Mitigation of Turbulence in the Zygo Verifire™
Team Members
Madelyn Sabatini- Project Coordinator, Customer Liason
Steve Kim- Document Handler, Testing Coordinator
Nicholas Franco- Scribe, Manufacturing Coordinator
Advisors
Paul Mark- Zygo Customer Contact
Dr. Jim Zavislan- Faculty Advisor
Mike Pomerantz- Technical Advisor
Vision
The goal of this project is to find a working solution to reduce the effect of turbulence taken on the Zygo Verifire. Currently, turbulence causes variation between measurements, due to changing temperature and air flow. Even if nothing is moved between measurements, the data recorded will change from measurement to measurement as a direct result of said turbulence. The team will be testing the system: without any abatement techniques, with a standard polyvinyl chloride (PVC) tube covering the cavity, with fine mesh window screening surrounding the cavity, and finally an air flow straightener which consists of a combination of a thick honeycomb aluminum screen and fine mesh window screening. The final deliverable will be in the form of a working prototype as well as a paper presenting a thorough analysis of data collected. While we seek a solution that adequately reduces turbulence for any type of lens, testing will be primarily conducted on Thorlabs 50.8 mm diameter wedged flats.
Background
This project looks to mitigate turbulence in the measurements of optical flats using a Zygo Verifire HD. Turbulence comes from two sources: changes in pressure and changes in temperature. The change in pressure in the laser beam path of the Verifire causes non-laminar air flow, which in turn creates turbulent eddies that change the index of the air in the laser cavity, and changes in temperature again slightly change the index of refraction in the laser cavity. These changes in index affect the path the laser beam takes, causing the measurements taken to have higher amounts of error, meaning that multiple measurements must be taken and averaged to get a better representation of the wavefront error from the part being tested.
The Zygo Verifire HD is a Fizeau type interferometer. This type of interferometer uses phase modulation to measure the deviation of a part under test from a near perfect reference part (either a transmission flat or transmission sphere). This measures the surface of the part as well as the transmitted wavefront. The length of the laser cavity must be precisely modulated, meaning different parts under test will require different cavity lengths. The measurements are taken using a camera within the system to capture multiple images of the fringes. The Verifire then utilizes the proprietary Mx software, which allows the images to be analyzed. This software is highly customizable, such as allowing users to set up the system to automatically take and save a certain number of measurements in succession.
Testing Setup
To account for repeatability, a specific testing setup was used. This setup started by placing the optical flat being testing 0.5 meters away from the transmission flat. The optical flat was then aligned to show around less than one fringe of error. Once the optical flat and transmission flat were aligned, MetroPro’s autosequence tool was used to take 36 measurements 1 second apart, allowing for a “time lapse” effect to more accurately see how eddies move through the laser cavity as well as aiding in keeping the conditions as consistent as possible. A ten second delay was put before the measurements were taken to allow testers time to move away from the system as to not contribute any additional turbulence to the system. Three parts in total were tested, with each side of the part having 36 measurements taken
Proposed Solutions
Solution 1: PVC Pipe
The first solution under test utilizes a 6-inch diameter PVC pipe. The pipe will be placed over the laser cavity, with a mount printed to line up the center of the pipe with the center of the HeNe laser of the Verifire.
The primary purpose of the PVC is to have the rigid structure of the tube block out any turbulence caused from fluctuations in air pressure from entering the laser cavity while having the smooth interior of the tube create a more laminar flow of air within the cavity. The low thermal conductivity of PVC of around 0.19 W/(m*K) also should help keep fluctuations of temperature from entering the laser cavity, as a low thermal conductivity indicates that the tube should not hold on to any heat from the outside environment.
The possible challenge with this solution is that, should there be any large amounts of turbulence that occur as the pipe is being placed, the turbulence may not be able to escape the inside of the tubing, therefore extra time must be taken before measurements start to allow the potential turbulence to dissipate within the tube after it is placed in the laser cavity
Solution 2: Window Screen
The second solution under test utilizes a standard aluminum window screen. This method will follow the same form as the PVC pipe, as it will consist of layers of window screen wrapped into a tube which will isolate the cavity.
Rather than fully insulating the cavity like the PVC, the window screen aims to diffuse any incoming sources of turbulence quickly into equilibrium. The aluminum has a high thermal conductivity compared to the PVC material, allowing more heat to be absorbed rather than entering the system.
A challenge of this solution is that the window screen tube is very lightweight, so it is not durable and can be easily deformed or knocked out of place.
Solution 3: Honeycomb Air Flow Straightener
Air flow straighteners are commonly employed in wind tunnel engineering applications where downstream turbulent flow is undesired. In principle, air flow straighteners are typically comprised of a length-wise thick honeycomb material that works by reducing the transverse components of velocity fluctuations (turbulence). Honeycombs are often used in conjunction with fine mesh screens that have open-area-ratios (the ratio between the area that fluid can pass through and the 2D area occupied by the screen mesh) higher than 0.57 to reduce the axial component of velocity fluctuations.
Selection of the honeycomb depends largely on two parameters: the length-to-diameter aspect ratio and the cross-sectional shape.
In wind tunnel applications, honeycomb cells with length-to-diameter aspect ratios of around 8-12 were found to be the most effective at reducing the turbulence created downstream of subsonic wind turbines. For our application, the expectation is that turbulence arises largely from HVAC systems with air flow characteristic of most other work environments. A study found that the mean wind speed measured by a static anemometer for 55 work areas within 27 factories (that encompassed a range of environments from office spaces to heavy industrial engineering) was found to be 0.3 m/s. This is much smaller than the mean speed characteristic of wind tunnels and thus a much smaller aspect ratio could be considered.
In consideration of the desired properties of the honeycomb layer as well as economic and marked availability, we have chosen aluminum hexagonal honeycomb screens from Performance MRP which are typically used for mass air flow sensors in combustion engines as the ideal candidate honeycomb structure. They are supplied in 12” x 12” rectangular sheets with ¼” cell sizes and are ½” thick lengthwise.
Results
PVC and Window Screen
To measure the effectiveness of each solution, the 36 measurements taken for each side under each method were averaged using the Mx software. This yielded a surface map which serves as an approximation for the surface with no effects of turbulence. Each measurement for that set of conditions was then subtracted from the ideal surface. All 36 differences for each set were averaged to give the deviation of the measurements from the idealized surface map due to turbulence, ΔZ. As seen in the figure below, both methods effectively reduced turbulence, with the PVC generally outperforming the window screen.
Future Work
Because of time constraints, the honeycomb air flow straightener results were not able to be fully analyzed. In the future, this method should be fully tested if the solution is deemed as viable. From “mini tests” which were performed for proof of concept, this solution proved effective, but thus far there is no quantitative data on how this solution compares to the window screen or the PVC.