Water Rocket Parachute Deployment Mechanism




Introduction: Water Rocket Parachute Deployment Mechanism

There are different ways to eject parachutes from water rockets. The technique we like to use is called Side Deployment (aka Horizontal Deployment) where the parachute is deployed in a perpendicular direction to the rocket body axis. The other common technique is to separate the nosecone and have the parachute come out in-line with the rocket axis. This is how most amateur pyro rockets do it.

We use the side deployment technique because we often have a camera and altimeter mounted in the nosecone. Having them rigidly attached to the rocket body gives them a more stable platform on the way down.

There are a lot of different ways to achieve side deployment as well. We like to construct ours so that the outer aero-shell is separate from the actual mechanism itself. This makes construction and adjustment much easier. It also allows the mechanism to be easily reused if the aero-shell becomes damaged.

The procedure below outlines the general steps of how we construct ours. No specific dimensions are given here as these will depend on the bottles you use and how much space you want to allocate for the parachute.


- Corriflute (Coroplast, Twinplast, Correx ) corrugated plastic sheet
- 2 PET bottles
- Cardboard
- 2 rubber bands
- Paper clip
- String
- Wire
- Skewer stick
- Ping pong ball
- Timer (Tomy timer, flight computer etc.)


- Tape
- Contact glue
- Scissors
- Craft knife
- Long nose pliers

More water rocket instructions such as these can be found at our main website:

Step 1:

Get a clean bottle with a nice aerodynamic shape and straight sides. Remove the label and clean off the glue with mineral turpentine.

Step 2:

Cut off the neck and the base off the bottle.

Step 3:

Glue half a ping pong ball into the hole left by the neck of the bottle. This gives the nosecone a nice rounded shape. If you are using different sized bottles look through the kids toy box because there are bound to be plastic balls of varying diameters. Don't let the kids see you though.

Step 4:

Cut two circles out of some Corriflute sheet. You can find this stuff almost everywhere. Old signs make a good source.

WARNING: In the interest of public courtesy when obtaining these signs, make sure you don't get the ones with "Wet Paint" written on them.

Make sure the circles are a somewhat loose fit in the bottle. This will allow the mechanism to be removed from the bottle for servicing.

Cut out a larger rectangular section from the Corriflute sheet. The size of this will vary depending on your bottle size and the height you want to make it. Make it taller to fit bigger parachutes. The corrugations should be oriented vertically. Now slice only one side of the rectangle half way along.
Bend the rectangle to make a 'V'.

Step 5:

Next, cut 4 strips of cardboard and bend them 90 degrees along their lengths to make four 'L' shaped brackets.

Step 6:

Glue these to the edges of the V with contact glue as shown in the photograph.

Step 7:

Now, make all the necessary holes in the V to support your release mechanism. You can use a Tomy timer, or as we are using here an RC servo motor connected to one of ourflight computers.

Make holes and slots along the vertical edges of the V. These will hold the rubber bands that eject the parachute.

Step 8:

Cut out another smaller rectangle (the ejection plate) from the Corriflute sheet. This will be used to eject the parachute. Make sure the corrugations run horizontally. Cut on one side only of the ejection plate to make 4 cuts along the corrugations.

Step 9:

This lets you to place the rubber bands inside the corrugations. Use a piece of tape to close the cuts again.
NOTE: When attaching the rubber bands to the V make sure that the cuts in this small rectangle are facing away from the parachute. This ensures that the rubber bands can't come out through the slots.

Step 10:

Securely attach all your release mechanism components to the V.

Step 11:

Glue the V to the two previously cut out circles.

Step 12:

You want to make sure that no pieces are overhanging the circles because you will need to be able to slide the entire assembly into the nosecone aero-shell. This allows you to remove the entire mechanism for servicing.

Step 13:

Keep the inside of the V clear of all major protrusions. You can mount things like batteries in the far end of the V.

Step 14:

Slide the mechanism into the aero-shell. You can push it up as far as it can go. This will stop it from moving upwards when the rocket starts decelerating shortly after burnout. You can mount it lower, but you need to glue some stops to the inside of the nosecone to prevent the mechanism from moving up.

Step 15:

On the aero-shell, mark out the boundary of the mechanism where the parachute will come out.

NOTE: You should make the hole only a little smaller than the cavity. This will help prevent the parachute from snagging on the way out.
Also mark any access holes to buttons and controls, and don't forget the altimeter vent hole if you have one fitted.

Step 16:

Now cut out all the necessary holes. We often only make small holes for the switches to help streamline the aero-shell. To make small holes just heat a large nail or screw on the stove top and push it through the plastic.

Step 17:

Next we make the support ring for the mechanism. This holds the mechanism in place against positive G's. Cut a cylindrical section from another bottle and make sure its diameter is a little narrower than the aero-shell bottle.

Step 18:

Curl one end of the section on an old frying pan on low heat. You can use the curling technique shown in the Splicing Video.

Step 19:

Trim the bottom of the section so there is approximately 1 or 2 cm of straight wall left.

Step 20:

Glue the entire mechanism to the top of the support ring.

Step 21:

Fit the rubber bands into the holes made earlier along the edge of the V. We use two skewer sticks on the edges to give the rubber bands a nice rounded edge to sit on. The sticks are held in place with a small piece of sticky tape.

Step 22:

Detail showing how the rubber band is threaded. Experiment with the rubber band size to give you different tension on the ejection plate.

Step 23:

You can add extra tape to the base to ensure a good join between the mechanism and support ring. (Silver tape shown here)

Step 24:

The parachute door comes next. How this is attached and how the door latch mechanism works will depend on your design. The door hinge can be on the side or top or bottom. We put ours on the side as it works better with our latch.

If you have your door with the hinge on the side, you can easily make the hinge from some packing tape. If you have your hinge door at the top or bottom you may want to use some other method as the hinge needs to be fairly small due to the curvature.

Step 25:

Cut out the door so that it is at least 5mm bigger than the hole all the way around. Often you can use the same bottle that you made the support ring out of.

Leave a small tab on door (visible in the previous photo) and bend it back over so that it makes a loop. Because this part of the door will have a lot of force pulling on it, we secure the bent over tab with some wire. You can use a heated paperclip to "drill" two holes through the tab and then thread the wire through it.

Step 26:

Attach the door to the aero-shell. This can be achieve with strong tape. You may want to use a piece of cloth instead that is glued to the door and shell for extra strength. In practice we find the tape is sufficient. Two layers of clear packing tape were used in the photo.

Step 27:

Now its time to make the other side of the latch that holds the door. This is made from a strip of left over PET bottle used to make the door and support ring. Again we secure the bent plastic with wire the same way as was done on the door. Here you can see the 4 holes made by a hot paperclip wire.

Step 28:

After securing the plastic loops with wire, cut out the slot for the door latch with a craft knife. Make sure the door loop is a couple of mm smaller to make sure the latching mechanism comes apart cleanly.

Step 29:

Using some long nose pliers turn a paperclip into the latch pin as shown. Make sure the end of the pin is sanded or filed to a nice smooth finish to stop if from getting caught in the latch mechanism.

Step 30:

Insert the latch pin through the latch and door, and tape the latch to the aero-shell. This makes sure that the latch and door are well aligned.

NOTE: Don't make it too tight. The door should sit snugly against the aero-shell, but the pin should be free to move.

Step 31:

The aero-shell is now almost complete.

Step 32:

Attach a length of string to the servo motor arm. (horn). Use a fairly strong thread or nylon line.

Step 33:

Insert the mechanism into the nosecone and thread the string through the appropriate hole above the latch.

Insert the pin in the latch and attach the string to the pin so that the string is fairly tight.

Step 34:

Okay time to go get some coffee here comes the tricky bit.

Now test the mechanism to see if it unlatches the door. You can cut the pin to length or adjust the string until all works properly. Secure the string knot with some glue to stop it from coming loose.

Step 35:

To secure the mechanism inside the aero-shell just tape the support ring to the inside wall of the aero-shell. You could secure it in other ways if you are expecting high G forces. You can use nylon screws through the support ring and the aero-shell.

Step 36:

The side deployment nosecone is now complete. You just need to attach the parachute to the rocket. Tying the parachute string somewhere near the center of gravity will ensure that the rocket falls mostly sideways helping to increase the amount of drag on the way down.

Step 37:

All that remains is packing the parachute and the nosecone is ready for flight. We normally tape the nosecone to a pressure tested bottle and keep the entire nosecone ready for when needed. The whole thing then just screws into either a Robinson or tornado coupling.

Make sure you test the deployment mechanism a few times before flight.

Step 38:

For lots more instructions, and videos on how to build water rockets please visit:


Be the First to Share


    • Unusual Uses Contest

      Unusual Uses Contest
    • Paint Challenge

      Paint Challenge
    • Laser Challenge

      Laser Challenge



    4 years ago

    can we use 1n4007 diode instead of 1n4001?


    11 years ago on Step 7

    dude i wanna ask how to make that parachute stuff can u plzz tell me step by step cause i wanna make this thing can u buy that stuff somewhere maing's better though


    11 years ago on Step 7

    dude i wanna ask how to make that parachute stuff can u plzz tell me step by step cause i wanna make this thing can u buy that stuff somewhere maing's better though


    12 years ago on Introduction

    I know this sounds stupid but where's the center of gravity supposed to be on a water rocket? i will be fitting mine with this parachute mechanism plus I am using RC to control the parachute to try fly it back to me.


    Reply 12 years ago on Introduction

    Actually it is a very reasonable question. You should always work out the Cg for a dry rocket. This is because most of the flight the water rocket actually has no water in it. The Cg should be about 1.5 rocket diameters ahead of the Cp. When flying with fully open nozzles the water is ejected in the first couple of meters of flight so there really isn't that much time for the weight of the water to have an effect on stability. When flying with restricted nozzles you tend to want to make the rocket a bit more over-stable, ie. move the Cg a bit further up as the water stays in the rocket longer. Very good idea about flying the rocket back under parachute. :) Will you be using some kind of steerable parafoil?


    Reply 12 years ago on Introduction

    When making your Instructable, I found a few problems (with my build). As I am using a "tommy timer", I found that half a rotation of the wheel, pulled the pin out within 2 seconds. I have an 'arm' on the slowest moving wheel (to connect the string to, like your servo) When I twist it once round, the pin is pulled out releasing the parachute, like I said "within 2 seconds", which unfortunately is a bit to short of time for when I want my parachute deployed. I do not intend to fault your design, but I have found a way on how to fix this. Instead on having a pin pulled out i now wrap an elastic band around the rocket to keep the door shut and then wrap it round the "tommy timer". When that un winds it releases the parachute. I think your design is brilliant, along with the instructable. I'm just saying this so if someone designs it with a tommy timer and have the same problem, they can fix it. If there is a way to fix it.. I would like to know Thanks Oscar


    Reply 12 years ago on Introduction

    I've only got 2 servos - so if all fails i would make it deploy two parachutes, one high up when rocket is starting to fall and the other a few feet from the ground for a perfect soft landing - haha


    Reply 12 years ago on Introduction

    yes, I am investigating the use of a parafoil like parachute. ...still looking into designing the shute though, from scratch (my parafoil kite is a bit to big) I have got all the details to build the rocket from your site - thanks!

    Thats amazing, how high dose it fly (sorry i just skimmed the ible)? Do you have a video of it launching? iv seen thinks like this but never this intricate.


    Reply 12 years ago on Introduction

    This particular rocket goes to around 400 feet. Here is a bigger rocket that uses the same parachute deployment mechanism.


    Reply 12 years ago on Introduction

    That is amazing, it must been difficult to build and pressurized that rocket..


    12 years ago on Introduction

    Is there an instructable to make this rocket?!


    Reply 12 years ago on Introduction

    do you have problems with the centre of gravity with this rockets with the boosters at the bottom of the rocket?


    Reply 12 years ago on Introduction

    Yes and no. Just after liftoff the rocket is way too tail heavy and the Cg is actually behind the Cp, but the water drains very quickly so that in less than 1 second the rocket is stable again. We use a 2m long guide rail to try to get it pointing in the right direction as it builds speed. In an earlier rocket the fins were a little on the small size and we did get unstable flight during the initial part of the flight. Increasing the size of the fins fixed that problem up.