Arduino Gimbal With Micro Servos

27,377

205

26

Introduction: Arduino Gimbal With Micro Servos

I find gimbals really fascinating even just to look at. So lately I decided to make my own since the ones out there are so expensive, and making stuff is so much cooler than buying stuff.

I quickly realized that they are usually made with a gimbal controller board and brushless motors. And they are super easy to make, so maybe later I will make one of those. But I also have a lot of components laying around, and just because i was curious i decided to make a arduino gimbal with micro servos

luckily I found this instructable by woojay wich became the basis for my gimbal.

I had no hopes that it could even support my action camera, but to my surprise it did. It has some delay, but its a fun and interesting project and its really cheap.

Step 1: Partslist

The parts needed for the arduino gimbal with micro servos is:

2 x 9g micro servos

1 x arduino (i prototyped with the uno, and in the end i used the nano just because it so little, you can use wich one you like)

1 x mpu6050 accel/gyro sensor

2 x 10k resistors

Power supply (i ended up with 4x1.5 AA batteries) IMPORTANT, the power output from the arduino will not be enough to drive the servos itself, you NEED external power.

And the casing and arms. I printed this item, my design can be found on thingiverse. And if you have a printer yourself, i would love to see some other designs, mine was rough and could probably be a lot more refined.

Step 2: The Prototype

So i searched a lot around the web to find anyone else making a gimbal with micro servos. And i couldn't find much info, so I assumed that this wouldn't even work since the internet wasn't full of info. But there was one instructable that made it all happen. It was a prototype of the system made by woojay. This really helped a lot since I am not a programmer in any ways.

So i went ahead and built his system with his code and it worked without any issues.

The only thing to be clear about here is the calibration of the MPU6050. Everyone of these sensor are unique in there own way, therefor you need to run the calibration program stand alone on your MPU6050 first to read its characteristics. The the value you get out of you chip must be edited in to your specific program for the gimbal.

For the mpu6050 to work you also need I2Cdev and the MPU6050 Libraries installed.

For this sample to work, the I2Cdev and the MPU6050 libraries need to be installed. Find the calibration program and wiring diagram attached. Also check out woojay's instructable, he has good wiring diagrams.

Step 3: Going Forward

So i made woojay's system, then it is time to turn it in to a real gimbal. This is all about creativity and how you want to make you specific system, you can use my step files and print it, or make something completely different.

I used a nano instead from this point to make it smaller and a cheap protoboard to make it one unit.

I put the mpu6050 inside the main box. I've read that maybe putting it on the camera side (the arm the camera sits on) might be a better solution. So that is up to you. Everything would still be the same if you choose to do so.

As you can see from pictures and video the first arm configuration i design was to flexible, maybe I should have seen that one coming, but anyways I am giving you a heads up now if you should want to design your own. I also added a switch between the battery and arduino.

Any questions leave a comment, and read more at my blog

Share

    Recommendations

    • Water Contest

      Water Contest
    • Metalworking Contest

      Metalworking Contest
    • Fix It! Contest

      Fix It! Contest

    26 Discussions

    Why just 2 and not 3 axis? There is a lot of difference!

    The I2Cdev library can not be inserted into Arduino 1.8.4, it does not contain the correct files .. Something like that says :-( Please help, thanks ..

    enabling interrupt detection (arduino external interrupt 0).... Nothing happens after printing this... What should I do?

    How do you download those libraries, because it isn't working for me.

    Why not use 5v for gyro vcc instead of 3.3v?

    hey please help! i am not able to open the code .Please share the code here or please mail gimbal code to my mail - theultimateforcerobo@gmail.com .

    thank you, nice works

    hi there. i want to a question why we have two file code. i don't know why we have two file..tks if you reply. i really want to do that.

    I made the project with a Nano and it and it works great! Many thanks to Friedrikeide and Woojay! I will use it to show some teens how a gyro can be used to change the direction of a rocket. That's quite a leap from this project, but it's a start! Does anybody have suggestions for changing the servos from pitch and roll to pitch and yaw? Or even better, to use servos on all three axes.

    Gimbal 2 Servo project.jpg
    1 reply

    hi there. i want to a question why we have two file code. i don't know why we have two file..tks if you reply

    Looks like a fun project! Question - I may be wrong but I think there is a wiring error in your small picture of the servos, UNO, 6050, and the 9V battery. The 6050 GND seems to be connected to the UNO VIN. Shouldn't UNO VIN get +9V and 6050 GND get ground from someplace?

    I actually just finished building this! Thank you but the servos keep spazzing out. Any idea why? I calibrated the gyro too

    1 reply

    how cool you build it :D no i have no idea why, i have not encountered this issue. maybe someone here with more knowledge then me might help you..? or if you ask under woojays post?

    This is exactly what I was looking for! Arduino nano, 3d printed, those cheap 9g servos... wow! Awesome!

    Gonna have to look into the chip calibration but it doesn't look too difficult.

    Do you think the servos could support the weight of a smartphone?

    2 replies

    Thanks! The chip calibration just sounds difficult, it is not. I dont know about a smartphone. It might be just a bit to heavy.

    Nice quick proof of concept. Small servos are not meant for such a large cantilevered load. The life expectancy of this design could be improved by having the axes supported by a shaft running through a hole in your printed plastic parts and the servos only being used to impart rotary motion either with printed gears or via servo horn to a wheel on the drive shaft (s) via pushrod. This will reduce the off centered load on the servo output shafts which should improve the smoothness of the pan tilt motions.