I’ve seen teams on here recommend the use of these larger pivoting bearings, but I am having a hard time imagining a use for them that isn’t rather contrived or over-complicated for most teams. If you’ve used one of these in the past and think it’s a good example, would you mind posting about it?
Prospective applications I can think of:
Turret for shooting. I can see this for very advanced teams, but I figure most teams should just pivot the robot instead of adding this extra complexity.
Arm rotation. But what would be the advantage of using this larger bearing instead of just a couple of smaller ones or some brass/plastic bushings?
We used one once, for a shooting turret. It mostly worked, but we did have problems with it.
Like most other parts that are made for something else, and adapted to robots, these bearings do take some effort to make work, and they don’t work as well as you’d like.
Team 5980 used a 22in Lazy susan bearing in 2018. We had the elevator and intake system mounted on the bearing. Helped us place and grab cubes, and made climbing with another team easier.
3946 used one for a wheeled shooter turret rookie year, feeding the ball into the shooter through the hole. The real weakness we found in this type of bearing is that it is intended as a thrust bearing (that is, supporting a load parallel to the axis of rotation) and most are not robust as at radial loads (that is, the recoil as we shot the ball sideways). I believe we broke two of them completely apart, spilling the ball bearings on the floor and went to a larger one (~12" hole) to mitigate (but not actually solve) the problem. When we mounted an air cannon on a turret a few years later, we used a drive train gearbox (TB-mini IIRC) mounted with a vertical axis.
If you’re also going to have a radial load or shock component and need that hole in the middle, I’d recommend something more like a slewing ring.
There isnt anything a lazy suzan accomplishes better than the 3 bearing and washer sandwich. The one pro is that it is a bit less manufacturing on your part, but they arent meant for robots, there are lots of stories of them just falling apart because they are not robust.
I think you are underestimating how difficult it is for many teams to make large rotating assemblies like turrets. Many teams don’t have the machining capability. Lazy Susans make it easy, and if you go for the high-load or heavy duty versions they are fairly robust. We had trouble getting someone local who could machine a round ring for use in our 2012 robot.
With regards to ‘turning your robot’ being easier, I think a number of teams would tell you otherwise. The momentum, scrub, and mechanical factors limit how accurately you can turn a robot. PID control will have hit-and-miss results.
Now that motion profiling is easier, you can do velocity and acceleration controlled turns to really “tune it in”, but it’s still much much easier to get accurate turns on a low-mass low-friction turret. Depending on how much accuracy you need, a turret may be the right way to go.
237 used one for a turret back in 2006, worked good for that.
I’d be hesitant to use one for an arm or anything like that where the side loads will change dramatically as it rotates, especially if it’s at the base of a long arm that outreaches. That’s just asking for issues due to drastic weight shift, impacts of the arm traveling back to it and likely having a short service life due to that and failing or binding at the most impromptu time.
Any heavy duty type bearing like that to handle those loads will be physically heavy to start with and likely not practical on a FIRST robot.
Can you pass an object to be launched through a bearing-and-washer sandwich? If not, then that is definitely something that a lazy susan can potentially do better.
The basic deal with a lazy susan bearing is that you have a large, stable turning bearing. It doesn’t cost an arm and a leg. It’s relatively sturdy as-is. And you can put stuff up through the middle of it to make handling said stuff a hair easier.
Granted, your random store-bought lazy susan bearing might not be built for robots. But that doesn’t necessarily mean that there aren’t lazy susans that are built for such applications, that may be findable with just a little bit of google-fu.
Our 2017 turret setup used a circle of 6 of these stackups, but we’re seeing good results with circles of 4 and 3 on our demo bot turret and dye rotor.
For 10 hardware, we use the following stackup:
1601 bearing 10 washer
R3 bearing 10 washer
1601 bearing
add washers as necessary depending on plate thickness.
You can definitely pass objects through the center of a bearing washer sandwich turret. That’s exactly what 254 in 2016, 148 in 2017, and a number of other teams have done with their turreted shooters.
That’s not to say that lazy susans don’t have benefits over a custom version. If you don’t have the machining resources, they’re definitely an option*. But in general a custom bearing washer sandwich setup is probably cheaper and more robust than a lazy susan.
*whether they really provide a competitive advantage for teams without resources to make a custom setup over the extra programming and driver training time is a different question
195 used lazy susan bearings for their turrets in 2017 and 2019. Both years, to my knowledge, they have held up without breaking, and they are simple. The ones we used were EDMed, so depending on your situation, it may or may not be in your favor to do something similar.
Yes you are correct, but turrets are, from what I can gather, are about 2 things: quicker aiming, and dealing with defense. Depends on what compromise you want to make.
987 has used a lazy Susan/slew ring on a variety of robots now for turret applications.
In 2007, the scoring arm was placed on a turret allowing it to rotate and reach in any direction around the robot:
In 2012, it was used to turn the shooter with assistance from the kinect for targeting:
In 2013, a larger one was used in combination was a rotating cam to allow the turret to rotate on two axis:
In 2016, the robots arm was mounted on a slew ring to allow 2 axis of rotation again:
In 2017, 987 had a turret with similar functionality to their 2012 turret, with only one axis of rotation to allow the robot to shoot balls from any angle (made shooting from the field hopper much easier):
We did a very large bearing sandwich design for our turret this year and it worked flawlessly. All of the parts for it we’re machined on our CNC router.
