Servos are bad. Change my mind

So why would you use a servo instead of a motor? Pros and Cons?

I don’t think it’s fair to just say “servos are bad” without specifying anything. There are tons of different kinds of servos out there, from $5 mini servos from China to specialized high power or linear servos.

7 Likes

I don’t agree with the statement “Servos are bad”. Yes, there are quite a few cases where using a motor has a benefit over servos, such as high load situations (elevator, main drivebase motor, etc.). However, for something like a pin that holds a part of your robot in, a simple plate which swivels up and down to keep a game piece in your bot, a shifter for your gearbox (though a pneumatic shifter is the better way to go, but that’s a different conversation altogether), or any application in which all the object has to do is go from point A-> point B and is under little to no load, then it’s definitely much more worth it to go with a servo since it’ll be smaller, and lighter than a motor and whatever gearbox you need, and can save you a motor controller since it goes directly into the PWM port of the Roborio.

tl;dr: servos can be pretty useful, just depends on the application.

9 Likes

Umm… a servo IS a motor so…

12 Likes

All servos are motors, but not all motors are servos.

Also completely agree with others above. Totally depends on the situation.

6 Likes

Servos, on an FRC sized robot, under the size limitations currently in the rules, aren’t very useful.

but, if the rules were relaxed (to allow high power 12V servos), or someone made a 50 to 300W, 100 to 500:1 gear ratio servo with FRC legal components (as discussed here), they’d be very useful. There are many teams that don’t have the chops (or time) to code up a position-feedback PID controlled motor & gearbox. Sometimes you just want to attach an actuator to something that just works.

6 Likes

“Bad hot take. Change my mind” threads are less than useful. If you’re going to take that stance, provide some reasoning behind it.

25 Likes

Cunningham’s law is a powerful thing

10 Likes

Servos are nice for “pinning” type mechanisms where you just need to actuate something once. Pull a pin and then the mechanism drops, etc.

3 Likes

Yep, although I find pneumatic cylinders are lighter and more reliable in this application (provided you have a pneumatic system already)

In 2017 our hopper had a 3" flap at the front which would swing down at the start of the match to help catch balls from the chute. All we did was zip tie a little servo to one side such that the arm could be rotated to hold the flap behind the frame perimeter, but would go to the “0” position and release the flap as soon as the robot was enabled. Super easy.

You can see the servo at the right of the clear flap here:

Servos used incorrectly are bad. Change your thinking.

7498 hatch intake used 2 servos to hold hatch panels in place securely. Watch some match footage and see what 2 servos used correctly are capable of.

3 Likes

I’d say that pound for pound, the servo is lighter–depending of course on the type/power of servo and type/power of pneumatic cylinder. But the cylinder will tend to be stronger as far as what can be pushed/pulled.

Also, WAY back in the day, it wasn’t uncommon to see cameras on pan/tilt units, which were powered by servos.

I am glad some people agree with this! Although, what I think you mean is, the small scale hobby servos available to FRC teams right now are insufficient. Powerful, robust servos exist and would work in this application, they just tend to be either too expensive or too intimidating for students to use.

DJI has made great strides in this, making (relatively) cheap servos that are powerful and also easy to program:

15 watts?

Yes that was more to demonstrate the form factor and nice software interface. You could build the same exact thing scaled up with a bigger motor, and sell gearheads to match.

I just wish their software would allow more than 7 CAN connected devices on a single bus (unless I very much misread the documentation when I looked at these for a project)

FRC-legal servos (meaning meeting both the price point and the ability to be powered from the RIO or a REV servo module) are unlikely to have enough power to serve as an FRC manipulator. However, there are still a number of FRC applications where a servo may be the right answer (examples, not necessarily complete):

  • aiming a camera or other sensor (or perhaps a flashlight or laser pointer)
  • shifting a baffle or pulling a retaining pin on a robot without pneumatics
  • flicking a switch or rotating a dial (potentiometer, needle valve, secondary pressure regulator, focus dial)

Didn’t 118 use a servo on their shooter hood in 2017 to adjust the exit angle?

This is true.

OP - my personal hot take - the title (combined with the lack of followup or specific technical details), in the context of the nature of the CD technical forum, doesn’t convey a true desire to learn. Being precise and considerate does convey that meaning.

I will still give you the benefit of the doubt.

To regurgitate the message of others concisely: FRC-legal servos are a valuable asset while designing some mechanisms, with many reasonable permutations where using a servo is the best option. Whether a specific scenario demands a servo or not depends on the specific scenario.

Care to dig any deeper?

1 Like

There are some powerful servos out there now that are FRC legal. We used one this year to poke the cargo ball into our shooter and never really had any issues with it. Just depends on the application, but it’s great when you can use one and not have to use a heavier motor / pneumatic solution.