Introduction: L.A.R.S. (Launch and Recovery System)

About: While I'm not busy doing my job as a Solutions Architect...I'm a huge dork. I sketch crazy ideas. I design those ideas. I build those designs. If I have to start over, it was a bad bad ideas.


This project is a Launch And Recovery System (LARS) made up of various models and assemblies. All together, they represent a recovery system suitable for a low-altitude water rocket. The rocket is made up of several sections, fabricated from 1.5 liter SmartWater bottles.

The entire system is comprised of several elements:

  • Launch Pad
  • Main Body
  • Recovery System

Purpose and Drivers

The inspiration for this project (as most of my projects) originated with my nephews. Long story short, years ago (when my young nephews weren't so big) they wanted to set off some fireworks on Independence Day. Normally, not so bad, but that year was different: we were planning a long weekend at their grandparent's cabin, in McCall, Idaho. If you've never been to Idaho, it's very dry. If you've never been to McCall, Idaho: it's very dry and there is a LOT of trees, shrubs, grass and other fuels perfect for forest fires. As it was already an incredibly dry year and Smokey the Bear was holding the "HIGH" fire risk sign outside the forest service office, I sought an alternative.

At the same time, I saw it as an opportunity to demonstrate what I try to impress upon young, scientific minds: STAND OUT. Think outside the box and come up with the solution to multiple problems. It may stand out like a sore thumb at first, but the best ideas usually do.

Also, as it turns out, I'm SUPER cheap. Not so much for saving money, but I just see so much that can be done with ordinary things. Most of the time, ordinary things that are meant to be single use.


I would be remiss if I didn't mention this first; I am constantly reminding my nephews "Safety, first." This fits into the overall goal since the primary propulsion is water and air. Obviously, this doesn't pose any serious fire danger.

As I built more sections of the propulsion system, this added volume for more highly-compressed air (i.e. more propellant). Of course, this is directly proportional to the max altitude attainable. Continuing with this logic a bit further: yes, this meant A MUCH MORE DANGEROUS velocity when returning to earth.

The realization of how dangerous this could be became apparent after we had our first successful launch of an early prototype, using multiple sections in the fuselage. Take a look at Rocket Boys , on YouTube.


Minimizing the cost of any build is important. In my scenario, I thought it was equally important to ensure the cost of each use was minimal. I mean, c'mon - who wants to put in a ton of work for a one-time use system?

Anywhere possible, I used junk: water bottles that would go to the trash, an old flight case from the army surplus store, a broken chair umbrella from a local sporting goods store, a burst air hose from Harbor Freight and even a broken, pop-up sprinkler head - basically, I like to teach the kids there's no such thing as trash; it's a product that needs repurposing.

Just ASK

In many cases, things you want to use are not necessarily "for sale" and, if they are, may not be worth their price. The original, green umbrella I used for the parachute was on a broken $28 item at a store. I took it to the counter and said I'd give them the $4 I had on me. I showed them it was broken anyway and explained I only wanted the umbrella material. Voila! we instantly had a parachute.


It's nearly impossible to talk about keeping material cost down without also considering how the product design lends itself to reusability. If we could not easily use the rocket over and over again, we may as well be setting the forest on fire with bottle rockets and M-80s.


After only 1 test launch of our first prototype, I realized the entire rocket system needed to be as compact as possible. At the same time, I wanted to loan it out to anyone who wanted to try it. I didn't want to write a bunch of detailed assembly instructions or rent a moving van for transport.

In the end, I used the box where I stored all the parts as the launch pad. A few modifications made it possible to attach air compressor fittings, hose and no-return valve while leaving plenty of room inside for all sections of the rocket propulsion/recovery systems.

...of course, after the most recent launch, I realized the entire system could stand to be more portable, ha.


Launch Pad

  • 1 x Old army surplus shipping container
    • I believe we picked this up at The Reuseum - a great place to check out if you're in Boise, Idaho ;)
    • Spoiler Alert: they don't actually stock these types of army containers since they changed locations.
  • 1 x Air Compressor hose (~ 15' ...or whatever distance you feel safe)
  • 1 x Air Compressor hose (~ 2' ...this just goes inside the box)
  • 1 x No Return Valve - Here's one
    • Lots of options out there. Get something cheap.
  • 1 x 90º elbow -- goes from ¼" compressor thread to garden hose thread (something like this )
    • I forget where I got this, but given this is a super smart (and super-forgiving) audience, I'm guessing you can figure out how to go from one thread to another after a few minutes in D&B or TSC.
    • Maybe combine with something like this .
  • 1 x Threaded, rotating air compressor male adapter (we reused something like this one , from Harbor Freight)
    • This is used to attach the short piece of hose on the inside of the surplus container
  • 1 x Gardena Female Adapter


  • 12 x Smart Water Bottles ( 1.5 Liter )
  • 4 x Essentia Water Bottles ( 1.5 Liter )
  • 1 x Custom Nozzle (i.e. 3D Printed 2 Liter Bottle Thread on a Male Gardena hose connector)
  • 1 x Tube of Sikaflex adhesive (like this Construction Sealant )
  • 1 x Role of Duct Tape (or strapping tape)
    • I used a combination of white and black duct tape on the fuselage sections. I thought it gave the rocket more of an Apollo-era look, ha.

Recovery System: non-printed

Recovery System: Circuit

  • 1 x Arduino Pro Mini
  • 1 x 24 Pin DIP Socket Adaptor ( link )
  • 1 x 5V Voltage Regulator ( link )
  • 1 x Piezo buzzer ( link )
  • 1 x 220Ω resistor ( link )
  • 1 x MPL3115A2 module
    • The one used here is no longer available
    • I started working on an alternative that logs data and leverages a gyro along with the barometer - the code on Github (see lars ) should still be applicable if you have an alternative
  • 1 x Push Button Latching Switch
    • If you haven't noticed, I'm a fan of Tayda for small things in small quantities for a small price

Step 1: Launch Pad

The launch pad is a multifunction piece of equipment. Made from an old, equipment box I found at Boise's infamous Reuseum , it serves three functions:

  1. Fueling mechanism (...for lack of a better term)
  2. Storage for the rocket body pieces
  3. Launch pad


Before talking about fueling the rocket, it's important to take a minute to think about the basic scientific principles supporting it. From Wikipedia:

A water rocket is a type of model rocket using water as its reaction mass. The water is forced out by a pressurized gas, typically compressed air. Like all rocket engines, it operates on the principle of Newton's third law of motion.

Breaking it down a bit, the reaction mass is simply something used to push against. It pushes against the ground and the air pushes against the water. The air also pushes against the bottle. Action is the expanding air, reaction is an object with the least mass (i.e. the bottle rocket) is forced away from the launch pad.

Pressurizing the rocket needs to be an additive process. It requires pushing air into the vessel without allowing air or water to escape. This is accomplished using a little device called a "non-return valve" (sometimes referred to as a check valve).

The rocket is attached via a Gardena Hose connector on the launch pad. The bottom of the rocket has a nozzle with a contour matching the Gardena Tap Connector. I modeled a nozzle in Fusion 360 combining the profile of the Gardena Tap Connector with the threads of a 1.5 L bottle.

The hose connecting it all together was taken from a burst air hose. I found it at a Harbor Freight parking lot sale for a couple bucks - looked like someone returned it because it burst open. When I saw it, I knew it needed spliced. I thought this would be a good time to cut a short piece to permanently attach to the launch pad.

Future Improvements

I'd like to either buy or 3D print a few additional parts. I would like the connection points on the box to sit flush with the rest of the box's surface. With those little garden hose threads and air compressor connection sticking out, it is difficult to store without getting damaged. Also, it's prone to damage when transporting to a launch site.


When disassembled, the sections of the rocket body fit nicely in the box. Also, I had enough room left over to keep the recovery assembly in there as well. I keep those parts in a separate box in case the other pieces moved around and smashed the 3D printed parts.

Launch Pad

When it's time to launch the rocket, we need something to help prop up the rocket. Also, it's imperative that it starts off in the right direction (i.e. Newton's First Law of Motion). To achieve this, I used two pieces of aluminum trim channel connected by two 3D printed pieces.

The bottom 3D printed bracket has space for a ¼" hex nut. I used this size since it's a standard for mounting attachments to a camera tripod (I'll talk about that in a second). Once I sanded down the sides of the nylon hex nut, it fit snugly into the bracket.

My brother is a great outdoor photographer and, as with all experts in their field, breaks or upgrades their equipment. While at his house, I noticed a tripod in the trash. The only thing wrong with it was the vertical height adjustment. Other than that, it was a good tripod. I didn't know what I would use it for at the time, but it had a lot of great parts.

When I started building the rocket launch pad, I reached a point where I needed something portable to hold up the guide rails. BOO YA - repurpose the broken tripod. It adjusts nicely in several directions, great for uneven ground at your launch site. Also packs up and fits well in the launch box.

I also designed and 3D printed a piece that fits between the rails and attaches to the rocket main body. This particular piece holds the rocket close to the rails and doesn't allow it to tip over. I attached this to the main body with some 3M 414 Scotch® Extreme Mounting Tape. When I designed the piece I recessed two spots where the foam tape goes, so the piece sits flush with the curved, plastic surface.

Future Improvements

I'd like to 3D print some connectors that make it possible to cut the guide rails into smaller segments. With smaller segments, I can store the launch rails in the box as well. Trying to transport the rails in 8ft lengths - in a Jeep, no less - was a pain. Also, having it all put together made the plastic pieces prone to snapping (which they did) in the car.

Step 2: Main Body

The main structure of the rocket is comprised of multiple, identical components. The components are made up of two 1.5 L SmartWater bottles and one 1.5 L Essentia water bottle.

  1. Cut the bottoms out of the 1.5 L SmartWater Bottles
    • Be sure to leave a bit of the curved part. This is more surface area for the Siaflex adhesion.
  2. Cut the top and bottom off a 1.5 L Essentia bottle
  3. Insert the bottom of one of the cut SmartWater bottles inside the cut Essentia bottle
    • Try to align the bottom of the SmartWater bottle with the midpoint of the Essentia bottle
  4. Insert the bottom of the other SmartWater bottle in the other end of the Essentia bottle
    • Push the SmartWater bottle down until it touches the other SmartWater bottle

Mark the Bottles

Mark the bottles to accurately align them if you have to. (SPOILER ALERT: you will have to reassemble it later).

Use a method that works best for you. I like to draw a few, obvious marks across the two, different bottles. When the center is filled with Sikaflex and you can't see where the two meet in the middle, it's a more useful guide.

Gluing the Bottles

I like to coat the inside of the Essentia bottle with a layer of Sikaflex. It's a great adhesive and also allows for some expansion. This is a beneficial feature since the bottles tend to expand when filling with compressed air. Also beneficial when it crashes to the ground (...yes, you will likely have a crash at some point) - the bottles are easier to bend back into shape and reuse.

Connecting the Sections

Once you have all the sections glued together, connect them with something called a "Tornado Tube". THIS IS THE BIGGEST POINT OF FAILURE.

The plastic of the bottles and the plastic of the tornado tubes is quite rigid. They don't always fit together perfectly and a lot of air can leak out when filled at a high PSI. Also, when using garden hose gaskets at the contact points, there is a risk of over-tightening to a point where the gasket is forced inside the bottle. When this happens, it basically renders the gasket useless since it's no longer sealing the connection between the bottle and the tornado tube.

I plan on creating my own connections with a dual-extrusion 3D print. I think there could be a easy way to print a rigid exterior (for the threads) and a flexible seal at the middle (to replace the hose gaskets). I'll post those plans when complete.

Step 3: Recovery System: Circuit

Parachute logic is a lot like other scenarios in life: deploy too soon and bad things can happen; not deploying at all, bad things happen.

I wanted to make sure the parachute doesn't deploy until the rocket started falling. Given the components I had at my disposal, I chose to use a barometric air pressure sensor that provides an accurate measurement of altitude.

The entire system needs protection from the elements. I designed the payload to accommodate the circuitry and sensors. I didn't want to take the whole thing apart every time I wanted to activate or reset the system, so I designed the payload with an external switch.

When the system is enabled, an initial measurement is taken - this is our "ground level". As the altitude of the rocket increases, it's new altitude is saved and compared to the next measurement taken. When the saved value is higher than the newly measured altitude, the rocket is assumed to be falling.

When working with the recovery system on the ground, it's possible the parachute will deploy accidentally. The code actually doesn't consider the rocket to be "flying" until the measured altitude is at least 1 meter above the initial ground level measurement taken when the system was turned on.

Once the rocket is considered to be falling, the parachute gets deployed. This is done by activating the attached servo far enough to unlatch the pop-up sprinkler head attached to the 3D printed parts. Of course, the spring in the sprinkler pushed out the umbrella parachute, it falls to the ground, everyone laughs their butz off and so on and so forth.

The circuit was comprised of three, main parts:

  1. Arduino
  2. Servo
  3. Barometric pressure sensor


I originally created a custom board with a bare-bones Arduino on it. When I tried to revive it for this article, it decided to stop working :\

I used an Arduino Pro Mini, but it's a bit of overkill. It's also a lot larger than the previous version. The larger size required a redesign of some 3D printed parts - I'm sure there are some differences in the parts from the photos I've posted (...sorry for the inconsistency).

I posted the code in a public repository on Github. Checkout LARS.


The latch is actuated by a common SG90 servo. The servo gets its power directly from the voltage regulator, not through the Arduino.

Barometric Pressure Sensor

The particular breakout I used in this project was something I found on Tindie (...but has since been retired). It uses the MPL3115A2 sensor. This provides the Arduino with an accurate readout of the current altitude.

Step 4: Recovery System: Enclosure

The recovery system includes several, simple products you probably have laying around. For example, the parachute is made from an old, broken umbrella and it's deployed with a compressed spring from a pop-up sprinkler. Don't sweat the small stuff either - I used a paper clip to connect the servo horn to the sprinkler head latch. Even some of the raw material you can find from random places, like the aluminum extrusion I found in a Goodwill bin.

In another design, I used some fiberglass fence posts in place of the aluminum shown in the pictures. The fiberglass was laying around from some backcountry trip (I think) someone took, they're used to create an impromptu electric fence for horses. Not super-important to this design, but want you to think about alternatives.

Design Influences and Changes

I knew I would one day share this design with friends and family (, never thought it'd be on Instructables, ha). I also assumed not everyone would have the same brand of seltzer water available in their local store. While there is room for much improvement, I modified my design to allow for various sized bottles to fit on the top.

The best way I could think to ensure a secure, yet removable fit, was to use a flexible material. Enter: honorable MakerDojo is strong with da ninja on my side.

With a dual-extrusion print, I could create a piece with a rigid bottom and a flexible top. The flexible part was stretchy enough to squeeze inside the bottle and strong enough to apply the pressure needed to keep the bottle in place.

Step 5: Recovery System: Parachute

This has to be one of my favorite parts of the design. I mean, c'mon, how many times have you thought of floating around all Mary Poppins-style with an umbrella? It was fun to finally see an umbrella actually function like a parachute.

I found one umbrella on a broken product at DICK'S Sporting Goods - I offered them a few bucks and they took it. I found another one while diving through bins at Goodwill. Of course, I found an awesome, old golf umbrella at Second Use (in Seattle) after those other two. Golf umbrella would be awesomely huge and make a great parachute.

Whichever umbrella you choose, make sure it's attached securely to your rocket. When the parachute deploys, depending on the size/weight of your rocket, the force of the parachute opening could be significant. In my case, I attached a piece of flexible bungee cord I had (...I think my sister was throwing it away from some broken trunk cargo net thing). With that bungee cord in place, it reduces the amount of stress on the plastic parts when the parachute is deployed.

Step 6: Overview of 3D Printed Parts

Jump to...

  1. Full Assembly
  2. Recovery System Parachute Plunger Thingy
  3. Recovery System Compression Guide
  4. Recovery System Payload Threaded Top
  5. Recovery System Payload
  6. Water Rocket to Gardena Adapter

The Full Assembly

This model basically contains all the other models, but assembled as it would be in production. It includes the non-3D-printed reusable parts as well (just for reference).

Recovery System Parachute Plunger Thingy

Back to top ↑

Recovery System Compression Guide

Back to top ↑

Recovery System Payload Threaded Top

Back to top ↑

Recovery System Payload

Back to top ↑

Water Rocket to Gardena Adapter

Back to top ↑

Step 7: Wrap Up

There's a lot here I didn't mention. I suppose the more I wrote, the more I realized I forgot to document along the way. This article ended up more like a cross between a comprehensive "step-by-step" but calling out all the variables so you can try it on your own. For us, we had so many different designs and methods that didn't work out but presented new learning opportunities for both me and my nephews. It's a great, little adventure to set out on with your students and any other budding, young scientist.

All in all, it's a fantastic exercise in reusability, creativity and technology - a great one for kids and even more fun for parents.

Be bold, think different and as always let your ideas stand out.

Reuse Contest

Runner Up in the
Reuse Contest