Prototype for Advanced Bottle RocketFollow article
The Endeavour rocketry society of the University of Edinburgh is working on developing a water bottle rocket for an Outreach project. This is happening in collaboration with RS. The following article will summarise the build of a prototype for the water bottle rocket. There are still improvements to be made, but we wanted to share the process and maybe the community on DesignSpark even has ideas in the comments for further improvements.
The building of the prototype began with planning and compiling ideas from different sources to create our own version of an advanced water bottle rocket with a payload. There is already a lot of content on water bottle rockets out there, so it was important to us to find a unique take on the concept. We tried to combine ideas of sustainability and data science in the rocket. The aim was to include these aspects by making the rocket both recoverable and using it to measure data.
In the following, this article will walk through the development and building of the prototype. A project with full instructions will follow. The aim is to create a toolkit that makes this type of water bottle rocket build that anyone can recreate anywhere. This article is going to be useful for anyone attempting to create their own water bottle rocket, at the end I will compile what has worked and what can be improved upon as a quick overview.
1. The Basic Outline of the rocket
The rocket is composed of two bottles, one being the pressure chamber and one the payload chamber. It starts from a wooden launch pad. The launch itself occurs by pumping air into the sealed, water-filled pressure chamber with a bike pump. The cork sealing the pressure chamber will fly out and the water pressure will propel the rocket forward.
We attempted to make the rocket recoverable by including an electronic parachute release mechanism. And we attempted to include data-collecting sensors. Both of these areas can still be improved upon, but the basics have been working well so far.
2. Building the rocket stand
Due to the materials we already had on hand we decided to start the build with the construction of the launch pad. As a material, we decided on wood and initially glued the wood, but I can already say that that was a mistake. It is worth investing in some screws and some L-brackets to make sure the legs of the stand stay at a 90-degree angle and do not cave in under the pressure of the starting rocket. We aimed to keep construction simple and produced the stand out of a single plank of wood. This also ensured the legs were long enough to fit the head of the bottle and the head of the pump underneath the launch pad surface. We were thinking about adding stabilizers to the top of the launch platform, but this ended up being unnecessary as the fins on the rocket acted as stabilizers. This completed launch pad construction.
3. Constructing the pressure chamber
Next, we decided to construct the pressure chamber of the rocket to get an idea of scale for the rocket and test some of our theories about the rocket stand. The bottles used were meant to be straight-edged, with a large volume and made from fairly solid plastic. The final choice for us was Irn Bru’s two-litre bottles.
For the pressure chamber, we decided to keep the bottle as intact as possible, without any cuts or holes in it, since this could compromise the stability of the design. Therefore we just left the bottle whole and decided to only glue and tape things onto it and make incisions into the other parts instead. The fins for example, which we already applied in this step, were taped and glued to the outside. This was done by creating an area that would fold away at a right angle from the fin and therefore be parallel to the bottle. This is best seen in pictures. The technique did end up working out for us, especially with sufficient taping the fins held up well against the stress of starting and a couple of crash landings.
Then we chose a cork to fit into the bottle opening. Here we are not sure if we have optimised this part of the rocket yet since the cork did withstand a certain amount of pressure. Possibly more pressure and therefore a higher height could be achieved with a more well-fitting cork. Our strategy was to shave down a sparkling wine cork since those did not have holes in them from corkscrews. The cork also had to be adjusted in length so that the needle of the pump could fit through and penetrate. The cork was initially too long so that the needle of the pump did not reach the inside of the rocket.
This concluded the assembly of the pressure chamber. It fit well onto the launch pad. We then took the circumference of the lid of the bottle and chose the next biggest circular drill to drill a hole in the launch platform. The fins were taped at a height so that they could act as stabilizers. This meant that a bit of the bottle poked through the launch platform. Therefore the fins need to be taped on a bit higher than initially thought. Overall this step was fairly straightforward with the right equipment.
4. Assembling payload
This was possibly the most challenging part of the build. Without this component, the bottle rocket would still work completely fine, but this was the extra challenging part we wanted to build. And of course, here the support from RS really shows. So the goal was to create a parachute release mechanism and then a payload with environmental sensors. The parachute release mechanism is modelled after information on these two websites: Water Rocket Construction - Side Deployment (aircommandrockets.com) and Arduino-controlled Water Rockets: 19 Steps (with Pictures) - Instructables. The environmental payload is our own design, but with information from the documentation of both the sensors. (Adafruit Accelerometer MMA 8451 and Parallax Altimeter MS5607)
The parachute release mechanism was mainly debugged by putting components in smaller circuits and running test code on them before assembling everything into the bigger circuit. One key thing was to make the ground as straightforward as possible and then to ensure that all switches were connected in the right orientation. Since I had very little experience with circuit boards, the whole payload was a big challenge for me. For anyone wanting to attempt a challenge like this: I do think that this is a wonderful project to get out of your comfort zone. Of course, it is a bit more challenging than usual beginner projects, but it is incredibly rewarding to try and figure this concept out yourself with some help online. In the end, we were able to create a parachute launch mechanism, that has two modes, one is triggered by a countdown, the other by a tilt switch. This tilt switch mode should be the most useful for triggering the is parachute launch as it should be triggered at apogee when the rocket slightly tips over. When the launch mechanism is triggered it moves a servo motor into a different position. This releases rubber bands that are holding the parachute in place. The different modes can be accessed through a user interface consisting of multiple switches and LED’s.
The other part of the rocket payload is the environmental aspect. We decided on several sensors. Here we tried to make them as useful as possible for a small rocket. We decided on a Temperature, Pressure and Altitude sensor. This came in one sensor from Parallax. This was of course very practical as it took up little space. And then we also decided on an acceleration sensor that could provide some interesting information in the short time span that the rocket will be flying. This sensor is made by Adafruit.
In the assembly of the payload, there were some small hiccups. We hope to resolve them in the further development of this water bottle rocket. The Arduino we used for the flight computer could make the sensors work with the parachute launch mechanism but not with an SD card as well. There were not sufficient pins on the Arduino Nano. This of course complicates things, because space inside the water bottle rocket is incredibly limited. This became a bit of a struggle later in assembly as well. So to anyone that is interested in making a payload with a water bottle rocket should very much consider this. Another issue is that there seems to be an inherent bug in connecting the sensors to the SD card. Both print to the terminal exceptionally well and the code works well. And it is possible to just write to the SD card while they are all connected in the same circuit. Then, as soon as the code tries accessing both sensor data and writing to the SD card the system stops working well. Either the SD card can’t be accessed or the sensors start printing strange values (mainly everything equal 1). We have not yet managed to understand the bug between the SD card and the sensors. This is a bit of a strange bug but should be resolved soon in code. This is definitely the biggest area to still work on.
Then the soldering started. Once again there was very little experience on our part here. It was attempted to keep the soldered circuit as small as possible. Another key thing was that the tilt sensor was on his separate little piece of the circuit board so it could be oriented in any orientation to be triggered at the appropriate time. When soldering the usual tips apply: Do not connect things that shouldn’t be by using too much solder and please ensure all components are oriented the right way, especially LED’s, we made some unfortunate mistakes in this regard. In the end, the soldered circuit is definitely much smaller than the breadboard circuit, which is quite helpful for fitting it into the rocket. Nonetheless, if we could have made it any smaller this would have been very useful. We soldered with jumper wires, due to a limited time frame and lack of access to better resources. This made the “back” of the circuit board quite bulky and should definitely be reduced and optimised.
6. Making the payload chamber
The payload chamber needs to fit on top of the pressure chamber, therefore it is going to be made out of the same type of bottle. The top and bottom will be removed from the bottle so that it is basically just a cylinder. Then the bottle can be pushed onto the pressure chamber. This creates a space for the payload. The inner structure of the chamber can be created with cardboard. This is necessary to create the needed hold for the electrical components. Of course, the heaviest components are the battery and servo motor. These should be inserted in holes in the cardboard and then additionally secured with tape.
Holes were cut into the outside of the bottle to create a parachute release chamber opening and access to the user interface of the parachute release mechanism. Here placement was very difficult. Mainly because the cables were too long. It was however very useful to keep all buttons, LED’s and switches on one side and the cables on the other. This set-up could however certainly be optimised. Especially since we did not incorporate the sensors at this point. These will also take up a lot of space and therefore space-saving is a very big priority.
7. Parachute Release Mechanism
To release the parachute we tried to make a spring-loaded type of release mechanism. This means there is a piece of plastic indented under pressure, that will release when the servo motor releases the pressure. This makes the parachute pop out.
The parachute has to be wrapped in a very particular way so that it unrolls and unfolds very easily. Here it is best to look up a Tutorial on the internet. There are multiple great resources out there, like Parachutes (aircommandrockets.com).
This mechanism seems to work well, except for the fact that the “door” that keeps the parachute in sometimes does not open as easily or wide enough. This is because we only bent the door open, which means that the door was only folded plastic, this is of course not particularly elastic. This slowed the parachute release, so this might be an area to review in the future. Apart from that the spring-loaded type of mechanism is very easy to construct and quite successful.
8. Assembling nose cone
We used a top third of a bottle and cut out the lid. Then we inserted a half-sphere made from Styrofoam into the hole and tried to streamline the edge as much as possible. We pretty much only used glue to secure this sphere. Then this nose cone was basically used as a lid to put on top of the payload chamber since this only had a flat cardboard lid otherwise. Here the parachute was secured as well. Then the nose cone was secured with tape.
This concluded the build of the rocket. We made some flying tests and I will insert a video of our most successful trial. During these tests, we noticed that one of the main problems is that the rocket does not fly high enough for the parachute to unfold and slow the descent. This is another area that should be improved in the future to optimise this project.
Much more detail will follow in a designated project here on DesignSpark, where we will also share code and schematics. This prototype taught us all a lot and showed us that we had to develop many new skills to complete this project. Yet there are also still some areas to be improved upon. I will show them in the following list. Any tips are appreciated in the comments. Hopefully, these issues will be resolved by the time the full project is published.
In the final testing the issues that we have found are:
- The rocket does not fly high enough to give time for the parachute to unfold.
- Possible improvements:
- Fly higher (better cork fit, larger pressure chamber, more pressure)
- Make lighter
- Make parachute unfold quicker
- Use another type of propulsion
- The SD card does not yet work with the sensors.
- Work on code
- Use different sensors/ SD card
- Payload to large
- Use smaller wires so they take up less space.
- Possible improvements:
Things that already worked well:
- Stability of test stand after adding screws.
- Take-off of rocket (Mechanism works, but should be made more efficient).
- Assembly procedure makes sense and is easy except for soldering.
- The parachute mechanism works very well.
- We had fun building this prototype!