Additional external bearing required or relying on motor bearing 1

[Hubert.]

Hi All,

I'm in the process of designing my own 3D lidar scanner (mostly for fun), but since I'm not a mechanical engineer I would love to get your input on bearing requirements.
The lidar scanner platform basically consists of two motors each being able to turn the lidar in 1 axis. The platform will probably be used later on to mount different and potentially somewhat heavier units (up to 500 grams) than the currently installed lidar unit.

The current design looks like this:

I'm relying on two gyems motors, that both have their own microcontroller and 16-bit magnetic encoder. In my tests these motors provide excellent precision positioning, but data on the internal bearings is limited.
This makes my somewhat worried about the bottom motor, which is listed here: [URL unfurl="true"]http://www.gyems.cn/794879.html
[/url]

This motor would have to carry the weight of most of the scanner, and although it is plenty powerful I wondered what the impact on the lifespan of the bearings would be. The only information I can find on their website is that they use two roller bearings.
I was thus playing around with the idea of adding a bearing of my own in the form of a tapered roller bearing to take up both the axial and radial forces. Besides this potentially being overkill, the other problem with this solution is that the bearing would rest on the bottom (stationary) part of the scanner. Thus, if you would lift the scanner by the top part the weight of the bearing would essentially hang onto the bottom part of the motor (as the motor is the only thing holding both parts together). Since the tapered roller bearing adds a fair amount of weight might this do more harm than good?

Are there different bearing solutions that could also tie the rotating portion and the stationary base together, or is it just not needed?

 

[drawoh]

Hubert.,

Most motor specifications provide the maximum shear and moment loads on the output shaft. If your scanner exceeds the load capacity of your motor, you have three choices...

[ol]
[li]Select a larger motor that can take the output load. [/li]
[li]Select bearings big enough to hold your scanner, and connect the motor through a flexible coupling.[/li]
[li]Select bearings big enough to hold your scanner, and select a frameless motor.[/li]
[/ol]

You can buy angular contact spindle bearings that are lapped so that they are pre[‑]loaded when assembled face to face.

If you design your system for an adequate level of rigidity, it probably will be very strong.

--
JHG

 

[BrianE22]

Can you provide a sketch of your proposed bearing setup? Tapered bearings are normally used in pairs. The 2 bearings react the axial preload applied through the bearings.

 

[Hubert.]

Hi JGH and Brian,

Thank you for your reactions.

I already own the motors, and would thus like to use the ones I have.
The main reason for using these motors is that they have a very accurate internally running feedback algorithm, so directly mounting them to the moving parts eliminates the introduction of any possible slack.
So this basically eliminates the suggestions by JGH, but I'll definitely keep them in the back of my mind for future designs!

Below are illustrations of my bearing setup for the bottom part of the scanner (the top part is not really a concern).

In the picture above the motor is selected, making it clear that the bottom stationary yellow part is only connected to the blue spinning part trough the motor itself.
This would normally mean that when one is to lift the scanner by the blue part the motor would have the weight of the yellow part pulling on it and the heavy bearing too.
To prevent this I have added the red part, which grasps onto the inside of the roller bearing from below thus lifting/pressing it into the blue part. As this inner part of the bearing constitutes about 3/4 of the weight doing this should help a lot.

To preload the bearing I'm planning on leaving a little gap between the motor and the yellow base. So that when I screw the motor to the base I can apply some preloading force by tightening the bolts.

I hope this makes sense, and I'm very interested to hear what you think of this solution!

 

[drawoh]

Hubert,

Controlling preload by tightening screws is incredibly difficult. The tensile strength of a single M4[×]0.7 A2[ ]stainless steel screw is[ ]4KN, almost certainly way in excess of the load your bearing can take. Tiny changes in torque will massively affect the force you are exerting.

You can order thin section bearings where the balls contact in an X[‑]pattern. The bearings are pre-loaded radially, axially, and in moment. Unfortunately, they are not cheap.

--
JHG

 

[Hubert.]

I hadn't really expected that to be so hard, but it makes sense now that you mention it.
However, in this particular case, wouldn't the flexibility of the 3D printed material in between the point where the motor is screwed down and the contact point with the bearings cause enough flexing to prevent such loads?
(essentially acting as a kind of spring)

 

[drawoh]

Hubert.,

Controlled spring loading of the bearings is a good solution. This makes the preload not dependent on how you tighten the screws. It also allows you to tighten the screws down hard, which is good design practise.

--
JHG

 

[BrianE22]

It's a taking a bit for me to see what is going on but it looks like you are clamping the inner race of the bearing between the blue and red pieces? If so, how is torque transmitted from the motor to the blue piece? From what I can tell you would have to control some dimensions fairly tight to get clamp load to the motor shaft and also the bearing inner race. There might be some axial compliance in the red piece - maybe that's enough.

 

[Hubert.]

Thanks for your tips JHG! I'll make sure to design it for enough spring action.

Brian, yes the inner race will be clamped between those two pieces. The motor consists of two parts, the whole top part (contained within the red piece and attached to the blue part) spins, the bottom part just under the red piece is fastened to the stationary yellow part. These motors don't have shafts, they have a rotating top "halve" to which you can attach things:

 

[Tmoose]

Is that a typical taped roller bearing with -
- A "cup" which is just the tapered ID outer race
- A "cone" consisting of the assembly of the inner race, the rollers, and a cage that keeps all the components together.

I do not see in your unlabeled images any feature to keep the "cone" from being disassembled from the cup, or even tipping when radial load is applied ?

 

[BrianE22]

These motors don't have shafts, they have a rotating top "halve" to which you can attach things:

Yep, by "shaft" I meant the top rotating plate of the motor. How to you transmit the motor torque from that plate to your blue yoke? Not just the 4 screws, correct?