Citation
Calibration of Microelectronic Five-Hole Probes

Material Information

Title:
Calibration of Microelectronic Five-Hole Probes
Creator:
Gerding, Michael Dylan
Publication Date:
Language:
English

Notes

Abstract:
A calibration rig was designed for a microelectronic machine (MEM) based five-hole probe. This probe will give a unique solution to characterizing unknown flows in uncontrolled environments. The calibration rig was to step through a series of angular positions ranging from positive to negative ninety degrees in either direction for both rotational axes. The probe tip is to be placed at the center of the two rotational axes. Two Physik Instrumente rotation stages were utilized as it was desired to have small angular steps of approximately 0.1 to 0.2 degrees for angular positions less than five degrees. The L-611ASD was selected as the cone angle rotational stage due to its minimal stepping capability of 0.7 μrad. The DT-80 was selected for the roll rotational stage as it had minimalistic mass and needed to be supported by a less rigid portion of the rig structure. The deflection of the probe tip resulting from the stiffness of the DT-80's stage was calculated to be 0.0005 degrees, which is negligible. The L-611ASD has uncertainties resulting in position error less than one percent whereas the DT-80 may have an error on the order of magnitude of 10% depending on backlash. ( en )
General Note:
Awarded Bachelor of Science in Aerospace Engineering, summa cum laude, Major: Mechanical Engineering and Bachelor of Science in Mechanical Engineering, cum laude, Major: Mechanical Engineering on May 8, 2018.
General Note:
College or School: College of Engineering
General Note:
Advisor: Mark Sheplak. Advisor Department or School: MAE

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Michael Dylan Gerding. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.

UFDC Membership

Aggregations:
UF Undergraduate Honors Theses

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

Calibration of Microelectronic Five Hole Probes Michael D. Gerding Undergraduate Honors Thesis Summa Cum Laude B.S. in Aerospace Engineering Interdisciplinary Microsystems Group Spring 2018

PAGE 2

pg. 2 Acknowledgements I would like to thank Dr. Mark Sheplak and Haocheng Zhou for guiding me through this design process. I have learned a lot from my time working in the lab. I would also like to thank David Mills as well as all of the other IC2 members who spent time helping me improve my design during a team review. I wish to thank all of my IMG lab mates for providing a friendly, constructive learning environment. Finally, I would also like to thank all the members of Boei ng who spent time aiding with my design. Abstract A calibration rig was designed for a microelectronic machine (MEM) based five hole probe. This probe will give a unique solution to characterizing unknown flows in uncontrolled environments. Five hole pr obes are very sturdy pieces of equipment that have been used on mechanisms such as aircraft to calculate both the speed and angularity of the flow along the aircraft through the measurement of the pressures at a number of port s at the tip of the probe. Tra ditionally five hole probes have a time delay from when pressures are experienced to when pressures are recorded due to long pneumatic tubes connect ing the probe to a pressure transducer. A MEM based five hole probe eliminates this delay. Previous MEM base d five hole probes have not utilized optical pressure transducers, which allow the port pressures to be measured without compromising the accuracy of the measurements. The calibration rig was to be ran at low Reynolds numbers and step through a series of a ngular positions ranging from positive to negative ninety degrees in either direction for both rotational axes. The probe tip is to be placed at the center of the two rotational axes. Two Physik Instrumente ( PI ) rotation stages were utilized as it was desi red to have small angular steps of approximately 0.1 to 0.2 degrees for angular positions less than five degrees. As such a roll and cone angle coordinate system was utilized for the rig as opposed to a pitch and yaw coordinate system so the torque requir ements could be minimized, allowing the use of extremely precise and delicate rotation stages. The L 611ASD was selected as the cone angle rotational stage due to its minimal stepping capability of 0.7 rad. The DT 80 was selected for the roll rotational s tage as it had a mass of 0.8kg (compared to the 2.6kg of the L 611ASD) and needed to be supported by a less rigid portion of the rig structure. A calibration rig was designed around these two rotational stages to place the tip of the probe in the middle of an eight inch by eight inch wind tunnel exit area. The rig was designed to allow the adjustment of leveling feet to better position the probe tip. The deflection of the probe tip resulting from the stiffness of the DT stage was calculated to be 0.00 0 5 degrees, which is an order of magnitude smaller than the innate manufacturing uncertainties present in the product. The L 611ASD has small enough uncertainties to the point that any resulting position error will be less than one percent whereas the DT 80 may have an error on the order of magnitude of 10% depending on the effects of the stage s backlash. As such the DT 80 may be unable to meet the 0.1 to 0.2 degree step specification reliably. Once the motors are received from the supplier in Europe the ac curacy of the stage can be verified experimentally.

PAGE 3

pg. 3 Contents Acknowledgements ................................ ................................ ................................ ................................ ....... 2 Abstract ................................ ................................ ................................ ................................ ......................... 2 Introduction and Lit erature Review ................................ ................................ ................................ .............. 5 Design Development ................................ ................................ ................................ ................................ ..... 9 Expected Performance ................................ ................................ ................................ ................................ 15 Moving Forward ................................ ................................ ................................ ................................ .......... 18 References ................................ ................................ ................................ ................................ .................. 20 Appendix ................................ ................................ ................................ ................................ ..................... 21 Table of Figures Figure 1: Five Hole Probe Layout [3] ................................ ................................ ................................ .......... 6 Figure 2: R oll and Cone Angle Rig [3] ................................ ................................ ................................ ........... 6 Figure 3: Pitch and Yaw Rig [3] ................................ ................................ ................................ .................... 7 Figure 4: University of Florida's Open Return Wind Tunnel ................................ ................................ ....... 8 Figure 5: L 611ASD (left) and DT 80 (right) [6] ................................ ................................ .......................... 10 Figure 6: An Experimental Pitch/Yaw Configuration [2] ................................ ................................ ........... 11 Figure 7: "Arced" Rig Configuration [1] ................................ ................................ ................................ ..... 11 Figure 8: Calibration Rig ................................ ................................ ................................ ............................. 12 Figure 9: Top of Calibration Rig ................................ ................................ ................................ .................. 13 Figure 10: Parallel Shaft Coupling ................................ ................................ ................................ .............. 23 Figure 11: Shaft Collar ................................ ................................ ................................ ................................ 24 Figure 12: Horizontal Arm ................................ ................................ ................................ .......................... 25 Fi gure 13: DT 80 Coupling Plate ................................ ................................ ................................ ................. 26 Figure 14: Vertical Arm ................................ ................................ ................................ ............................... 27 Figure 15: Table Top Sheet 1 ................................ ................................ ................................ ...................... 28 Figure 16: Table Top Sheet 2 ................................ ................................ ................................ ...................... 29 Figure 17: L 611 Coupling Plate ................................ ................................ ................................ .................. 30 Figure 18: 2020x52.95in ................................ ................................ ................................ ............................. 31 Figure 19: L 611 Table ................................ ................................ ................................ ................................ 32 Figure 20: L 611 Shaft ................................ ................................ ................................ ................................ 33 Figure 21: Support Base ................................ ................................ ................................ ............................. 34

PAGE 4

pg. 4 Table of Tables Table 1: PI Rotation Stages [6] ................................ ................................ ................................ ................... 10 Table 2: PI Stages Base Specifications [6] ................................ ................................ ................................ .. 16 Table 3: PI Stages Uncertainty Specifications [6] ................................ ................................ ...................... 16 Table 4: DT 80 Calculations ................................ ................................ ................................ ........................ 17 Table 5: Bill of Materials ................................ ................................ ................................ ............................ 21

PAGE 5

pg. 5 Introduction and Literature Review This paper will detail the design of a calibration rig to be used in tandem with a microelectronic machine (MEM S ) based five hole probe. A five hole probe has one central hole and four evenly spaced periphery holes pres ent on its tip. Figure 1 shows a visual representation of this layout. Note that the probe being development by IMG will have a hemispherical tip as opposed to a coni cal tip. The probe is connected to a pressure transducer which will read the pressure at each of the holes. For a given probe orientation relative to the flow and a given speed of the flow, the pressures of the five holes relative to one another should be consistent. For instance, i f the probe is pointed directly into the flow, the central hole should have the largest pressure reading while top and bottom holes have equal pressure readings. Moreover, If the bottom hole is perpendicular to the flow it will t hen have a larger pressure reading than the center hole, which in turn will have a larger pressure reading than the top hole. However, for a probe to be able to determine its angle relative to a flow and the velocity of said flow, it needs a calibration da tabase of example flows at a variety of angles and velocities. At each calibration condition, non dimensional coefficients of various pressure differences can be recorded and stored for reference in unknown flow con ditions [3 ] To perform such calibration a rig is needed to step the probe tip through a variety of angles relative to the variable flow speed of a wind tunnel. To do so, the tip of the probe needs to remain at the same point in space. This is done by placing it at the intersection of two axis o f rotation. Two coordinate systems can be used in this process; pitch and yaw or roll and cone angle. Either of these coordinate systems can be used effectively and then converted using explicit geometrical rel ations when convenient [3 ]. Figure 1 shows a five hole probe while label ing rotation ax e s. Note that the actual five hole probe will have much smaller holes. Figure 2 gives an example of a roll and cone angle rig whereas Figure 3 gives an example of a pitch and yaw rig Note that there are no studies describing which calibration rigs perform optimally and provide the best positioning accuracy. As such the design presented in this report is an attempt to create a simple yet accurate calibration rig by looking at examples existing in the real world. This design will be automated, which will make it inherently more precise and time efficient than manually operated calibration rigs.

PAGE 6

pg. 6 Figure 1 : Five Hole Probe Layout [3 ]. Note that the nomenclature is not used in this paper. Figure 2 : Roll and Cone Angle Rig [3 ] Note that the nomenclature is not used in this paper.

PAGE 7

pg. 7 Figure 3 : Pitch and Yaw Rig [3 ] The purpose of developing a MEM based five hole probe is to increase the accuracy of readings with respect to time. In traditional probes, air must tr avel through long pneumatic tubes to reach sensing equipment which can take a significant amount of time when compared to the amount of time taken for the flow to change. A MEM based five hole probe removes this lag from the process through the use of optical pressure transducers embedded in the tip of the probe Many past attempts to develop a five hole probe with a better frequency response have not resulted in the manufacture of a functioning MEM based five hole probe. One MEM bas ed probe was developed in the past; however, it did not utilize optical pressure transducers. The piezo resistors that were utilized as an alternative caused high heat generation which altered the flow field around the probe compromising the results. Optic al pressure transducers will not cause such a problem. Five hole probes have been used on aircraft, rockets, and in the calibration of wind tunnels Boeing uses five hole probes to measure the wake behind a model aircraft in a wind tunnel. This data can be used to calculate accurate drag and lift data [8 ] The re are several other flow measurement techniques which multi hole probes compete. One of these is hot wire anemometry (HWA). This technique places a heated up wire in the flow and uses the convection heat transfer to measure the flow. Three such wires would be need ed to make three dimensional measurements. However, the small size of the wires makes them very vulnerable to breaking when compared to the five hole probe. Laser Doppler Velocimetry (LDV) us es the doppler effect of lasers to characterize the flow when pointed at particles seeded in the flow While this method is non intrusive and does not place a solid object in the flow, this method requires a transparent flow field and could only operate in research facilities Additionally, LDV requires a relatively expensive and a time consuming set up process. Particle Imaging Velocimetry (PIV) is another nonintrusive flow characterization technique. In facilities the flow is seeded with particles and pic tures can be taken as the particles move downstream allowing velocity measurements to be taken. Again, this method requires transparent fluid and can only take place in a facility. Note that only multi hole probe s also provide

PAGE 8

pg. 8 pressure information, allowin g for more than flow kinematics to be studied. Only the multi hole probe is suitable for a flight test. Vanes can be used to measure flow angle on flight vehicles, but not the velocity. Additionally, vanes are not mechanically robust and are prone to break ing in unfavorable w eather conditions [8 ]. This project is being mentored by both Boeing and the National Aeronautics and Space Administration ( NASA ), both of whom have required specifications for the five hole probe and calibration rig NASA differentiates small flow angles from large flow angles as being less than five degrees. For small angles NASA requires angular steps of 0.1 to 0.2 degrees from the calibration rig. For large angles, NASA requires angular steps of one to two degrees from the calibration rig. The probe is to be calibrated for angles with the bounds of plus or minus ninety degrees for both axes of rotation. This is assumed to yield a five hole probe with an angular error of no more than approximately five percent of the flow angle being measured However, the purpose of this paper is to discuss the design of the calibration rig and not the performance of the five hole probe. However, the probe will rely heavily on the calibration process to remain accurate. Boeing simply requ ires angular steps of one to two degrees for all angles. As such, the NASA specifications were pursued to the best of ability assuming that the Boeing specifications would be met regardless. It was desired to be able to control the rig using national instr The initial phase of experiments on the five hole probe will be at low Reynolds numbers using Figure 4 The opening is an eight inch square cross section Figure 4 : University of Florida's Open Return Wind Tunnel

PAGE 9

pg. 9 Design Development The design was developed around the rotational stages that specification s as they are the most crucial component of the calibration rig. OTS rotational stages were considered to be more precise than a custom made design due to gear trains not being perfectly efficient Additiona lly, it was desired to control the calibration rig through the use of LabVIEW. Physik Instrumente (PI) is a unique supplier as it sells rotation stages that are accurate to the scale of rad. Additionally, it was ascertained that they supply a LabVIEW libr ary for use with their motors which would make any programming task simpler in the future. Since the stages are accurate to such a small increment of motion, they naturally have relative low holding torques when compares to less accurate angular positionin g devices. As such a pitch and yaw coordinate system was ruled out as a possibility for the calibration rig, as it will naturally place a larger torque on the pitch rotational stage than would be present if it were a roll rotational stage (see Figure 6 and Figure 9 ) [2 ] Note that the probe being manufactured will not be a hooked p robe, thus reinforcing that idea that a pitch and yaw coordinate system is unobtainable. s that stepper motors would be a natural fit for the application. This would allow the rig to take an incremental ste p in angular position and pause to take data sequentially for a long period of time. All PI rotational stages that could be driven by a stepper motor either had an unlimited travel range or was limited to ten degrees in either direction. Since ninety degre es of motion is required in either direction, only the stages with unbounded ranges were considered. Table 1 details the remaining rotation stages (NS mean s not specified) Of the stages available the L 611ASD had the second smallest minimum incremental motion of 0.35 rad. Note that PI defines mini mum incremental motion as the smallest movement that a stage can execute repeatedly. As will be detailed in the expected performance section, this minimum incremental motion is adequate for the specifications. The PRS 200 was not selected for its minimal m otion as this level of precision will unnecessarily increase expenses. Furthermore, the stages are relatively expensive, costing tens of thousands of dollars. As such the L 611ASD was selected as the cone angle rotation stage. The L 611ASD was not selected for the roll rotational stage as it has a significantly higher mass which will put more strain on the cone angle rotational stage and will cause an increased amount of bending in the rig structural components. The DT 34 had the highest minimum incremental motion so it was not considered adequate. 350 rad is approximately 0.02 degrees. This is ten percent of the maximum step sized desired for the calibration rig at small angles. Thus, the inherent uncertainties in the DT 34 will have a larger effect on the positioning than the other stages. Both the DT 80 and RS 40 are less than a kilogram in mass and are considered lightweight. The DT 80 is considered superior to the RS 40 despite having twice as much mass due to its far lower backlash error. Backlash erro r is defined as the error incurred when switching directions. However, it is also characteristic of the error incurred when stopped at a new angular position. As will be discussed in the expected performance section, the DT 80 might not be accurate enough The DT 80 was justified in being selected as it was available in the expected performance section. The DT 80 was selected as the roll rotation stage [6 ] Figure 5 shows the selected motors.

PAGE 10

pg. 10 Table 1 : PI Rotation Stages [ 6 ] Motor L 611ASD DT 80 RS 40 PRS 200 DT 34 Minimum Motion 0.7 70 87 0.5 350 Mass (kg) 2.6 0.8 0.4 8 0.15 0 350 1500 0 NS Figure 5 : L 611ASD (left) and DT 80 (right) [ 6 ] Note that the design was restricted by the use of a large wind tunnel as opposed to a smaller nozzle (which many other experiments have favorited as in Figure 6 ) [2 ]. The torque restrictions of the motors also acted as a design constraint as mentioned before. Thus a curved design as in Figure 7 is not realistic [1 ]

PAGE 11

pg. 11 Figure 6 : An Experimenta l Pitch/Yaw Configuration [2 ] Figure 7 : "Arced" Rig Configuration [1 ] To provide visual context for the upcoming discussion, Figure 8 is provided as a view of the designed calibration rig.

PAGE 12

pg. 12 Figure 8 : Calibration Rig The wind tunnel to be used for this project has the bottom of its opening at almost five feet off of the ground. As such a support structure must hold the major rig component up off the ground. 2020 support legs supplied by 80/20 are used to connect all rotating components to the base of the support. These 80/20 legs are connected to a stainless steel plate that is half an inch thick with a two foot, square cross section. This relatively thick plate was made out of steel rather than aluminum to lower the center of mass of th e rig. This prevent s the rig from tipping over and the expensive rotation stages from being damaged. This concept was used as opposed to an entirely 2020 support system as it would be more expensive. Leveling feet are included with a four inch threaded len gth to help make the probe tip perpendicular to an incoming flow. The leveling feet can be adjusted to make the central hole of the pressure probe have the max pressure reading while the pressure readings of the periphery holes are symmetrical per the roll angle of the probe. The leveling feet are rated for 250 lbs each (1000 lbs total) [4 ]. The probe was to be positioned in the center of the wind tunnel opening. The center of the support base plate was placed at the center of the four inch threaded length of the leveling feet and the 2020 support legs were then dimensioned accordingly to do so. The tip of the probe is just under seven inches above the highest point of the 2020 legs, making the legs just under 53 inches in length. This gives the rig equal distances of play in either direction for each foot. For a discussion of the top of the c alibration rig, Figure 9 was provided as a visual reference Note that fasteners are not included in any figures due to time constraints although they have been specified in Table 5 2020 Support Legs Support Base Plate Leveling Feet

PAGE 13

pg. 13 Figure 9 : Top of Calibration Rig Note that the probe is modeled as a three inch long cylinder for simplicity. At the time of the writing of the report, the probe has yet to be fully manufactured. The probe is to have a hemispherical tip with a diameter of five millimeters. As such an off the shelf (OTS) five millimeter aluminum shaft collar was used to hold the probe in place. Probes often increase in thickness after a few inches of distance from their tip. As such a different collar may be needed in the future. An OTS shaft collar was utilized rat her than a custom design to ensure a tight grip around the circumference of the probe. Holes had to be made in the collar for it to be fastened to two shafts approximately 3.6 inches long. These shafts might also vary in length depending on the final geome try of the probe and the location along the probe where it is to be gripped. Typically, probes are gripped near the tip to prevent any vibrations while measuring the flow. However, this project will start by testing the probe at low Reynolds numbers, thus minimizing the potential for vibration in the first place. In the current configuration, the probe is being gripped an inch away from the tip Regardless of where the probe is gripped, the probe tip must be at the center of the axis of rotation of the L 61 1. Minimizing the effects of components other than the probe altering the flow is also a priority. As such as many components as possible are desired to be kept out of the flow or at least kept at a distance behind the probe tip. The widest five millimeter shaft collar was selected so that the two 0.25 in shafts have as much clearance as possible with the probe (due to an unknown final geometry). This width was 25 mm, or slightly less than an inch [7 ] The DT 80 has an aperture size of forty millimeters and was not a constraining factor for this consideration. The shafts also benefit from being longer rather than shorter by allowing a larger radius of curvature for any cords exiting the rear of the probe. The two shafts are tapped on each end for a 5 40 fastener. These shafts can be order ed with one side already tapped. Thus, they can be cut to length and given a matching tap on the cut end. This will help minimize manufacturing costs. The shafts are compo sed of 1045 Steel, which will help prevent the shafts from deforming under weight. Both shafts are fastened to a coupling plate made of 6061 A luminum which mates to the pattern of threaded holes on the surface of the DT 80. The probe is positioned to be pr ecisely at the center of the DT aperture The DT 80 is fastened to the vertical arm using both of its tapped holes as well as its bolt holes. The vertical arm is 0.25 in thick, three inches wide (to be able to mate to the DT 80), and 7.25 inches tall Probe Placeholder Vertical Arm DT 80 Coupling Plate Horizontal Arm L 611 Support Table L 611 Coupling Plate Table Top Turn Table

PAGE 14

pg. 14 while being made out of 6061 A luminum which is cheaper than stainless steel The height of the vertical arm was made to both give enough space to mate to the DT 80, as well as place the probe tip at the center of the wind tunnel opening while keeping the r emainder of the rig out of the flow entirely. The probe tip is approximately six inches above the bottom most face of the vertical arm. This distance only needs to be four inches to keep the remainder of the rig out of the flow while keeping the probe at t he center. Material could be removed from the vertical arm to make it more aerodynamic if deemed necessary in the future; however, this would increase manufacturing costs. The vertical arm is fastened to the horizontal arm at its bottom using a n OTS eight hole, gusseted angle bracket supplied by 80/20. The gusset will make the joint more rigid so that the position of the probe tip is not as affected by the bending of structural components. The horizontal arm is 0.25 in thick, 2.5 in wide, and thirteen inche s long while being made out of 6061 Aluminum The width of the horizontal arm had to be at least two inches to mate with the angle bracket yet was made 2.5 in to help improve rigidity as well as leave less of the vertical arm hanging over its edges as the exact center of mass of the DT 80 is unknown. If torsion of the horizontal arm is determined to be a problem, the cross section can be altered. The connect ion between the two arms uses fasteners which were verified to fit regardless of the space constrain t of the nearby table top. The horizontal arm is connected to a shaft which is also fastened to the L 611 coupling plate, which serves a similar purpose to the DT 80 coupling plate. This shaft is produced in the same manner to those connecting to the DT 80 coupling plate. Both rotational stage coupling plates are made out of 6061 A luminum to decrease cost. They are also not expected to experience very high forces. The horizontal arm is connected to the shaft at the center of its length. It extends equally f ar in both directions so that an appropriate counterweight can be placed to generate a similar moment to all the components connected to the upper half of the vertical arm. This counterweight is to be designed after the rig is manufactured initially so tha t an appropriate position of the center of mass can be determined experimentally. This would minimize any moments on the turntable. The horizontal arm is connected in two places to the 6.18 in bolt circle of the OTS lubricated steel turntable supplied by M cMaster Carr. The turntable is rated for heavy machinery so the amount of weight placed on it should not be an issue [4 ]. If the bending of the horizontal arm is determined to significantly impact the position of the probe, it could be made out of steel ma king it more rigid. Such a determination would be made experimentally after the rig is built. The horizontal arm may need to be made longer in the future to provide more space for the geometry and cables of the probe. Note that at the time of the writing of this report, the PI rotational stages have yet to be shipped from Europe. As such, any center of mass determinations have not yet been made. The table top is a one foot square plate with a thickness of 0.25 in which connects the turn table to the 2020 l egs This too is made out of 6061 Aluminum. A hole was made for the L 611 shaft to pass through the table top. The table top is connected to the 2020 legs using the same angle bracket as was used on the arms. This same bracket was utilized to connect the 2 020 legs to the support base plate. This helps improve the rigidity of the structure. A separate eight inch by one foot by 0.25 in plate made of 6061 Aluminum was used to support the L 611. OTS f our hole wide, gusseted angle brackets were used in this case to ensure the bolt pattern would not interfere with the positioning of the L 611. All fastener clearance holes above the top surface of the turn table as well as the connection between the L 611 coupling plat e and the L 611 shaft were made to be close fit clearance holes. This helps ensure the accurate positioning of the probe tip at the center of both axes of rotation All other

PAGE 15

pg. 15 fastener clearance holes were dimensioned to be free fit clearance hole. As such, the L 611 will be less likely to have torque placed on it as a result of manufacturing imperfections within the tolerances of the designed parts. Fasteners were selected so when they were used with a nut, the fastener were always long enough to reach thro ugh the components it was connecting as well as through the entirety of the nut. When fasteners were to be threaded into a hole, it was made sure that the fasteners would always have at least three fully engaged threads. Expected Performance This section locate the probe tip at the center of two axis of rotation as well as incrementally change its angular position about said point. This will be done entirely using motor specifications and estimated weights of designed parts. Manufacturing tolerances will not be considered as this is best done experimentally which would require the rig to have been assembled, which was not possible at the time this report was written as th e rotational stages had yet to be delivered Table 2 and Table 3 summarize the rotational stage base specifications and uncertainty specifications respectively. In Table 2 the cost is only of the rotational stage, not any controller or cables required. The minimum motion is that which is capable of being executed repeatedly through the use of a controller that was purchased whereas the step size is that of the stage. to that which is perpendicular to the axis of rotation. The design resolution is the theoretical minimum mo vement of the stage. In Table 3 tilt stiffness characterizes the tilting of the axis of rotation downwards under weight. Axis deviation is the possible linear distance the axis of rotation can be from the center of the rotational stage The repeatability is defined as how close the stage can return to a given position after a deviation from that position. The DT the same d irection whereas the L Returning to the same position from the same direction is negligibly affected by backlash error. As such, backlash error is considered more important as the stages wi ll have to stop rotating frequently to record data. The backlash error is allegedly eliminated from the L 611 through the use of an integrated angle measuring system [6 ]. This allows the exact position to be measured regardless of backlash.

PAGE 16

pg. 16 Table 2 : PI Stages Base Specifications [ 6 ] L 611 DT 80 Total Cost ($) 8 914 2 744 Min imum Motion (deg) 0.00004 0.004 Permissible Torque (Nm) 3 0.1 Mass (kg) 2.6 0.8 Axial Load Capacity (N) 100 20 Tilt Load C a pacity (N) 10 Permissible Tilt Torque (Nm) 40 5 Step Size (deg) 0.02 0.01 Design Resolution 2.00E 05 0.01 Table 3 : PI Stages Uncertainty Specifications [ 6 ] L 611 DT 80 Tilt Siffness deg/Nm 0.001719 0.00859 Wobble (tilt about rotation axis) (deg) 0.000859 0.006 1 30 2.5 30 Repeatability (accuracy when returning) (deg) 0.0002 0.01 Backlash (deg) 0 0.02 To estimate the performance of the calibration rig as well as confirm that the DT 80 could withstand the weight of the probe clamping parts, weight and moment calculations were required. The parts included in this calculation were the DT 80 Coupling Plate, the Parallel Shaft Couplings, the Ruland 5mm aluminum shaft collar, and an est imated probe shape. The probe was approximated as a three inch long solid cylinder All parts were considered solid shapes and fastener were neglected. Despite the parts being various steels and aluminums all were approximated as Stainless Steel with a de nsity of 8000kg/m 3 as an overestimation [5 ] This value provides an overestimation regardless of fasteners being neglected and the change in materials. Note that the Parallel Shaft Couplings are not of definite length as the probe geometry was not certain at the time of this calculation. For the purpose of this calculation, they were estimated to be approximately 3.6 inches in length, so that the probe i s gripped an inch away from its tip. All measurements were rounded up so an overestimate would be acquired. Equation (1) details how the weight of each part was calculated whereas (2) shows how the moment arm for each part was calculated. Finally, (3) appr oximates the moment of each part. is the overestimated density, A is the cross sectional area of the solid part, l is the length of the part along the rotational axis, m is the

PAGE 17

pg. 17 minimum distance from the part in question to the rotational stage, W is the weight of a part, M is the moment of a part g is the acceleration due to gravity (9.81m/s 2 ) and L is the moment arm of a part Note that these calculations assume that the rotational axis is perfectly horizontal and that each part is symmetrical about th e axis of rotation. (1) (2) (3) Table 4 gives the results of the aforementioned calculations. Table 4 : DT 80 Calculations Part L (m) W (N) M (Nm) Probe 0.0910 0.1174 0.01068 Shaft Collar 0.0943 0.231 0.021 8 Parallel Shafts 0.0514 0.44 8 0.0230 Coupling Plate 0.0031 8 1.286 0.00408 Sums 2.08 2 0.059 6 It can be seen that the total weight which the DT 80 is experiencing is less than 10N and the moment by the stiffness given in Table B, the DT 80 axis of rotation will sag at most 0.00 05 degrees about the horizontal. Note that since the parts are design to be symmetrical about the axis of rotation, the 0.1Nm of permissible torque about the axis of rotation should not be reached. Furthermore, the L 611 is considered to operate at its specifications assuming the turntable is properly lubricated. Additionally, all moving parts must be accelerated slow enough so that the resulting inertial forces do not overpower the rotational stages. As such, an angular position error of less than one percent is expected at steps of 0.1 degrees ( from the repeatability specification ) The DT 80 can expect an angular position error of approximately five to fifteen percent depending on the effects of the backlash error and repeatability error while operating with steps of 0.2 degrees given its specifications in Table B The exact error will likely be less than fifteen percent as the DT 80 will not be required to co mpletely reverse its direction of rotation. The DT 80 is only required to stop rotating for data collection. This affect can be minimized by having the L 611 rotate as much as possible in lieu of the DT 80. In other words for each position of the DT 80, th e L 611 will rotate through all of its positions. As such the DT 80 may be unable to meet NASA specifications at small angles. This would need to be

PAGE 18

pg. 18 verified by experimentation. However; using a second L 611 might cause other issues with accuracy, such as the rig structure bending more under increased weight. Additionally, this will significantly increase the cost of the rig unnecessarily at such an early stage of the probe design. The effect of the 0.00 05 degrees of tilt due to weight simply adds to the ef fects of the tolerances given in its specifications found in Table B (which are more significant) The effect of the probe tip not being perfectly centered at the two axes of rotation and the axes being tilted are considered to be only verifiable by experi ment due to the complexity of the situation. The same is considered for any manufacturing imperfections. Note that an L 611ASD would also bend less than 0.0005 degrees under the weight it would experience as the roll rotation stage. Moving Forward At th e time this report was written, a bill of materials (BOM), part drawings, and an assembly file have been completed (see appendix) Future work would include adding all f asteners specified in the BOM to the assembly file and creating assembly drawings for t he manufacture of the calibration rig. Additionally, the assembly file can be used alongside a finite element analysis simulation to estimate the deflection of the probe tip. However, this would be best done once the exact weight and center of mass of the two rotation stages are known as well as the geometry of the probe Additionally, the effect of the leveling feet would to be difficult to account for in the simulation. Thus such a simulation is not a priority as it may be more time efficient to simply account for any deflections experimentally rather than computationally. The calibration LabVIEW code can also be written before the manufacture of the rig utilizing the PI LabV IEW library. This would see the stages incrementally stepped through different angular positions using a for loop with pauses for collecting data. The exact duration of the pauses can be determined once the rig is manufactured by measuring the settling tim e of preliminary data. Testing procedure may also be formally drafted. For instance despite the fact that the experiment will be run at a low Reynolds number, testing can be done at a variety of low Reynolds number to see if the calibration changes signif icantly throughout a given range. It was also postulated that a hot wire anemometer can be run at the same time as the calibrated probe. This would give another result which can be compared to that of the probe when an unknown flow is characterized. Note t hat the anemometer will only be able to measure the speed of the flow if only a single wire is utilized. This still can be used to build confidence in the operation of the probe. Once the rotation stages have been received, much more work can be completed A counterweight can be developed so that the turntable does not experience a moment Experiments can be devised to generate angular position uncertainties for the stages through observation rather than reference a data sheet. This will help to better ch aracterize the effects of the backlash error. If necessary A rotary encoder can be implemented behind the vertical arm to allow for closed loop control of the DT 80 and allowing backlash error to be directly observed. The L 611 already has a sensor for cl osed loop control. Additionally, the controllers specified in Table 5 are capable of closed loop control. How level the rig can hold a probe like object can also be ve rified. This can be done experimentally by clamping a laser pointer in the rig and using trigonometry to estimate any deviations while the rotation stages are being turned. The calibration rig could be surrounded by a sort of trifold to

PAGE 19

pg. 19 record laser point positions on. Note that the leveling feet can also be adjusted to make the rig as level as possible. This could be done in addition to verifying that the top and bottom ports of the probe read the same pressure in a flow field. This would mean the probe is level with the flow. Ro lling the probe 180 degrees and observing the same would verify the orientation of the probe.

PAGE 20

pg. 20 References Institute Qualifying Project Report, 2013, pp. 40. Miniature Five Hole Probe with Embedded Pressure Sensors for use in Extremely Confined and Complex Flow Areas in [3 ] Engineering Digital Collection, vol. 123, 2001; doi: 10.1115/1.1334377. Mc Master Carr Supply Company, Available: https://www.mcmaster.com/ https://www.engineeringtoolbox.com/metal alloys densities d_50.html Physik In strumente, Available: https://www.physikinstrumente.com/en/products/productfinder/ Ruland Manufacturing Company Inc., Available: https://www.ruland.com/shaft collars.html Doctor of Philosophy Proposal, 2017, pp.12 30.

PAGE 21

pg. 21 Appendix Note that parts closer to each other in Table 5 are closer to each in the physical rig assembly when possible. All figures of part drawings tolerance dimensions that have thre e decimal places with 0.005 in. Also note that unnecessary sheet format blocks have not been completed This is in part because assembly drawings have yet to be finalized Table 5 : Bill of Materials Sub Assembly Part Name Drawing # Quantity in Sub Assembly (Total) Possible Vendor Vendor Part # OTS? Probe Interface probe P3 1 no 5mm aluminum shaft collar P2 1 Ruland ENCL25 5MM A no, must be drilled M2.5x8 steel hex screw NA 1 Ruland ENCL25 5MM A yes, comes with collar parallel shaft coupling P1 2 McMaster Carr 8017T1 no, must cut and tap raw stock 5 40x0.5'' 316 SS BHS NA 4 (6) McMaster Carr 98164A438 yes DT 80 coupling plate P5 1 NA no M4 0.7x10 DIN912 SHS NA 4 McMaster Carr 91292A116 yes DT 80 rotation stage NA 1 PI 64439200 yes, comes with DT 80 M4 0.7 DIN 912 fastener NA 4 PI 64439200 yes, comes with DT 80 18 8 SS hex nut M4 0.7 NA 2 McMaster Carr 91828A231 yes vertical arm P6 1 NA no, must be drilled 10 series 8 hole al. gusseted corner bracket NA 1 (9) 80/20 inc 4138 yes 1/4 20x5/8'' SS BHS NA 8 (32) McMaster Carr 97763A264 yes 1/4 20 5/32'' hex nut NA 8 McMaster Carr 94805A029 yes horizontal arm P4 1 NA no, must be drilled Table to Interface Connection 5 40x0.5'' 316 SS BHS NA 1 (6) McMaster Carr 98164A438 yes 1/4 20x9/16'' BHS NA 2 McMaster Carr 91255A524 yes 1/4 20 t nut NA 2 (58) 80/20 inc 3382 yes Probe Table lubricated turntable NA 1 McMaster Carr 1544T2 yes 10 32x9/16'' BHS NA 4 McMaster Carr 91255A027 yes 10 32 SS hex nut NA 4 McMaster Carr 91841A195 yes table top T1 1 McMaster Carr 89015K137 no, must be drilled 2020x52.95'' T3 4 80/20 inc 2020 S maybe, can order to length or cut 10 series 8 hole al. gusseted corner bracket NA 4 (9) 80/20 inc 4138 yes

PAGE 22

pg. 22 Sub Assembly Part Name Drawing # Quantity in Sub Assembly (Total) Possible Vendor Vendor Part # OTS? 10 series 4 hole wide gusseted corner bracket NA 4 80/20 inc 4134 yes 1/4 20x5/8in SS BHS NA 24 (32) McMaster Carr 97763A264 yes 1/4 20x5/32in tall hex nut NA 24 (32) McMaster Carr 94805A029 yes 1/4 20x0.5'' BHCS NA 24 (40) 80/20 inc 3061 yes 1/4 20 t nut NA 24 (58) 80/20 inc 3382 yes L611 shaft emulator T5 1 McMaster Carr 8017T1 no, must cut raw stock 5 40x0.5'' 316 SS BHS NA 1 (6) McMaster Carr 98164A438 yes L611 coupling plate T2 1 McMaster Carr NA no M6 1x10mm SS BHS NA 4 McMaster Carr 97763A823 yes L611 rotation stage NA 1 PI L 611.9ASD yes, included with L 611 L611 table T4 1 McMaster Carr NA no M6 1x60mm SS BHS NA 3 McMaster Carr 92095A254 yes DIN 433 6 washers NA 3 PI L 611.9ASD yes, included with L 611 M6 1 SS hex nut NA 3 McMaster Carr 96621A110 yes Support/Rig interface 1/4 20x0.5'' BHCS NA 16 (40) 80/20 inc 3061 yes 1/4 20x7/8'' SS BHS NA 16 McMaster Carr 97763A346 yes 10 series 8 hole al. gusseted corner bracket NA 4 (9) 80/20 inc 4138 yes 1/4 20 t nut NA 32 (58) 80/20 inc 3382 yes Support 5/16 18x4'' swivel leveling feet NA 4 McMaster Carr 6111K457 yes 5/16 18 lock nut NA 4 McMaster Carr 6111K457 yes, comes with leveling feet support base S1 1 NA no Other orders stepper motor controller NA 2 PI C 663.12 yes stepper motor cable 3m NA 1 PI 720090000 yes L611 to controller converter NA 1 PI C 815.AC32 0300 yes

PAGE 23

pg. 23 Figure 10 : Parallel Shaft Coupling

PAGE 24

pg. 24 Figure 11 : Shaft Collar

PAGE 25

pg. 25 Figure 12 : Horizontal Arm

PAGE 26

pg. 26 Figure 13 : DT 80 Coupling Plate

PAGE 27

pg. 27 Figure 14 : Vertical Arm

PAGE 28

pg. 28 Figure 15 : Table Top Sheet 1

PAGE 29

pg. 29 Figure 16 : Table Top Sheet 2

PAGE 30

pg. 30 Figure 17 : L 611 Coupling Plate

PAGE 31

pg. 31 Figure 18 : 2020x52.95in

PAGE 32

pg. 32 Figure 19 : L 611 Table

PAGE 33

pg. 33 Figure 20 : L 611 Shaft

PAGE 34

pg. 34 Figure 21 : Support Base