Embedded Files
Optical systems require specialized mounts for precise control of optical components, but these mounts are not always available commercially and can be quite cost-prohibitive. 3D printing certain opto-mechanical elements can be a solution to these problems. This project serves to start the development of a library of designs for 3D printed optical mounts and other components used around the lab.
Methods
Methods
CAD and Rapid prototyping
CAD and Rapid prototyping
All designs in this project were modeled using Onshape, a cloud-based computer-aided design (CAD) software that supports real-time collaboration. Each part was developed through an iterative design cycle: (1) A rough model is created based on the functional requirements. (2) The part is printed and tested for functionality and mechanical stability. (3) Based on the test results, the design is refined and reprinted.
This rapid prototyping process enables optimization over multiple iterations. The top left image shows the design progression for a 2-axis kinematic lens mount developed during this project.
Slicer Software
To prepare a CAD model for 3D printing, it must be "translated" into g-code, which instructs the printer how to build the object layer-by-layer. This translation is done using a slicer. For this project, I used PrusaSlicer, an open-source slicer software developed by Prusa Research. In the slicer, users can adjust print parameters such as layer height, infill pattern, and print speed to suit the needs of each part being printed.
The printer
All parts were printed on an Original Prusa MK4 (most recent version is MK4S) fused deposition modeling (FDM) printer known for its reliability and print quality.
"Evolution" of the design of a 3D-printed kinematic lens mount developed as part of this project.
Parts for the 3D-printed kinematic mirror mount developed in this project, viewed in PrusaSlicer.
The Original Prusa MK4 printer used to print all components in this project.
Optical Mount Development
Design
Design
As a proof of concept, I designed and printed a 2-axis mirror mount (modeled after the Thorlabs KM100) and assessed its stability and precision. After evaluating its performance, I went on to design a custom 2-axis lens mount for precise coupling of a laser beam into a hollow-core optical fiber.
While the primary structure of each mount is 3D-printed, precision components, such as actuators, bushings, and springs, were sourced from Thorlabs to ensure mechanical stability and fine adjustment capability.
All design files are available in the links and resources section at the bottom of the page.
3D-Printed Kinematic Mirror Mount
3D-Printed Kinematic Mirror Mount
3D-Printed Kinematic Lens Mount
3D-Printed Kinematic Lens Mount
Here the mount is deconstructed to show the inner workings of the kinematic mechanism.
3D-Printed Kinematic Lens Mount
3D-Printed Kinematic Lens Mount
Testing
To evaluate the precision and stability of the 3D-printed mounts, I performed beam-steering tests using a HeNe laser directed onto a Thorlabs CMOS camera. I measured the displacement of the beam centroid on the camera as a function of discrete turns of the actuator.
The plots below show the beam displacement (in pixels) versus the number of actuator turns for both the 3D-printed mirror mount and a commercial Thorlabs KM100 mount.
Solid points represent measurements from the 3D-printed mount. Error bars represent the standard deviation of beam displacement, scaled x5 for visibility.
Open circles show data from the commercial mount. The size of the circle represents the standard deviation of beam displacement, scaled x5 for visibility.
These results show that the 3D printed mount is as accurate as the Thorlabs mount, but has less precision, as seen by the higher variance in pixel displacement measurements. There is little coupling between the degrees of freedom for both the 3D-printed and commercial mounts, i.e. the x-pixel doesn't move when the y-actuator is adjusted and vice versa. While the 3D-printed mount does not fully match the mechanical performance of its commercial counterpart, it demonstrates strong potential as a cost-effective and customizable solution for optical setups that perhaps don't require a ton of precision.
The results of the lens mount tests are shown in the plots below. Unlike the mirror mount, which used 100 TPI actuators, the lens mount is designed for high-precision alignment, so commercial micrometers were used in this mount to allow for finer movement.
The plots display the beam centroid displacement on a CMOS camera as a function of lens translation in millimeters, with the x-axis movement shown on the left and y-axis movement on the right. Each data point corresponds to a 0.5 mm increment of lens travel, and error bars represent the standard deviation of beam displacement, scaled ×5 for clarity.
For each millimeter of lens movement, the beam centroid shifted by approximately 280 pixels, with an average error of about 10 pixels. Given the CMOS camera’s pixel size of 3.45 µm, this corresponds to a displacement of (966 ± 35) µm, resulting in a relative error of less than 5%.
These results indicate that the 3D-printed lens mount, when combined with commercial micrometers, provides highly repeatable and accurate beam steering. The low variance and consistent displacement per unit travel make it well suited for tasks requiring precise optical alignment.
Other Library Components
Self Diffraction and Transient Grating Masks
The 3D printed masks pictured on the right are used in frequency-resolved optical gating (FROG) and dispersion scan (DSCAN) setups for characterization of ultrafast laser pulses. In these techniques, overlapping copies of a laser pulse in a nonlinear medium generate a signal beam through nonlinear optical processes. By scanning the time delay between pulses and measuring the spectrum of the nonlinear signal at each delay, the pulse can be reconstructed in time and frequency. These masks are designed to split a laser pulse into multiple beam copies with precise spacing and geometry.
Two nonlinear optical processes were implemented in the setup using the masks:
Self-diffraction (SD): The pulse is split into two beams and focused such that they cross in a third-order nonlinear crystal, forming an intensity grating that diffracts the beams to generate the signal.
Transient grating (TG): The pulse is split into three beams in a geometry that resembles three corners of a square. Two of the three beams create a transient grating in the nonlinear crystal, while the third delayed beam diffracts from the induced grating to generate the signal pulse.
Using 3D printing enabled quick fabrication and precise control over mask hole sizes and spacing, allowing the beam crossing angles to be easily optimized for our setup.
Threaded adapter
In addition to the mounts and the masks, I designed and printed a custom SM1 to C-mount thread adapter to mount a lens on the Thorlabs camera used to image the beam in the kinematic mount stability testing. This adapter demonstrates that accurate fine-pitch threads (40 threads/inch) can be reliably produced using 3D printing. The same SM1 threads were also incorporated into the kinematic lens mount, allowing a standard retaining ring to be used for lens mounting.
This project established a foundation for a growing library of 3D-printed opto-mechanical components for use in optics laboratories. The printed kinematic mounts demonstrated stability and repeatability comparable to their machined counterparts; however, long-term stability was not assessed, and the mounts may be less suitable for applications requiring high levels of precision. Additionally, the hardware cost for the kinematic mirror mount was similar to that of a KM-100 mirror mount from Thorlabs, limiting the cost advantage in this specific case. However, the SD and TG masks and threaded adapter highlight the key strength of 3D printing: rapid design iteration and customization for specialized experimental needs where commercial solutions may be unavailable or impractical.
Links and Resources
Links and Resources
Optical Mount and Component CAD Drawings
Optical Mount and Component CAD Drawings
Student Biography
Student Biography
My name is Leah Reid, and I'm an undergraduate in my senior year at Stony Brook University majoring in Physics with a specialization in Optics. In my free time I enjoy rock climbing, reading, and playing Dungeons & Dragons.
Page updated
Google Sites
Report abuse