Through this project, we aim to design a filter that removes dust from low-velocity gas-solid suspensions which are characteristic of lunar surface systems, but could be used for regular medical purposes as well.
We have produced a physical helical filter prototype using 405nm UV-cured transparent resin surface-treated with polycrylic spray. We also tested the filter by installing one end in the “Particle Zenith Device”, described below, and the other end in an industrial shop vacuum.
Currently, we are looking for additional volunteers with backgrounds in mechanical engineering and leadership to bring this project to the next level. If you are interested in joining the team, contact comms@pubinv.org or go directly to this calendar to schedule an interview.
A particle in a fluid forced through a helix feels a "centrifugal" force. If it is denser than the fluid, as most particulates in air are, it will be pushed to the ouside of the helix. A slit or opening there may preferentially allow the particle, or particle-heavy fluid, to leave the helix. The helix is thus acting as a simple "sorting" device. The remain fluid flow in the helix is cleaner than at was before the particle left.
Helices can be wound around each other and packed together to create compact geometries with many ports. This project is focusing, however, on optimizing a single helix for now.
This has a great advantage of fiber-based filters. In a sense it does not clog---or at least it does not continually clog degrading its performance. It may be possible to produce a filter that requires very low pressure difference to clean the air this way. Finally, instead of the maintenance of replacing or cleaning the filter, the filter is in theory renewable by just dumping out the filtered particles.
A fully developed version of this filter could be very commercially valuable system on Earth and a safer system in Space.
NOTE: “Re-entrainment” is a significant problem in our design, which could affect the experimental performance reported below.
We tested the effectiveness of the filter design using a range of particulate sizes by using two teaspoons of the following materials:
- powdered sugar (fine particle size distribution),
- table salt (medium particle size distribution),
- cornmeal (coarse particle size distribution), and
- mineral salt (large particle size distribution).
We introduced these individually into the gas-solid suspension generator. The particle sizes varied from 50 μm (powdered sugar) to 1500 μm (mineral salt). Suspensions were generated by injecting compressed air at the base of the chamber with the Particle Zenith Device, forcing the particles into an upward flow and creating a circulating gas-solid suspension that passed through the helical inertial filter.
In order to come as close as possible to simulating a low-gravity environment, the Particle Zenith Device projects particles upward in a semi-enclosed tube. It has a cone-shape funnel at the bottom attached to a source of pressurized air. This jet of air somewhat “fluidizes” the fallen particles, and blows them with a velocity nearly proportional to the air pressure. By adjusting the pressure, the maximum height of the particle, or zenith, can be partially controlled.
At the zenith, the particle momentarily has a minimal velocity. When the inertial filter is placed at that height, it extracts particles as it would in a low gravity environment. The top is open, allowing air to flow out and balance the air injected by the compressor. However, since this is a relatively small flow, the top can be mostly covered.
Particles which hit the lid generally fall back down into the funnel to be recirculated. A small number of particles escape into the room or get sucked into the helical inertial filter.
A sketch of this design is pictured below:
Schematic diagram of the particle zenith device
Photos of the Particle Zenith Device are included below. The attachment in the middle is where the filter is inserted into the particle chamber. The far right image shows white (powdered sugar) gas-solid suspensions emerging at a low velocity. It is possible to quantify and engineer the dynamic behavior of the suspension at the region of minimal velocity to better align with practical uses in future research. Below is an image of the helical and paper filter stages loaded with cornmeal (which has a distinct color) and powdered sugar after experimental runs.
Particle zenith device assembly and operation and filter condition after experimenting with cornmeal and powdered sugar.
We collected the material that passed through the filter ("pass") and the material that successfully remained trapped in the helical filter. We measured the weight in grams using a lab-grade precision scale to quantify the pass rate (the percent of material that escaped through the filter). Each trial lasted for 120 seconds, and the vacuum pressure at the filter outlet (which is interlocked with the shop vacuum) was 1.75 PSI. The results are shown below.
| Material Type | Material Passed (g) | Total Material collected in filter (g) | Pass (%): Filtrate / Permeate | Caught (%): Residue / Retentate |
|---|---|---|---|---|
| Powdered Sugar Trial 1 | 0.115 | 0.445 | 25.8 | 74.2 |
| Powdered Sugar Trial 2 | 0.115 | 0.296 | 38.9 | 61.1 |
| Powdered Sugar Trial 3 | 0.17 | 0.513 | 33.1 | 66.9 |
| Table Salt Trial 1 | 0.178 | 0.662 | 26.7 | 73.3 |
| Table Salt Trial 2 | 0.060 | 0.282 | 20.0 | 80.0 |
| Table Salt Trial 3 | 0.106 | 0.432 | 24.5 | 75.5 |
| Cornmeal Trial 1 | 0.862 | 2.984 | 28.8 | 71.2 |
| Cornmeal Trial 2 | 0.453 | 2.215 | 20.5 | 79.5 |
| Cornmeal Trial 3 | 0.501 | 2.454 | 20.4 | 79.6 |
| Mineral Salt Trial 1 | 0.050 | 0.219 | 22.8 | 77.2 |
| Mineral Salt Trial 2 | 0.076 | 0.283 | 26.9 | 73.1 |
| Mineral Salt Trial 3 | 0.128 | 0.487 | 26.3 | 73.7 |
We measured an average caught/retentate rate of 73.8% across all tested materials. These results indicate that the helical filter consistently retained the majority of the material regardless of particle size, with pass/filtrate rates remaining within a relatively narrow range (20.0-38.9%).
This suggests that the helical geometry successfully induces inertial separation across a broad spectrum of particulate sizes. The relatively similar performance across materials may further suggest that filter geometry and flow dynamics, rather than particle size alone, are dominant factors that govern collection behavior in the current design.
However, further study is needed to determine whether modifications to filter geometry, slit placement, chamber design, or operating conditions can reduce these losses, or whether they are inherent limitations of the helical filter architecture itself. This is one aspect of the project we are seeking to continue developing.
After this initial test, we isolated the filter with the shop vacuum attached and packed the filter chamber with cornmeal to test a “pre-loaded” filter in clean air. We ran the vacuum to assess what portion of material was re-entrained through the slits and collected this material in a paper filter. We observed a 16.2% pass, or re-entrainment, indicating that a substantial fraction of captured material remained susceptible to subsequent transport through the filter, limiting overall collection efficiency.
These tests were completed during our initial contract with NASA, in tandem with the M-COG’s development. This work established the feasibility of helical inertial filtration for particulate capture in low-velocity gas-solid suspensions and provides direction for future investigation. While the current prototype demonstrated measurable particulate retention across a range of particle sizes, the observed pass and re-entrainment rates indicate opportunities for further optimization.
Therefore, we will need to continue research and development of the filter to identify potential design improvements. Currently, this is our plan:
- First, we will run additional trials with each material. The first five trials will replicate the existing protocol to improve statistical confidence in the baseline results. In the additional tests, the filter will be rotated by 90° to evaluate whether slit orientation relative to gravity influences particulate capture efficiency and pass rate.
- Second, we will test two modified filter designs. One design will incorporate wider helical turns to alter particle trajectories and time within the filter. The second design will modify slit placement to evaluate the effect of extraction location on collection efficiency and re-entrainment behavior.
- Third, we will investigate the influence of the surrounding collection chamber geometry. Possible modifications include increasing the chamber depth, enlarging particulate collection regions, and incorporating additional passive collection features within the chamber walls. These modifications could reduce the likelihood of captured particles re-entering the airflow and subsequently clogging the filter.
The OpenSCAD files in this repository are highly parameterized. The filter should not be considered one 3D printable module, but rather a toolkit or “design kit” for making models of any size, number of coils, slit geometries, etc. It aims to make testing new printable designs as easy as possible. Of course, this requires some understanding of the OpenSCAD programming language, and reading the code to understand the configuration. This design is captured in the figure below.
Highly parameterized OpenSCAD helical filter design
At present, this file produces a “cartridge” with a tapered cylinder which is convenient for testing within a standard vacuum cleaner or shop vac hoses. Below is pictured a 3-bin, 6-turn cartridge. A “cap” can be produced from the same file so that the bins are all closed.
In future work, we need to modify our design to have much deeper bins. The OpenSCAD system minimizes the labor of producing such a design. Included below is a photo of the filter ready to be plugged into a shop vac .
3D printed and assembled filter ready for installation.
An early prototype looked like this:
Early prototypes which used barbed fittings.
We altered this approach in favor of a system that did not use barbs because otherwise it became difficult to control the particulates in the tube.
HEPA Filter Testing for Life Support Systems on Artemis Lunar Missions, accessed November 23, 2025, https://ttu-ir.tdl.org/bitstreams/01236268-2dc8-46ab-a3fe-6d9d5cbd9797/download