I wanted a continuous servo for a project I was working on. I'm not actually sure I'll use it, but I wanted to try it. Anyways, I attempted to do it on my own and ruined a perfectly good servo. I looked up how to do it and found a robotics website that explained a not very good way to do it and then someone commented a great way to do it. Based on that little bit of info I was able to figure exactly what he meant and I thought I would share what I learned since FliteTest has no articles on this matter.
I removed the stickers off of the servo. Not really necessary, but I had hot glue stuck on this servo and the sticker came off with it. It is probably a good idea to remove the stickers anyway. Then you can scuff up the plastic so glue sticks to it better.
Step 1: Remove the four screws on the bottom. I recommend taping the bottom piece to the middle piece of the servo. You don't have to open up the bottom at all do it this method and it can be difficult to shove all the wires and the chip back in.
Here is a picture of the bottom removed. I did this before I looked up how to hack it. You do not need to remove the bottom.
Here is a photo of the servo taken apart.
Step 2: Open up the servo and remove the top gear. This is the one that holds the servo arm and it is the only piece that needs modification.
Step 3: Cut off the plastic tab on the top gear. Also, you will need to bore a bigger hole into the bottom of it. The way the servo works is that this gear is attached to a potentiometer. The potentiometer tells the chip what potition the gear is in. By making the hole bigger, this gear is no longer attached to the potentiometer, it moves freely, and the servo thinks the gear is still stuck in the middle. Do not bore out the entire hole. I'm sure if you did then you could no longer mount a screw into the top of this gear. I actually just used the tip of an exacto-knife and I slowly removed a little bit at a time. I kept testing it to see if it was loose enough to not grip the shaft.
This is a photo of the bottom of the taken apart servo. I was originally going to cut the wires off of the potentiometer and re-wire them so it thinks it is stuck in the middle position. I was unsure how to do that and when researching, I stumbled on the method of removing material from the top gear. It makes it a lot easier and fool-proof. Besides, I think you have to do that anyway, otherwise the potentiometer stops the gear from spinning any further from a mechanical stop built into it.
This is the top view of the taken apart servo.
Step 4: Before putting it back together, you need to put the potentiometer to the center position. I eyeballed mine and I did not get it exact. Which means I have to put some trim on it to get it to work properly. To prevent you from making the same mistake, hook up the servo to a powered Rx. Make sure there is no trim on your Tx. Turn the potentiometer until the motor doesn't spin. If you want, you can use a small drop of thin glue to hold the potentiometer in place. I did because I didn't want to run the risk of the top gear gripping and turning the potentiometer. You need to remove all of the gears to glue it. Make sure you keep track of which gear goes where.
Step 5: Put the servo back together. Then marvel in your awesome continuous servo. The servo will go both forwards and backwards. The best part is that the more you move the stick on your Tx, the faster it spins.
You might be wondering what applications there are for this. Mostly robotics. I know FliteTest used one for the RC blimp they made. I wanted to test out how fast it spins to maybe use it to power an RC car. I think I would have a very slow RC car.
Log In to reply
I have played around with continuous servos for positioning, and frankly they can be quite a pain. The signal line of the servo is quite sensitive and quickly ramps the motor up to full power once it gets away from center.
Log In to reply
Log In to reply
What if you used a third servo mounted on the camera tilt/pan system and used a Y-connector to hook it up to the pan servo. That way both servos move together giving you twice the pan distance. Maybe that would be bulkier than one 360 degree servo. It was just a thought.
Log In to reply
In this article I say that you might want to glue in the potentiometer, but actually, you HAVE to. I made another one of these for a project and even when the glue was drying, I had it hooked up to test it, the spin of the gear will move the potentiometer out of place. So after you know that the potentiometer is in the correct place, leave the glue to dry for longer than necessary with that glue.
The potentiometer shaft is bigger on top, where the top gear is put onto it. You could possibly sand that down, but that would just be more work. The outside case of the servo holds that top gear in just fine. You don't need the shaft to hold it very well.
The top gear can be rather hard to pull off and sometimes the potentiometer pops up too. Make sure you push it back into place.
I will also comment that after making a pair and a project out of these (RC CD Racer), I noticed these servos, at least with my cheap HobbyKing Tx, don't always fully stop when the Tx is in the center position. The servos are rather touchy about being centered. It might be better to modify the wires of the potentiometer. I don't fully remember how they work. I'm not sure if you can just cut the wires or if you have to solder two of them together. To clarify, it's not like the servo motor moves a lot. It very, very slowly moves, or at least you can hear the motor trying to move. If I play with my stick I can always get the servo to stop. I just thought I would mention that this modification isn't perfect due to how touchy the servo is about being in the middle position. (But it should still be perfectly fine for most applications where you would want a continuous servo. If you wanted exact positions, you would use a regular servo.)
Log In to reply
Log In to reply
Log In to reply