DIY Mini Gimbal for Air and Ground Video

Snarls

Gravity Tester
Mentor
Hey everyone, I thought I should share my build here to get some feedback and hopefully inspire others to build their own DIY equipment. I am currently working on a mini gimbal to compactly stabilize my GoPro Hero 2. The goal is to mount the gimbal on the end of an extendable pole to use for filming on the ground. This is actually the second gimbal I have built using this technique, the other being on my 500mm quad, and the third I have ever built. The frame is comprised primarily of plexiglass that has been bent and shaped into the desired pieces. I can't afford to work with carbon or a 3D printer so I find plexiglass to be a nice material to build with. I built this gimbal over the course of two days choosing dimensions on the fly so I do not have any plans to give, but I hope I can teach you the techniques necessary to build your own.


The arms of the gimbal are made from strips of two ply plexigalss. Superglue cured overnight is great for bonding the two layers together, although acetone may also work.
2layer.jpg

I find it is easier if you cut the slots before you bend the glass. The slots allow for adjustment of the arms to facilitate balancing the gimbal. To cut them I use a dremel with a plastic cutting bit and the flush cut mount. A fence helps keep a straight cut and scrap pieces keep the flush mount level with the piece to be cut.
Cutting.jpg

To bend plexiglass you need some concentrated heat. A small tea candle will work just fine. I marked a line where I wanted the bend and held the glass shortly above the flame while moving in slow circles to avoid heating one spot too much. You'll know if one area is too hot if you see bubbles forming. After some time the plexiglass becomes flexible around the bend area. Once its ready I carefully bend the piece around a 90 degree form. A chamfer on the edge of the form helps because the plexiglass will not bend with a sharp corner without repeated heatings.
Heating.jpg
Bent.jpg

Completed bent arms with slots and motor mounts drilled.
Arms.jpg

Layout of the major components for this gimbal. Very little parts required.
Layout.jpg

Here is the plate that will be tightened to the motor with the slotted piece in between to hold the arms in place. It is made from a single layer piece of plexiglass. One side is covered with a very thin covering of hot glue to add grip.
Plate.jpg

Arm mounted to the motor. To fix some clearance issues from bad planning I used a thinner metal plate instead.
MetalPlate.jpg

Assembled mini gimbal with GoPro mounting strap.
Assembled.jpg

Gimbal mounted on my tuning stand and balanced with GoPro. To balance I simply loosen the two screws on the motor whose axis is in question and slide the arm whichever way is required. The control board is mounted with some carpet tape on the back of the stand.
Mounted.jpg

Board.jpg

I did some very quick tuning on the computer with the PIDs I use on my other gimbal and the initial tests are looking great. The mount between GoPro and gimbal works, but is kinda iffy so I may have to revise it. Now I need to create a way to mount this gimbal onto a pole to create my ideal "gimbal on a stick." I might also add some wire rope vibration isolation to help get smoother shots. Test footage will also come shortly.

If you found this DIY helpful, let me know if you want to see an article on DIY gimbals and or bending plexiglass. I scratch build a lot of my equipment so if you have any suggestions for other things I should make, let me know. Thanks for reading!
 

Snarls

Gravity Tester
Mentor
Sorry, it took me some time to get this gimbal up and running along with getting test footage. Here is what it looks like now:

Finishing the Build: I used an 1" diameter poplar dowel as the handle and added a flat area at the end to mount the gimbal controller and battery. I anticipated that the gimbal was going to have its limits at extreme angles so I thought a hinge I can tighten at any angle would be a good thing to have. I built one out of some poplar and plywood. After testing and filming I took it all apart and painted the handle black to give it a more professional look.

Hinge1.jpg

Hinge2.jpg

final2.JPG

final1.JPG

Testing:
When filming for the test video I took note of the limits of a two axis GoPro sized gimbal. With a handheld gimbal, especially one on a pole, it can be quite difficult to control yaw. This is most noticeable in the video when I am mountain biking, due to the fact it is really hard to hold the gimbal with one hand while navigating rocks with the other. In contrast, the shots where I'm longboarding and biking are very clean on the yaw axis because it was easy to hold the pole still.

The second limit of the gimbal I found is with large lateral movements. It can be hard to keep such a small and light gimbal exactly at the same height off the ground. The gimbal can hide rotation, but not side to side or up and down movements. This is once again exemplified by the mountain biking.

Also worth noting is that at high speeds the gimbal may shake a little because of the amount of air resistance.


Overall there are certain situations where this gimbal performs, and where it lacks. With wheels and flat surfaces, the footage can look great. On bumpy trails or yaw inducing activities, the footage can still be shaky.

What's Next: Finding the limitations of a two axis gimbal, it seems logical for me to experiment in building a three axis gimbal, possibly even one for a larger camera. I can not afford the parts for that project now, but I'm sure it will happen one day.
 
Last edited:

RAM

Posted a thousand or more times
For the bike you might try holding the pole upside down. It's and old diy stabilizer trick.
 

markyoe

Senior Member
Hey Snarls! Nice work on the project! If you were to do a 3 axis gimbal, what control board would you use and where would you get the firmware?
 

Snarls

Gravity Tester
Mentor
Hey Snarls! Nice work on the project! If you were to do a 3 axis gimbal, what control board would you use and where would you get the firmware?

Thanks markyoe! I am actually working on a 3-axis gimbal build for my TBS Discovery. I'm hoping to make a compact 3D printed design, but I may end up using the same acrylic techniques. The controller I got is the Micro Storm32-s controller from Hobbyking. I mostly got it because it is really cheap and small. I haven't delved into it too much, but it looks like it can be difficult to flash the firmware at first. The firmware itself is downloaded from a GUI that you get in the Storm32 Wiki. There is a larger version of the board which may be easier to use. The Storm boards look like the best after Alexmos boards which are a lot more expensive.
 

yeeee

New member
Sorry, it took me some time to get this gimbal up and running along with getting test footage. Here is what it looks like now:

Finishing the Build: I used an 1" diameter poplar dowel as the handle and added a flat area at the end to mount the gimbal controller and battery. I anticipated that the gimbal was going to have its limits at extreme angles so I thought a hinge I can tighten at any angle would be a good thing to have. I built one out of some poplar and plywood. After testing and filming I took it all apart and painted the handle black to give it a more professional look.

View attachment 50508

View attachment 50509

View attachment 50507

View attachment 50506

Testing: When filming for the test video I took note of the limits of a two axis GoPro sized gimbal. With a handheld gimbal, especially one on a pole, it can be quite difficult to control yaw. This is most noticeable in the video when I am mountain biking, due to the fact it is really hard to hold the gimbal with one hand while navigating rocks with the other. In contrast, the shots where I'm longboarding and biking are very clean on the yaw axis because it was easy to hold the pole still.

The second limit of the gimbal I found is with large lateral movements. It can be hard to keep such a small and light gimbal exactly at the same height off the ground. The gimbal can hide rotation, but not side to side or up and down movements. This is once again exemplified by the mountain biking.

Also worth noting is that at high speeds the gimbal may shake a little because of the amount of air resistance.


Overall there are certain situations where this gimbal performs, and where it lacks. With wheels and flat surfaces, the footage can look great. On bumpy trails or yaw inducing activities, the footage can still be shaky.

What's Next: Finding the limitations of a two axis gimbal, it seems logical for me to experiment in building a three axis gimbal, possibly even one for a larger camera. I can not afford the parts for that project now, but I'm sure it will happen one day.
What motors did you use?