to man, then welcome! Either way, stay with me as I take you through my long, windy, design and construction process.
THE IDEATION PHASE
There was a whole host of options and possibilities which sprung to mind when I first heard the objective of this assignment. Those options were whittled down first by the rules of the assignment and secondly, by the resources I had and thus the feasibility of each of the options.
My first four initial thoughts had been; use springs, use elastic bands, use air, use force/propulsion. Using these four principles I devised the basic, general ideas of some mechanisms, as detailed below, which could each be modified or amended.
As I said though, learning the rules eliminated a few of these. But before some of these rules had come to light, I had dived headfirst into toying with some ideas. The first of which included the elastic band style concept. My family members laughed when I suggested my elastic band slingshot and clothes peg firing pin idea yet my engineering prowess prevailed and my very first ping pong launcher had sprung to life in no time.
My little pug Doug loved this new game of fetch and with some calculations, testing and fixed positioning, this contraption could’ve been extremely accurate. Alas, the final rules did emerge and as I’ll explain later, I eventually decided to scrap this idea in light of them.
So what rules and regulations do I need to adhere to here? Well, the ball must be launched over a 500ml bottle (approx. 20cm) and the launcher must be positioned no higher than 18cm off the ground. There is 20 seconds for the robot to stop and in a target zone and the attached catching surface must be <= 133cm². Points are awarded for the number of balls successfully caught (10 points per ball and only one can be fired at a time) and also for making the catching basket smaller. Also, the launcher can be manually triggered but you can’t manually reload the balls. Yes there’s quite a bit to take into consideration there but let’s see what we can do!
Once again considering the four initial possible categories of launchers, I knew straight off the bat that automatic reloading wouldn’t be possible with the first elastic band example I had drawn up and finding a new way to incorporate it would be difficult; or at least more difficult than it would be for some of the other options. This was also true for the pulley/trebuchet type example.
And then there were two. I did play around the pneumatics idea for a while. I had an air tank and an air compressor which could’ve been put to good use after all. I spent an evening or two (on the hallway floor, where all my best construction seems to take place these days) putting small prototypes together and it seemed liked a pretty good option. And yes, I also could’ve easily incorporated the reloading mechanism and landed a few balls to get some good points… but connecting a tube and valve to the firing cylinder just seemed too easy, and where’s the fun (or creative engineering) in that? Instead, I tried to find bits lying around the house that I could draw air out of but my little brother refused to sacrifice his footballs and besides, I don’t think they’d have enough air inside anyway. I also considered a tire, but the only ones I had were on my car and I wouldn’t be able to make my self-care trip to McDonald’s Drive-Thru if I did that. Also, as a student, Pirelli’s are a bit too pricey to be messing around with, just to shoot some ping pong balls.
So that just left the springs. There was a few variations and options here; would I attach 2 tension springs to the outside of my firing cylinder or 1 compression spring inside the cylinder? Well, I don’t know about you but once I have an idea in my head I just have to try it out. I knew that if I waited, I could ran-sack my dad’s mechanics garage in the morning but that was just too long a wait. So in order to begin my initial construction, I searched high and low around the house instead. Tension springs were nowhere to be found but a compression spring could be, in, would you believe, a toilet roll holder! I was warned I could use it on the one condition I returned it in the same condition I found it but yeah… I think I’m as well to just buy a new one at this stage…
Anyway, most of the initial ideation and thinking was done it was time to get BUILDING!
Prototype 1
In keeping with my bathroom theme, my initial build, to better formulate my concept design, was conducted using empty toilet rolls as the firing cylinder. I also tied together two allen keys which would be used to compress the spring. The slots on the sides housed the protruding allen keys which gave me access to pull them downwards for firing. An annulus attached to the bottom of the cylinder sat the spring and allowed for the exit of the allen keys as they were drawn back. The spherical hole would allow another loading cylinder to be attached. The positioning of this hole meant a ball could only enter the firing cylinder when the spring was completely compressed and ready to be fired. The diameter of the cylinder meant that only one ball could fall into the correct position at a time.
Now of course, as evident in the pictures of the knackered toilet roll, the testing took its toll and the cardboard wasn’t strong enough. This was inevitable however it was never intended to be used in the actual design and instead it served its purpose in helping me understand what I would need to do and to understand the dimensions.
Prototype 2
This version failed miserably I can’t deny. Instead of using an allen key like structure for firing, I tried to attach a disc to the top of the spring and pull twine back down the centre and out the bottom. This version didn’t require the slots or bottom hole and hence reduced the amount of cutting I needed perform on my new firing cylinder, an air freshener cannister. Stupidly, I hadn’t realised that my allen keys (in the first prototype) had been serving a dual function; to compress the spring AND to prevent it from deforming laterally. Without them, my spring wouldn’t pull back as desired as it had the freedom, of the entire diameter of the firing cylinder, to buckle in any non-axial direction. I realised that in order to prevent this, I was going to need a smaller cylinder to run down through the spring as was the case in my original prototype or else find a spring as wide as my cylinder! In the end, the latter is what I decided to for as once again, there was less cutting to be performed and less room for mistakes.
Prototype 3 and final product
Now that I felt I had worked out all the possible flaws in my plan, I got to work with a much more technical, final construction. As previously mentioned, my dad owns a mechanics garage specialising in trucks, so I filled my pockets with all the nuts bolts and steel I would require and then got to work welding everything together according to my calculations, designs and knowledge of the dimensions. Clearly, I’m not a very accomplished welder so thank god that’s not the basis of marking for this assignment! Enjoy this picture of Doug with my initial launcher, positioned strategically far away so you can’t see the dodgy welding.
Incorporated into the design were two functions to maximise the accuracy of the launcher and allow for adjustments. A threaded bar runs through the launcher and contains a nut inside. By tightening or loosening this nut I can reduce or increase how far the spring can extend and therefore tamper with the amount of energy I’m using to fire the ball somewhat. Tightening the nut will reduce it (spring cant fully extend) and loosening the nut will maximise it (spring can extend fully). It also ensures that I am pulling the spring back to the same maximum each time and can be seen in the second set of pictures further down in this post. The second variant is the angle at which the ball is fired. The firing cylinder is situated on a hinge, allowing me to increase or decrease the angle at which the ball is fired. Therefore, by being able to adjust both of these factors even after the construction of the launcher, I was able to perform a series of calculations and experiments to see what combination of values were most accurate.
In these pictures you can also see the washers and nut against which the ping pong balls sits. These washers needed to be offset against each other to
- hold the spring and 2. equally distribute mass and stop vibrations due to eccentricity which in turn, causes the ball to fire at odd angles left or right of the target straight ahead.
Now it kills me to say that after getting to this point, it came to light that my blind enthusiasm had resulted some dodgy understanding. I could only trigger the firing mechanism, not enable it. By pulling back the threaded bar each time I was adding potential energy to the device, something which was not allowed. However, there was a loophole I could exploit. I could employ multiple launchers and have them all pre-cocked at the beginning so technically, all the energy was present from the outset. I decided that after all the effort I’d put into incorporating the reloading system, I would still fire balls from this cylinder should I have any of the 20 second time frame remaining and could therefore still demonstrate the system’s capability.
This time around, I found two empty StixAll tubes and cut the tops off. I drilled through the end and found a threaded bar to perfectly slot in. Once again I cut my springs and bars to length and calibrated the system using nuts. I then used some very aptly named adhesive to attach the cylinders to either side of my original mechanism. To find out what this adhesive was called, check out the pictures ;P I also cut up some chopsticks to the size I needed and used them in conjunction with the seals off milk cartons/a vinegar bottle as a method to have my cylinders pre-cocked and ready to fire. By positioning the chopsticks directly above, rather than to the left or the right, of the bar, I could achieve a better shot; once I made sure to pull the seals directly upwards also. And so here was the final product!
Here’s a little video of the firing mechanism in action. Obviously I wasn’t holding the the launcher too tightly as I was trying to balance the phone to record as well, but you’ll notice a little kickback as I fire which will be discussed more in my next blog as I complete the course and reflect on the project.
The Principle
Now, all of this was possible thanks to the one simple principle I alluded to earlier; Hooke’s law. In mathematical terms Hooke’s Law states that F = kx where k is a constant which is characteristic of the stiffness of the spring and x is the springs displacement. The ability of the of the spring to return to its original shape post distortion is known as the ‘restoring force’. This is proportional to the amount of displacement or ‘stretch’ the spring is subjected to, by a factor of k. Now that my design was completed, I could use this equation to improve and work on the accuracy of the launcher in order to land maximum balls.
But what was my spring constant? Well a few quick measurements and a calculation and I soon had an approximate answer. P.s. if you’re wondering how I created my oh so lovely typed formulas, the answer is Latex! Soooo much faster than trying to use equations in Word, believe me.
From there I could use my original F=kx formula to suss out what force was compressing the spring.
From here I said that the when this force is released, the same force will be exerted on the ping pong ball as the spring restores its original shape. Therefore, in terms of the resulting projectile, the force calculated above is also equal to mass times acceleration of the ping pong ball.
Now that I knew the acceleration of the ball I wanted to know what the velocity of it would be as it exited the firing cylinder. Of course the firing happens so quickly that it would be impossible for me to ever record this value. Hence, I estimated to happen at about a tenth of a second
I also further analysed the equations of motion and derived relevant equations for my projectile as shown in the figure below.
From here, I could plug my values into the equations and find out whether I could achieve sufficient values for my projectiles maximum height and maximum range!
These values were more then enough, the maximum height would surely get over the 20cm bottles and my range was plentiful also. Of course they are quite theoretical and there were a lot of factors mitigated here such as air resistance, friction between the threaded bar and the bottom hole, friction between the spring and the cylinder, how smoothly the trigger was released etc. But given how big my theoretical values were, I didn’t stress. And good job too, because when I fired my launcher for the very first time, I got all the height and distance I needed. Check it out here for yourself.
Of course, this was my very first video, taken before the other two cylinders or firing pins were added. Once my actual device was built, positioned (by lining up with marks on the floorboards) and fired repeatedly, I was nicely confident I could land balls in relativelyyy the same area; it was looking promising. However, I may have slightly underestimated how small 133cm² actually is when I went on to add my attachments. There’ll be more discussion on that in my coding and completion of the course blog, which is coming up next! So stick around to see how I got on when I finally put everything together.
Cheerio,
Aoife
