Alright so here is what I am thinking. I have the major components on there way to me as we speak, 3 axis MEMS digital Gyro , A 3 Axis digital accelerometer, and 3axis magnetometer/Accelerometer combo in one chip. These are coming from ST Microelectronics.
The question is has anyone ever made there own IMU. I refuse to pay almost 100 bucks for one when I can get the chips for free to do it myself. I have atmega328's laying around just waiting to get used.
So if this project has been done before can anyone provide links or info about it. Board design, component placement, source code, etc... Thanks!
Replies
A lot of the manufacturers offer free samples of the components they make. Just the components though, you'd still have to design/build the appropriate circuit to make them work.
Daniel W Griffis said:
How did you get the parts for free?
I have, and I've documented the project on my site. I used a SparkFun 6DOF board and an Arduino. I then was able to fully simulate everything (hardware-in-the-loop) using X-Plane flight simulator. Worked awesome! Check it out on my site, NuclearProjects.com.
Nice! But I am not wanting to buy the IMU board (That would be the Sparkfun 6DOF) I am wanting to build mine. I am wanting three components on the board Gyro, Accelerometer, Magnetometer. It would be real easy to make the board and just put headers to access the pins. I've seen these IMU's with a MCU on the board and then that is interfaced with the Flight Controller. Then I have seen them like the sparkfun 6DOG where the components are interfaced directly with the Flight controller.
While your project would provide me with an example of interfacing the components directly with the flight controller, I also want to see the coding and hook up of how an IMU with a MCU is interfaced as well! Thank you!
What do you get per hour of work? :)
In comparison to getting your own stencils, paste, chips, routing your boards, going through a couple of test cycles and all the time you spend on that, $100 isn't really all that much for a working board.
That is one way to look at it... The other way is this. I am in school pursing an EE degree. So the experience I would gain from doing this project is worth more than everything you mentioned. I have the majority of the chips. I can do all the designing in Multisim and Utiliboard. I have un touched copper clad boards laying around. Some photo paper, a laser printer, an Iron, and some HCL or the Ferric stuff, and I have a traced board that I can then use to make a nice professional looking PCB. So in the end $100 is an amount of money that in my opinion is unnecessary for me to spend. Not to mention it would be something great to include in a portfolio for future job searches!
So with that being said if you happen to have any useful information I asked for in the first place please share I would love to learn. I would not ask about a starting a project if I did not have an Idea on the amount of work it would take to complete. I could understand your comment more if I was asking to build and code a flight controller from scratch, that is something I don't think I would attempt as it has already been done.
My goal here is to be able to build this and interface it with ardupilot preferably if not that then one of the other open source flight controller options.
Well, all APM boards are open source, so those are a great source of info. Also have a look at the ArduIMU v3 board, that's a much simpler board and doesn't have all the complexities of the main AP. Matter of fact, I believe it's available for $75. That arduimu is also opensource and actually uses a 328P. It's not enough to run an AP on it, but it should be enough to build a flight support module (wing leveling, etc).
Before you start, check if there's source code available for your chip. There's been some patent/licensing issues on the 6050. Configuring those IMU's is going to take some time, but once it works it... works.