Electronic stager and rocket computer using an Arduino

[teper]

We are thinking of getting an Arduino Pro mini and designing a control system ourselves. We have an idea that we would like to share with the community. Using an electronic staging system. An accelerometer detects when the rocket launches. when the rocket starts to decelerate, the burn is over. the controller waits a certain amount of time, and then launches the second stage. The parachute would either be released when the sustainer starts to deaccelerate and then the controller waits X seconds, or we use an uMAD to detect apogee. How does this sound?

[air.command]

Sounds like a good plan. Using an accelerometer to measure the deceleration gives you the benefit of optimizing the release in real-time. You can even use it to detect when the rocket is about to start decelerating, since the actual release mechanism can take a certain amount of time from the time that you activate it, to the time staging actually happens. Are you considering using the one controller for both the sustainer and booster (actuator on sustainer), or two separate units?
An accelerometer you might want to consider is this one: http://search.digikey.com/us/en/product ... ND/2444757 These are about as inexpensive as they come.
When staging rockets, I'd recommend using the the uMAD to detect apogee as opposed to time specifically. It's a little more difficult to predict time to apogee accurately if the second stage does not quite go vertical. It's especially useful if the stager fails to release.
I'd also tie the booster parachute release to the staging event. You'll find that generally during staging the booster's velocity reduces considerably, and by the time the parachute is deployed and fully inflated the booster is on it's way down already. Boosters typically don't fly high, so you need to deploy your parachute fairly quickly.

- George

[teper]

We are thinking about having the controller on the sustainer, and a string or wire trailing from the cargo bay to the staging mechanism. The Arduino is quite expensive. Do you think the TI Launchpad will work for this application? The accelerometer you recommended requires you to surface mount it. What equipment is required to do that? what we have is just a soldering iron. We were thinking of getting the accelerometer http://www.sparkfun.com/products/9723 and plugging it into the Arduino/launchpad. If you have any other suggestions please post them.

[air.command]

I believe either the Arduino or launchpad should work just fine for your application. It's really up to you which platform you choose. They should both be able to interface to the accelerometer. Maybe do a search on examples of people interfacing the specific accelerometer you choose to the two platforms, and see what code is available, and which programming tools you are comfortable with.

You really don't need any special equipment for soldering SMDs - just a fine tip on your soldering iron and fine solder. If you are needing to solder one particular SMD component, then there are lots of different breakout boards available that have the right footprint for the component. Like these: http://iteadstudio.com/produce/breakout ... r-smd-ics/

The Sparkfun accelerometer looks pretty good too and since it's already soldered to a breakout board would make it a lot easier to use.

If you decide to use the accelerometer for calculating apogee, make sure the accelerometer you choose will cover the range of G forces your rockets will experience. If you exceed the accelerometer G ratings, errors will start accumulating in your calculations due to clipping since you won't be able to measure the actual acceleration above the rating.

- George

[U.S. Water Rockets1]

We are thinking about having the controller on the sustainer, and a string or wire trailing from the cargo bay to the staging mechanism. The Arduino is quite expensive. Do you think the TI Launchpad will work for this application? The accelerometer you recommended requires you to surface mount it. What equipment is required to do that? what we have is just a soldering iron. We were thinking of getting the accelerometer http://www.sparkfun.com/products/9723 and plugging it into the Arduino/launchpad. If you have any other suggestions please post them.

The Launchpad is hard to beat because of the price alone. You can use other boards if you like, but you may find people will be reluctant to adopt your design because the build costs may price them out of the market. People in water rockets tend to like to keep the costs down in their builds because the designs tend to be much heavier than pyro rockets, so malfunctions tend to have much more destructive results.

TI also makes some small development boards in their EZ430 line. You might want to look at the EZ430-RF2500 which comes with a couple of modules that can talk wirelessly with each other. You could mount the accelerometer in the sustainer and have it wirelessly command the stager on the booster to release.

We've used the wireless modules for remote camera triggering in the past, and they are quite easy to use and program.

[teper]

We decided to go with the Launchpad. It came yesterday. We are having a problem getting the servo to pull the release mechanism. Under pressure, the locking tabs put to much force against the collar and it becomes hard to pull. I have included a drawing of how the stager works. If More drawing are needed, just ask. The string in the drawing meet, and go to one servo at the top. That servo is connected to the Launchpad. Would a lever work to make it easier for the servo to activate the stager? Maybe more pulleys?

[U.S. Water Rockets]

We decided to go with the Launchpad. It came yesterday. We are having a problem getting the servo to pull the release mechanism. Under pressure, the locking tabs put to much force against the collar and it becomes hard to pull. I have included a drawing of how the stager works. If More drawing are needed, just ask. The string in the drawing meet, and go to one servo at the top. That servo is connected to the Launchpad. Would a lever work to make it easier for the servo to activate the stager? Maybe more pulleys?

In concept, your stager design would work, but you would need to measure or calculate the amount of torque needed to pull the release and get a servo which can supply that amount of torque. Just going on q guess based on prior experience with gardena connectors, it probably means a large amount of torque and therefore a large servo. A big enough servo would probably draw more current than the Launchpad can supply, so you would need to connect the servo + and - directly to your battery. In this scenario, you power the Servo and the Launchpad directly from the same battery, but you select a battery voltage that is safe for both devices. Higher torque servos generally run off 5V or 6V so you would feed that into the servo power inputs and the servo control signal comes from the Launchpad. The Launchpad can accept from about 3.5V to 10V input, so it will be safe to connect the servo power directly to it as well.

The drawback of this is that you a big battery and bigger servos are needed for a lot of torque, and those get heavy. You may want to stay with a smaller servo and try and create a way to move the gardena that doesn't need a lot of torque, or to modify the gardena so that it requires less force to activate.

Think about that and see what you can come up with and keep us posted!

[arjan]

maybe this will work,

if you put some extra pulleys on the collar and lead the string over these pulleys back to the bottle and attach the string to the bottle you would need just half the torque, the servo needs to pull twice the distance but i think that wouldn't be a problem.
I hope you'll understand, i made a quick edit of your drawing, think that would make it clear.

-Arjan

EPSON001.JPG (186.38 KiB) Viewed 129 times

[teper]

Could we use the launch pad to control the 808 Car Keys Type #11 HD720P MiniDV Camera? We want it to take a video of the launch and when the rocket reaches apogee, take a couple still photos. Is this possible? Also, is there enough ports on the launchpad to connect 2 servos, an accelerometer, and the camera? Thanks for the quick responses you guys.

[air.command]

We decided to go with the Launchpad. It came yesterday. We are having a problem getting the servo to pull the release mechanism. Under pressure, the locking tabs put to much force against the collar and it becomes hard to pull. I have included a drawing of how the stager works. If More drawing are needed, just ask. The string in the drawing meet, and go to one servo at the top. That servo is connected to the Launchpad. Would a lever work to make it easier for the servo to activate the stager? Maybe more pulleys?

You've probably thought about many of the following issues, but here are a number of things to keep in mind while you are designing your stager:

- Bottles can stretch considerably when pressurised. A string running along the length of the sustainer will be pulled tight as you pressurise the rocket. It could potentially activate your stager while still on the pad. A single 2L bottle will stretch 6mm along it's axis when pressurised to 120psi. The more you stack the worse the problem. Consider locating the actuating servo closer to the staging mechanism, or have enough slack in your line to compensate. You can just run wires to the servo from the top of the rocket, but keep little loops in the wires to account for the stretch.

- Protect the servo from splash back on release. Quite a bit of water will fly forwards during staging.

- Consider how you will support the sustainer during the boost phase. One of the most common failure modes of a staging mechanism is that the sustainer tilts to one side during boost and either the rocket ends up flying in an arc, or the staging mechanism snaps due to the forces involved.

- You generally don't want to pull the gardena collar directly from a servo. You need to have a strong enough servo for that. It doesn't necessarily have to be a larger or heavier servo that draws more current but one that has a higher gear ratio but will be slower to respond. These are not as common and not as cheap. Consider using springs or rubber bands to provide the pull back force necessary to release the collar and use the servo to release a locking latch that does not require so much force. You'll probably also want to remove the internal spring to make it easier.

- Consider what type of non-return valve you will use in the "nozzle" (the purple component in your drawing) part of the booster, to stop water flowing back into the booster while on the pad.

- Consider what procedure you will use to safely abort a launch, and depressurise your sustainer.

- consider how your mechanism will handle collisions with the ground. Make sure it is mounted well so it does not get damaged on landing.

- Use silicone grease on all moving components. This makes things move a lot easier when the stager is pressurised.


- George

[teper]

To answer some of the thing you asked
We are using silicon greasing

We have removed the spring from the gardena

the sustainer is supported by a ring of PET

We have a one-way valve in the nozzle

Forgot about the stretch of bottles. we will have to compensate for that

thanks for the input.

[U.S. Water Rockets]

Could we use the launch pad to control the 808 Car Keys Type #11 HD720P MiniDV Camera? We want it to take a video of the launch and when the rocket reaches apogee, take a couple still photos. Is this possible? Also, is there enough ports on the launchpad to connect 2 servos, an accelerometer, and the camera? Thanks for the quick responses you guys.

The still photo mode of the 808 cameras (all of them) are extremely poor. You would be hard pressed to tell the difference between a still photo from the HD720P and a frame taken from the movie mode expanded to the same pixel dimensions. You would probable be better served by running the camera in movie mode and taking still frames from that and running them through a photo stitching program and making a panorama of them. U.S. Water Rockets pioneered the technique and taught all of the self proclaimed experts how to do it several years ago. We have a tutorial on that if you're interested in learning more. You will be much happier with the results of the panorama and it will be a lot easier to control this. Plus, you end up with a movie as well that may be a useful tool in troubleshooting your rocket. The sounds and things you can see on camera can often point you to a problem (i.e. you may spot a leak in your launcher you would not normally see because you are staying back a safe distance and don't see the leak).

There are definitely enough GPIO ports on the Launchpad to do what you want. Plus, you can free up more ports if you choose to not use the RS-232 MSP430 Applicatuion UART. That takes up 2 signals, which could read an accelerometer or something else. You could also eliminate the launch detect input pin because you would use the acceleration to detect launch.

Keep up the good work! These are excellent questions.

[U.S. Water Rockets]

To answer some of the thing you asked
We are using silicon greasing

We have removed the spring from the gardena

the sustainer is supported by a ring of PET

We have a one-way valve in the nozzle

Forgot about the stretch of bottles. we will have to compensate for that

thanks for the input.

If you use a significant number of pulleys to multiply your torque, you can most likely ignore bottle stretching because it will not pull the release very much since each pulley reduces the distance travelled and converts it into a force multiple.

The downside of this is that you would need to move the servo a long distance to move the release end far enough to trigger. A regular servo typically rotates less than 180 degrees, which will not result in a very large motion after the pulley reduction. A way to correct this involves getting a special type of servo that continuously rotates and does not stop, or to modify a servo by removing the feedback sensor that tells it how far it has moved, thus making a continuous servo. This could "wind up" a snall spool of line and pull a great distance through your pulleys resulting in significant force at the release over a the distance needed. There are plenty of tutorials around on how to hack a servo into a continuous one.

[teper]

I think the servo i have will be able to pull enough distance on the string.

[teper]

My servo rotates 180 degrees.
weight: 6.5 Grams/o.23 ounces
Torque: .8 kg/11 Ounces
Speed: .12 Sec/60 Degrees