July 2025 to Oct 2025
A rotating LiDAR mount concept for a Garmin LIDAR-Lite v3 sensor, focused on vibration control, bearing preload, and slip-ring-ready wiring.
![[sarit lidar mount v1]](/assets/[sarit-lidar-mount-v1].jpg)
First printed LiDAR mount prototype; the physical print exposed fit and rigidity questions that were not obvious in CAD.
Quick Read
- Started after the trailer hitch project wrapped.
- Designed around a Garmin LIDAR-Lite v3 sensor.
- Main issues were vibration, repeatable rotation, and wire routing through a rotating joint.
- Identified slip ring need for continuous rotation.
- Built v1 physical prototype.
- Iterated v2 around bearing preload and rigidity.
- Project stopped when team priorities shifted to power delivery.
Need
The technology team was integrating a Garmin LIDAR-Lite v3 sensor for object detection and distance measurement.
The mounting problem was different from the camera mount.
LiDAR data quality depends on the sensor staying rigid and repeatable. If the mount rotates, small wobble becomes angular error.
Research
The first phase was BOM and mechanism research.
- Vibration can be reduced with TPU pads, rubber grommets, or mechanical isolation.
- Continuous rotation needs a slip ring for power and signal.
- The sensor had six wires, which made slip-ring routing manageable.
v1 Prototype
The first CAD version was completed in early August 2025 and printed as a PLA prototype.
The print was not intended as final material. It was a fast physical check before committing to PETG, nylon, or another stronger material.
Having the prototype in hand made the rigidity and assembly issues easier to see.
v2 - Bearing Preload
v2 focused on bearing preload.
Without preload, even a good bearing can have small play. In a rotating LiDAR mount, that play becomes measurement error.
The challenge was designing preload into a printed assembly:
- Small forces.
- Tight tolerances.
- Plastic creep under sustained load.
- Assembly sequence had to stay reasonable.
I spent about three weeks refining the geometry, improving rigidity, and simplifying the assembly.
At one point the CAD assembly became hard to edit because of suppressed features and interdependent sketches, so I rebuilt it from scratch.
Outcome
The project was discontinued in early October 2025 because the Power Delivery System became more urgent.
The v1 prototype and v2 CAD were archived.
The work was still useful: the bearing preload problem had a design direction, the slip ring need was identified, and the next person would not be starting from zero.
What I Would Improve Now
I would build a simple wobble test fixture earlier.
For a rotating sensor mount, the useful measurement is not just whether the part fits. It is how much angular movement exists under load.