CONTACT FRONT!!! A blog on my ping pong launcher
The largest component of assignment 2 is the construction of a launcher device that can fire ping pong balls into a basket on top of the robot. The assignment brief comes with various requirements and optional extras. The main requirements are as follows:
· The device should use an energy storage device that converts potential energy into kinetic energy when released (the recommended design will use pneumatic force from an inflated balloon).
· The forces and energy stored by the device should be low, and not present any risk of harm.
· You should launch the ping-pong balls into the robots basket. Each ball is with 10 points. Only one ball can be fired at a time.
· Max surface area of the basket holding the ping-balls is 133cm² (approx.. perimeter of 340g cereal box) — bonus points for making smaller.
· New balls cannot be loaded manually after the robot leaves its post.
Of course, the standard disclaimers need to be given. This project was done during the COVID pandemic and therefore resources were limited and a large amount of ingenuity was required. At this point I will note that I only managed to acquire one ping pong ball however as you will see some of my designs do contain the mechanism or could easily contain a mechanism to fire multiple balls. In spite of this disadvantage, I am actually quite pleased with my work in this area.
When brainstorming a lot of potential options were available to me. I could develop a catapult, I could use the recommended balloon system, I could develop an air pressure powered system, I could make some kind of gun and even the idea of a spring powered launcher sprung to mind. OK, no more terrible puns, I promise. Ultimately, I settled on three designs, a catapult, a YOP powered launcher and a semi automatic Lego gun powered by elastic bands. These three designs were the easiest for me to build with my given resources and in my opinion the greatest in potential! Okay okay! sorry that was the last one! I mean that was probably so bad you didn’t even get it… potential energy…..potential. OK I swear I’ll stop.
YOP
This design was one of the most complex and is probably one of the more creative or unique designs considered by the class. In Ireland most schools traditionally are same-sex and having attended both all boys primary and all boys secondary school it’s safe to say that a lot of shenanigans went on. I have rather fond memories of lads who would bring YOP drinks to school. This particular bottle has a very interesting characteristic. It’s designed in such a way that it can be easily stepped on when empty with the cap closed but the force of the step fires the bottle cap a great speed. These things would be used to great effect normally aimed at the back of somebody’s head in the hallway or in a lunchroom as a makeshift cannon.
Back then I had no concept of what was actually going on. Well now, with nearly three years of engineering behind me, I think I can take a fairly good stab at a physics-based explanation. When the bottle is empty after consumption, the lid is replaced. This creates a constant mass of air. When placed on a hard flat surface like a table or the floor and we apply a force perpendicular to this flat surface we are now reducing the volume. We know from our thermofluids class that if we fix the mass and reduce the volume, the pressure is going to increase. We also know that force is pressure by area. If we were able to calculate the new pressure when the volume is reduced, will be able to calculate the force exerted on the bottle cap. The net difference in force exerted by the internal and atmospheric pressures wants to open the cap. This force combined with the reduction in structural integrity of the top of the bottle from the reduced volume creates a weak point at the bottle cap which the pressure acts against. When pressure exceeds a critical force the bottle cap is fired forward and we now have a typical mechanics problem which can also be modeled easily enough.
MATHS (I know, boring, just bear with me)
Taking the bottle as a perfect cylinder of radius 2cm and height 12cm
V = 3.14(2)²(12) = 150cm³
If we assume under force it shrinks to 1/3 of its original volume, our new volume is 50cm³
(101KPa)*(150cm³) = (50cm³)(P)
P = 0.3MPa
F = PA
Thus, F = 0.3MPa * (3.14*(15mm))
F = 0.3MPa * 0.071 = 0.0213N
The force of the atmosphere keeping the cap on is 0.007171N
That’s a ratio of 2.9 times the forces pushing the cap off as opposed to keeping it on.
An example of a completely unmodified YOP bottle firing can be seen in this video I pulled off YouTube. I think this video shows quite well the basic principle I’m going for. While this bottle cap isn’t a ping pong ball, this force can be harnessed as a propulsion mechanism for a launcher. For testing and evaluation purposes I built a very simple launcher. It is a one shot ‘fire and forget’ system. In essence it is a bottle with a barrel attached to the end of it and a ping pong ball rolled down the barrel. The ball rests on the bottle cap and when the bottle is compressed, the cap shoots forward. Similar to how you use the white ball to hit other balls in the game of pool, the bottle cap strikes the ping pong ball and fires the ball out the end of the barrel. As you will see in the second video of me firing, it turns out this is actually quite a powerful launcher. In fact it’s too powerful. It will be rather difficult for me to find a way to reduce the critical pressure and force using what I have in my house given that the bottle is a prebuilt structure. Before test firing my prototype as seen in the second video, I did draw up some ideas and schematics as to how a multishot variant of this may run. The main idea that I toyed with was to have a fixed barrel and a gravity fed loading system. When a ball fires, a new on falls into the barrel. To automate the reloading, similar to how an artillery weapon fires, YOP bottles would be used as shells which could be individually placed at the end of the barrel and fired. However this design presented multiple problems, the force required to compress the bottle is extremely great usually resulting in the bottle being placed horizontally and stepped on or punched, it will also be difficult to replace the bottle without the ping pong ball falling through the back of the weapon. I also cast some doubt whether that this will count as a manual reload or not and finally other designs proved to be more accurate, significantly more accurate.
Catapult
The catapult was an obvious choice! Plenty of examples online, they’re relatively accurate, easy to modify, to adjust for distance (i.e. over and undershooting) and are easy to make. The catapult I built was a spring powered catapult using two Lego spring pistons and Lego technic. An arm rotates around the fulcrum and at the end of the arm a holding device is placed. The other side of the fulcrum is a shorter arm attached to the springs. Being the quote unquote backup device, not an awful lot of time or energy was invested into it and it was mostly built using trial and error adjusting the length of the arm the number of springs and so on. No real thought was given to how to fire multiple balls bar the construction of multiple launchers which would technically meet the requirements albeit in a rather crude manner.
Seeing as this was a backup launcher, I never fully finished a mathematical analysis of it instead certain formula laws and principles were researched, collected and employed. In the event I needed to use this launcher, I would have continued with a full mathematical analysis however it was not required. Below are some of the formula useful notes and points of information I used and considered when assembling the prototype which is seen in the video above.
The math of the spring is discussed after the final launcher. Regarding the projectile, the following equations can come in handy in determining necessary characteristics of the catapult
Maths (Yay more boring!)
𝜔 = √ 3𝑘𝑥²/ 𝑚𝑟²
𝑣 = 𝜔r
(angular and liner velocity formulas for spring powered mangnels where x is the distance the spring is pushed back)
Link to source of the velocity equations
The above is a short selection of suitable equations to use for designing your catapult. As stated prior, the spring equations are talked about at the end of the blog and the above equations, while useless, were not used for any analysis as this prototype was ultimately made redundant.
The gun
This is the most complex and most developed design of the three I prototyped. The complexity of this design is less to do with the physics and more to do with the mechanisms and structure of the build. In fact, the only real physics-based elements are the trajectory of the ping pong ball and the force stored in the elastic bands. The complexity of this design is due to a number of parts, a magazine fed system, the semi automatic loading, the ping pong grenade launcher attachment, the small size of the device and the safety mechanism on the device. For this section of the blog it will be split between a written piece and several short video explanations as I feel an audiovisual explanation will be more intuitive and easier to follow than large blocks of text explaining abstract and niche concepts such as the sear and bent on a weapon.
As someone with military service, I understand the mechanisms and principles employed in firearms to make them function. By extension, this can be applied to a firearm style launcher as with the exception of a gas assembly they are almost identical. In the construction of my launcher I will take influence from various firearms from pistols, to the SMLE Lee Enfield and the GPMG FN mag. I also adopt concepts such as attachments like a rifled grenade or in this case, a rifled ping pong ball. These concepts on the mechanical mechanisms were shrunk and simplified to be made with Lego technic.
The first element that I’m going to discuss in a video will be a simple overview of the weapon such as a firing mechanism minus any attachments, it’s basic operating procedure such as how to load and fire the weapon and the safety catch.
In the next 3 short videos I’m going to briefly go through the magazine assembly, how it functions, similarities and differences to real life magazine mechanisms both standard and unique mechanisms in the real world, functional demonstration on how to load the weapon and cycle it to demonstrate the ammo counting capability that the weapon possesses.
Finally, I’m going to demonstrate the attachment and how it works before I resume writing to describe the concept behind the attachment, further ways to develop it and a brief functional demonstration.
Long before the advent of underslung grenade launchers like in the picture below of an M 203, there was a concept floated in the first half of the 20th century called the rifled grenade. This basically was a cup placed on the end of the barrel of the rifle and a (usually special) bullet would be used to launch a grenade from the cup as seen in the pic above.
These concepts helped inspire my design which I will now explain in the following two videos
Handling characteristics
Having built the launcher, I can note a few characteristics. Reloading the Lego ammunition can sometimes cause the elastic platform auto magazine to move in such a manner that it is no longer stuck to the elastic band. This can be a pain to fix but it doesn’t take more than 10 seconds to fix however ideally further engineering will be used. But alas, it is what it is. Regarding the trigger mechanism, it would work on occasion but it jammed so frequently that it was misfiring more often than firing requiring an IA (Immediate Action) drill. For practical purposes the weapons bolt and firing pin are the same thing and this extents slightly out of the front of the barrel. This means when the Lego ammunition has been fully used, ping pongs can still be fired regardless of the ammo count of the magazine.
Negative characteristics
Other negative characteristics of the weapon include insufficiently small magazine / pistol grip size making holding the weapon difficult and the trigger mechanism is too large again interfering with the ergonomics of the weapon. When considering the problem with the sear / trigger mechanism, I would recommend for further development that I either rework this mechanism or remove it entirely or replace it with a new system perhaps either a smaller version, a safety catch style switched like a modern weapon or a proper trigger instead. Another point to note is that the weapon requires to be cocked each time which counts as manually loading the shot each time. One of the downsides of basing the launcher off modern weapons is that they almost all exclusively use a gas operated system. Very simply put, on most weapons when a round is loaded into the breach, the bolt pushes a round in and creates a seal. When the trigger is pulled, the firing pen in the bolt strikes the back of the round causing the inside the round to ignite. The hot gases push the bullet out of its casing and down the barrel in the only direction it and the gas can travel. In semi-automatic and automatic systems, there is usually a second cylinder below the barrel. Near the end of the barrel there is a kind of valve which allows a lot of the hot gas to escape into the system below. The gas travels towards the firer and at the end of this system is a piston. The gas causes this piston to move back towards the firer and pulls the bolt above it back away from the end of the barrel allowing the empty shell casing to be ejected and a new round to be loaded.
Clearly, I have no gas and so this system will not work. There is also little to no development for alternate methods seeing as this method is so efficient, reliable and it’s the standard across all major manufacturers regardless of location, national standards, military, police or civil requirements and time period. I do intend over the summer to develop this a little further as it was quite a fun little project. I will definitely be revisiting the safety trigger mechanism and extending the magazine. Making the launcher load automatically will be a lot more difficult especially given the limited space. It is not impossible perhaps using one of the Springs from the catapult that I can now use since it is no longer required. Perhaps the spring could push the bolt back and allow another round entered the chamber however this will be difficult to control with the trigger mechanism. Another thing which I would like to develop further is the ping pong attachment, specifically some kind of vertical gravity fed magazine which allows extra balls to be dropped into the launcher when the one currently in the barrel is fired. Again, due to having no extra ping pong balls due to COVID and them not arriving on time, this wasn’t a major concern and this is on top of the fact that I had difficulty getting the balls into the target in the first place.
Probable cause of inaccuracy
I personally do not chalk this down to accuracy or reliability issues. In fact I believe the weapons are quite accurate. The main problems are that indirect fire weapons are a lot more difficult to aim than direct fire weapons and the fact that the robot would take about 45 seconds to reach the endpoint and would only stay there for 20 seconds making it extremely time consuming. It will be nice to be able to fire multiple balls in a quote unquote ‘spray and pray’ tactic but again it was not feasible as I hadn’t the balls. I also had a lot of trouble getting the ball to stay in the box when it landed as ping pong balls have a tendency to bounce as a result they will tend to bounce out of the box after they hit it as mentioned in my blog on adjustments to the robot. I had to pad the box in such a way to reduce this impact. Accuracy issues can also be attributed to the fact that each time I tried firing the robot and the launcher will be in a slightly different positions with the bolt pulled back to a slightly different distance from the front of the barrel each time and the physical location of myself and the gun will be slightly different each time. This makes aiming corrections extremely difficult and therefore puts it more down to luck rather than skill. While one could get skillful enough to overcome this, it would take significantly more time than I have and considering how late the robots arrived, the time wasn’t really there.
Conclusion
Overall, I’m delighted with the development cycle that I took for these launchers. During the brainstorm process, I came up with at least half a dozen feasible, buildable designs despite the COVID lockdown. I then took three relatively promising designs I managed to build 3 functioning prototypes, two of which were reasonably accurate and reliable bar the YOP launcher which while working was way too overpowered and will be too complicated to rectify. The catapult was simple to make, consistent and reliable and the gun could consistently fire its Lego ammunition and the ping pong ball from the launcher with relative accuracy in a smooth, easily repeated and fast operating cycles which would in theory allow multiple shots to be landed in the robots basket within 20 seconds. I believe these designs show a nice level of engineering with regards to things like principles, some basic formula and calculations, prototyping and design of things like the physical structure and mechanisms. I’ve already outlined changes and adaptations I could make to the devices but apart from this I don’t see much more I could have improved upon without overengineering. As a result, I would say that this has been very successful part of the assignment and I look forward to further developing my Lego gun over the summer. All that’s left to do now for this blog is a short maths section on the gun launcher.
‘Gun’ maths (last bit I swear!)
When conducting my research, I came across this website
https://www.wired.com/story/how-much-energy-can-you-store-in-a-rubber-band/
On this website, they quite a good bit of information about springs, how they work and their equations. The following are the important points that I extracted from it that were relevant to my launcher:
Rubber bands actually behave very close to ideal springs. Both springs and bands require more force to stretch them the further you pull. This leads to a proportionality between the force and the stretch distance which is represented by the spring constant (k in N/m)
The energy stored in a spring depends on the distance it is stretched and the constant as seen in the following equation:
The website then goes on to give some simple data and shows that it is a roughly linear relationship as evidenced by this graph:
Regarding this graph the website goes on to say,
“The slope of this line would be the spring constant with a value of 90.8 N/m. This means that if I stretch the rubber band a distance of about 20 centimeters (that’s around the breaking point) then there would be 1.8 Joules of energy stored in it.”
From my specific weapon, I actually tried multiple elastic bands to vary things like the power of the weapon and so we can use this equation to determine the exact range that we want. For example if we were to maximize range, we would preferably want a longer elastic band as there is no ideal ratio of length to force for instance.
I did determine however that the bolt had to be pulled back far enough to load a round which was also a distance that produced a range that was too great for the maze. Hence I had to ease the rubber band a little by letting the bolt go partly forward before firing which was done by eyeball estimation.
