Hello, long time no see!

I've had a very busy winter, and I haven't been able to do as much programming and etching as I would've liked (gCn, I'm looking at you). The weather has been too cold to etch effectively and I've had my weekends filled with coaching ski lessons.

Now that winter is slowly fading, I can get back to whats important: Rockets!

If any of you need a refresher from last year:





Anyway, I have a better idea for my rockets. Instead of just building more and more ridiculous things for the lolz, I have an actual goal in mind. I want to create a guided rocket using off the shelf parts.

It will work like this. An arduino or arduino-like board will be in the body, serving as the brain. It will be powered by a battery and control canards, most likely 4 of them. I plan on using servos for this. Also included will be a 9DOF board, and hopefully a GPS and maybe a small bluetooth adapter if I can find one small enough (so I can build many of them and program them after the fact).

This project will go in hand with my 3dof launch pad I'm creating (I have the circuitry, I just need to build it). This pad will launch anywhere from 1 to 9 rockets from launch tubes or rods, which may or may not be guided.

I have a new 3D printer I got a few days ago just for this project. Here's the preliminary test fit:



I'll fix the dimensions later to get the servos just right. This is a test bed of sorts right now, the final product will use smaller servos and use a smaller body. Large rocket motors aren't cheap!


I feel very confident that I will be able to at least make this have a rocket go straight up and stable; I work heavily with head tracking here at work and I have the resources to help me out with the code should I hit snags.

Tell me what you think!

Also, I have no clue how I'm going to do the parachute part of the rocket.

I think this is a very cool project! I've always liked remote controlled things like helicopters, drones, boats, cars, and whatever else there is, but a rocket is something new. I always though that they'd be too fast to control, but I don't know exactly because I've only used rockets that go up, and come down with a parachute. For the parachute, you can probably get a small one online, and then attach it to the nose of the rocket, and have it come off when it starts coming down. The rockets I used weren't very good, but the parachutes worked. The top part popped out, and the parachute came out. Since looking on amazon didn't yield any good results, I'd go with making one out of bags and such (plastic bags, trash bags, cloth, etc.). The part you have already printed is really cool, and I have no doubt that the end result will be even cooler than that! Good luck!

Play KSP or do anything else with rockets, you notice they tend to go up, down, left, right, at you, through the map, and (sometimes) to light speed.

As for the topic, this is super cool! I've never had the resources to build my own rocket from scratch. Good luck!

I printed the second version of the body last night, but its still a little big. I will cut it down this evening, and possibly try printing with ABS. I can get more PLA, but my work has an excess of ABS we no longer use.

I'm actually using a program called Art of Illusion https://sourceforge.net/projects/aoi/ to model in and then use Cura 3.2.1 to cut the exported STL to gcode and then the Repetier software to print. I use AoI because its what I've used for 3d modelling since I was like 9 years old. I know I should use something else like solidworks, but the learning curve I feel is so steep.

The electronics should be arriving this week.

I actually got in contact with a guy who did a very very similar project and he told me how he got power to the system (he used a 7805 regulator on a 9 volt battery). I've been reverse engineering his photos to get my rocket body to look like his because I feel that his fin arrangement is ideal (Two staggered sets so the servos across from eachother can be closer)

Potential boards to use:


The largest round board is the Flora, which is too large.

The smallest circle is the LSM9DS0 9dof chip
The middle sized circle is the UART GPS module

The narrow board is the Arduino Nano that my friend had in his backpack ("I buy them in bulk from China, I flash them myself. You can keep this one").

I think I'll use the Arduino Nano because it is running the same 328 chip that the arduino uno does, so it should be completely compatible with my test setup of 4 servos and an uno.

I may switch to a pyboard because it has 4 dedicated servo ports

UPDATE:

I will continue this project "full time" once school is out (one more week!)

I tried printing a smaller version of the test bed so I could put it in a rocket body and fly it with just the hardware to test CG, but I printed it for servos the size of an ant. I need to figure out how to do this.

I'll try again today and ask the hardware engineer guys at work how to scale it correctly.


***EDIT***

I was just thinking, and the hardware part isn't the bottleneck here. I need to get started on the software bit.

I will print out an arduino "case" that I can attach to the bottom of the test bed tube and print out a support disk for the 9dof at the top. This way I will have a completely self contained system ready to program my software on. Meanwhile, I'd be optimizing the hardware for the actual rocket on the side.

Please tell me what you think I should tackle first, hardware or software. I feel that the software shouldn't be too difficult because what I'm doing has been done many times before and I have the people here at work to help me with it. The hardware isn't too complicated but I don't have as much experience with it and I still need to figure out a recovery system that works every time and isn't too difficult to implement.

I am pretty sure your best bet is to start with the hardware. That's the more difficult part, so you'll have that out of the way first, and you'll have a solid test platform to develop your software on.

Well, after a stressful night of 3d printer shenanigans, I present to you:

The Arduino Rocket Test Bed!










As you can see, the test bed has 4 9g servos with faux rocket fins on them. I am using an arduino uno as the controller (328P). I have glued down the tube, case, and LSM9DS0 9dof together with hot glue.

I have power going to the servos from the arduino, and so far it seems to work fine.

This weekend will be the start of getting the 9dof working with the servos!

Here's a VERY rudimentary prototype of my code. Its just on an if chain.

I will gather data this week and make a filter.

This is really awesome!

How long has rocketry been a hobby for you?

About a year, actually. I just started last year when I was wondering if it was possible to buy rocket engines on eBay one day.
And here I am.

Update again!!!

I just had an epiphany that if I used the accelerometer, I can calculate the X and Y angles of thrust, therefor correcting them.

AngleX = atan(y / sqrt(x^2 + z^2))
AngleY = atan(x / sqrt(y^2 + z^2))

It won't work for spinning and it isn't easy to change it for angled launches, but as a demonstration, it looks pretty cool! I think I'll end up using the gyro for the project instead.

It feels nice to have a breakthrough.

Enjoy!

Interesting servo behavior with powering on.

So...this is the second Flora LSM9DS0 to die on me.

I think its because the inputs are supposed to be 3.3V but the arduino uno is 5V. I have it powered by the 3.3V output, but I think that the toll has finally been taken on the accelerometer. It now only reads the maximum values.

I will switch to an LSM9DS0 with a level shifter or an MPU9250 with a level shifter.

I can work on the gyro today, but it will only be a matter of time before I kill the gyro too. Luckily we have an excess of the boards in mind at work. I'll put it on the test bed tomorrow.

Will be fixing the code this evening.
Flying season is picking up, so I'm basically always up in the air, whether its in a Schweizer 2-33 or if its flying RC planes.

This week looks pretty dismal for flying, so hopefully strides will be made!

***UPDATE***

Currently rewriting the code because alot of it got lost and messed up during one of my sessions.

I've also been really busy with flying, I'm working towards soloing a glider this season. I've also gotten into FPV flying and extending the range of my RC planes from just my yard to a good chunk of Waitsfield valley.

This week, I will print out the final housing for the electronics and assemble the electronics. I will get the small servos and use those instead of the larger, more power hungry servos of my test bed.

Also as a side project for my planes, I've hacked my cheap 6 channel remote into a 14 channel remote with a 2 watt booster on the antenna.

I know its been a long time, but between flying, school, and work, there really hasn't been much time for this project. I am determined to get it launched before the snow falls.

I have been using Autodesk Fusion 360 at work, and boy do I love it! It speeds up 3d modelling to a point where its no longer a pain to do so.

I am using a Teensy 3.6 for the final version (which I'm working on now). I'm also using 4 3.7g micro servos and a small lipo battery. I'll need to wire the servos into a buck booster circuit, but those are easy to get/make.

I'll have a render soon of the progress.

Well, this is better than a render!





I need some slight modifications.

These include:

1. Bumping out the sides where the servo ears touch the body so they lie flush
2. Find a battery that can fit in the tube on one side with the teensy in there (other side will house the buck booster)
3. Add a shelf at the top for the 9dof
4. Make an engine block with room for the wadding/parachute
5. Print some fins!
6. Nose cone!

I'm thinking about putting in an sd card so I can log the flight telemtry

Noooo! The Flora LSM9DS0 is discontinued!

Not even ebay sellers seem to have this board for any decent price.
The whole point to this board was the fact that it is very small and can be powered directly from the Teensy with no conditioning.

It seems that I'll have to find another breakout board. I don't want to have to switch 9dofs, but if I do I can. I'm thinking about switching to the "generic" Adafruit breakout board. Its much larger, but still fits inside the housing. I can sandwich it between the Teensy and the buck booster with the battery on the other side:

Aww, that sucks! Why'd they discontinue it?
How much room do you have in the casing? Could a different board mess up the balancing? (I know this is self-stabilized but could it still affect the rocket?)
I hope everything works out for you, I'm looking forward to the finished product!