Скачать 188.16 Kb.
Fatigue Life Tester for Electroformed Bellows
Dr. Beth DeBartolo
NiCoForm, Inc. is a Rochester based company specializing in electroforming for engineering applications. They have a variety of metals used for deposition including, NiColoy™, a proprietary nickel-cobalt alloy, copper, and gold. Typical applications include optical and nanofluidic molds, catheter-tipping dies, IR cold shields, waveguides, and bellows. The NiCoForm Design Team was asked to design a fatigue-life testing apparatus to be used to gather empirical fatigue-life data for a direct comparison to theoretical fatigue life calculations. This report will provide a complete explanation of the design considerations behind the bellows fatigue testing platform and electronics cabinet.
The NiCoForm Design Team used the Engineering Design Planner™ (EDP) to establish the course of action we were to pursue. The facets of the EDP follow a logical path for the realization of a quality engineering design. This report reflects the first six chapters of the process. Initially the needs and design criteria set forth by NiCoForm had to be addressed. This information is found in chapter 1 of the text. The second chapter enumerates three of the ideas that we initially came up with during the concept development stage of the design process. Included in the project notebook are bills of materials for each of the designs considered. Immediately following, we progress to the feasibility assessment in chapter 3 where we look at some of the pros and cons associated with each of the proposed design solutions. Hereafter we list all of the objectives, in chapter 4, that our design is to meet per the needs of the sponsor. The final design shall be measured against these criteria. In chapter 5, a discussion of the analysis, and certain aspects of the different parts, ensues. Chapter 6 outlines the methods that were used to fabricate the test platform. The following chapter covers the mechanical redesigns and any additional components that were needed to complete our goals. Finally, chapter 9 covers the conclusions the senior design team had and the current state of the design.
Table of Contents Page
4.0 Design and Performance Objectives 16
4.1 Design Objectives 16
4.2 Performance Objectives 16
5.0 Analysis of Problem 18
5.1 Actuator 18
5.1.1 Actuator Support Flexure 18
5.2 Electronics 20
5.2.1 PLC Expansion Base 20
5.2.2 PLC CPU Module 20
5.2.3 EZ-Text Operator Interface 21
5.2.4 BEI Motion Controller 21
5.2.5 Pressure Switch 21
5.2.6 Solenoid 22
5.3 Bellows Adapters 23
5.3.1 Bellows to Frame Adaptor 23
5.3.2 Bellows to Actuator Adaptor 24
5.3.3 Adaptor Conclusions 25
5.4 Testing Platform 26
5.4.1 Support Rods 26
5.4.2 Base plate 27
5.4.3 Tester Top 27
5.4.4 Actuator Flange 28
5.4.5 Isolator Feet 28
5.4.6 O-ring 28
5.5 Feedback Using Optical Position Sensor 29
5.6 Electronics Cabinet 29
6.0 Machining and Assembly 30
7.0 Redesign of Mechanical Components 31
8.0 Testing 32
9.0 Conclusions 33
Appendix A 34
Appendix B 35
Appendix C 36
List of Illustrations
1.1 Bellows Produced by NiCoForm, Inc.
1.2 Axial Deflection
1.3 Stress Distribution in Compression
1.4 System Flow Chart
5.1 Adapter in Compression
5.2 Stress Concentrations at 50G acceleration
1.0 Recognize and Quantify the Need
1.1 Project Mission
NiCoForm Inc. produces electroformed metal bellows (See Figure 1.1, below). They have expressed a desire to use experimental fatigue analysis to establish life cycle data for bellows made from their proprietary metal, NiColoy™. Due to the differing and complex geometries of each different bellows, it is crucial to have empirical data to support any theoretical calculations on fatigue life. The mission of the NiCoForm Design Team is to produce a working fatigue-life testing apparatus for use by NiCoForm Inc.
Figure 1.1 Bellows Produced by NiCoForm, Inc.
1.2 Product Description
The production of electroformed bellows, a machine component with many different applications, is limited to only a few companies. NiCoForm Inc. is one such company. The electroforming process, as compared to the other methods of manufacturing bellows, hydro forming and welding, has a huge advantage in that it leaves the bellows in a stress free state. While fatigue testing equipment has been around for many years (since the early 1900’s), fatigue life testers for bellows are virtually non-existent due to lack of a market for such a machine. For our sponsor to successfully market their electroformed bellows, they will need empirical data for the life cycle of their proprietary materials.
Metal bellows have a variety of applications such as: use in the transmission of torque, either axially or between two offset shafts; use as a spring; use as a pressure activated electrical switch. Another possibility is a combination of these uses. Our design of the testing platform will be restricted to applications, which require no torque in the bellows, in accordance our project scope.
The scope of our design is limited to testing axial deflection as shown in Figure 1.2. The design team predicts the area for failure will be in the radii of the convolutions due to the high stresses seen in these areas. The stress distribution is represented in Figure 1.3. The main concern is with the high number of deflection cycles the bellows will need to complete in order to reach failure. NiCoForm, Inc. predicts the number of cycles will be on the order of 107. For the mechanism to perform a large number of test runs, a robust design is in order.
Figure 1.2 Axial Deflection
The design team’s task is to come up with a scaleable, robust prototype design of a fatigue life tester. It will consist of a testing platform that can be duplicated in the future to run multiple tests simultaneously. Also required is an adaptable control console to allow for the addition of more testing platforms in the future with minimal cost and effort to NiCoForm.
Figure 1.2 Stress Distribution in Compression
Our design uses a voice coil linear actuator to displace the bellows. The bellows will be held securely at one end, while being attached to the actuator at the opposite end. The operator will input his/her desired data into the HMI (human-machine interface). Input data will include the part number and/or description, a predetermined displacement distance, followed by a drive frequency for the test. The operator will then start the test. The bellows will be internally pressurized as a means of determining failure, and the test will commence. During the test the CPU (Central Processing Unit) card in the PLC (Programmable Logic Controller) will run a continuous cycle counter. Upon detecting a pressure drop within the bellows the test will automatically stop and the final number of cycles recorded to the CPU. The above represents a description of the process as seen in Figure 1.4.
Figure 1.4 System Flow Chart
1.3 Scope Limitations
This project involves two parts, the preliminary design and the creation of an operational prototype. It will be completed entirely by the members of the senior design team. The preliminary design will be completed by the end of the 2003 fall quarter with the balance being fulfilled by the end of 2003 winter quarter. The preliminary design will consist of two deliverables, the preliminary design report and the technical data package. The work done by the end of winter quarter shall yield a technical paper, an operational website, and a functioning prototype.
1.4 Stake Holders
The primary stakeholders for this project are NiCoForm Inc. and the senior design team. NiCoForm will have a product that will allow them to explore the properties of their proprietary material, NiColoy™. The senior design team will have obtained the experience necessary to work on other design projects from conception to implementation. Secondary stakeholders include customers of NiCoForm who will receive a high-quality product produced via a capable testing platform.
1.5 Financial Analysis
The working budget of $5,000 will go toward the following items critical to the Fatigue Tester design project:
1.6 Market Possibilities
This project will produce a specialized machine to a client who is interested in gathering empirical data about their own materials. Considering there are less than ten producers of metal bellows in the United States and even fewer that manufacture electroformed bellows, the market for bellows fatigue testing equipment is virtually nonexistent.
1.7 Background Research
The design team has been charged with designing and creating a specialized piece of machinery for a company who will use it for the express purpose of internal data collection. While our specific application is unique to the electroformed bellows, a relatively new product, the fatigue life testing process has been around for some time. Fatigue life testing dates date back to the 1930’s and 40’s, when according to Darrell Socie, Professor of Mechanical Engineering at the University of Illinois at Urbana-Champaign, research “was largely devoted to experimentally establishing the effects of the many factors that influence the long-life fatigue strength. Tests were usually conducted in rotating bending and the life range of interest was about 106 cycles and greater.” 1
Although fatigue-testing equipment has been around for many years, the specific application of our design is reserved for the companies who produce metal bellows. Information on this type of test equipment is held in close regard and is not generally shared with people outside the company. This makes research into previous designs virtually impossible. Our research was necessarily limited to perusing fatigue and fracture books and machine design textbooks.
2.0 Concept Development
2.1 Cam Driven Concept
The cam driven design concept incorporates an AC electric motor driving a shaft with a cam to move the bellows in compression. The bellows would be held at one end and the opposing end would be connected to a slide with a spring return, which makes contact with the cam as the shaft is spun. We would be able to produce the high frequencies needed for fatigue life testing by the use of an off-the-shelf motor running at 1750 rpm, or by using a multi-lobed design for the cam. Multiple cam designs would be needed for differing displacement lengths.
2.2 Pneumatic Actuator Concept
Pneumatics has been used to apply linear motion in many applications. The senior design team’s concept would use a high frequency valve system to drive a double acting cylinder. The system would utilize a high volume air compressor to operate. The design is somewhat similar to the above-mentioned cam-driven concept. One end of the bellows will be locked in place and the opposite end is connected to the piston of the double acting cylinder.
2.3 Voice-coil Actuator Concept
Voice coil linear actuators are a new addition to the motion control market. This concept has the same basic configuration as the previous designs. The linear motion driver is a voice-coil that is connected directly to the bellows. This design allows for high testing frequencies and accurate motion control. A vendor supplied motion controller drives the voice coil via PLC.
3.0 Feasibility Assessment
3.1 Cam Driven Feasibility
The senior design team found enough reason to discard the cam driven concept; the major concern with the design being the number of cycles needed to perform a fatigue life test on a bellows. Due to this high number of cycles, in the range of 106 to 108 per test, the machine will encounter a large amount of wear of it’s mechanical parts. The bearings, bellows holder slide, and even the return spring will likely fail after only a few test runs. The need is for a low maintenance drive system that has a life expectancy in the 1011 cycles range. Another concern would be the cost of manufacturing the individual cams needed for the different test displacements. The use of lubricants is of high concern in the electroplating field. A cam driven design will need oil or grease to perform at maximum efficiency. This would be unacceptable at the NiCoForm facility because it would adversely affect the plating processes taking place nearby.
3.2 Pneumatic Driven Feasibility
The pneumatic concept was declined for similar reasons discussed above. The longevity issues are a concern. A pneumatic system would need high-speed solenoid valves that have a finite life far below the design team’s vision of what is realistic. There is also the prohibitive cost of investing in a high volume air compressor. Noise is also a concern given the limited space that NiCoForm Inc. currently occupies. In addition, the use of lubricated air is prohibited at NiCoForm due to the apprehension of contaminating nearby plating tanks.
The concept of a voice coil actuated fatigue life tester has no objections from the senior design team. It is capable of high frequency movement. There are no elements of the actuator that will wear over time or need lubrication. The only noise that will be produced is from the cooling fans and the movement of air around the coil assembly.
3.4 Feasibility Conclusions
Based on the above-mentioned reasons and the needs of the sponsor, it is clear to the senior design team the voice coil concept is the most practical and robust option. The voice coil is theoretically capable of infinite life, barring any damage caused by the coil winding colliding with the field assembly. It is a non-contact linear motion device, therefore, there is no wear associated with extended use. This also means there is no need for lubrication of the system. The voice coil is capable of frequencies exceeding 100Hz, exceeding the performance objectives as seen in section 4.2. This concept will be further developed into a workable bellows fatigue life tester.
4.0 Design and Performance Objectives
In order for us to continue with our design, it is necessary to compile a list of design and performance objectives. These objectives are measurable quantities or attributes related to the customer’s needs and the basis for scrutiny of our completed design. The objectives are covered in the following chapters.
4.1 Design Objectives
The design objectives are a reflection of the needs of our customer. They are not crucial to the operation of the apparatus; they, however, will prove essential to agreeable operation. The design objectives are as follows:
4.2 Performance Objectives
Performance objectives are criteria against which the design can be physically measured. Our design team considers the subsequent objectives key to meeting the needs of our customer and the overall design of the tester:
5.0 Analysis of Problem
A thorough analysis of the fatigue-life testing platform was necessary to ensure the design criteria decided by NiCoForm, Inc. and the NiCoForm Design team would be met. The design and analysis were broken into a number of parts including actuator selection, testing platform design, stress analysis of the bellows adapters, and finally electronics selection.
Selection of an actuator was the most critical area in the design and analysis. It was imperative the selected actuator met the design criteria for total stroke and required force for the application.
A spreadsheet was generated (see Appendix A) which calculates the information necessary to choose an actuator such as acceleration, force, and generated back EMF. These values were calculated from the frequency of operation, the stroke of the actuator, and given information from the BEI specifications sheet as follows:
, , , , And
It follows that:
Now acceleration equals twice the distance, divided by the square of the time. This calculation is derived from one half of the triangular velocity profile.
It was later decided that although the above calculation is good for a starting estimate, the acceleration should be calculated for a sinusoidal velocity profile. The sinusoidal profile will be the likely waveform to drive the actuator. Here is how it was calculated:
Taking the time derivative, we get:
and taking the time derivative of that we get:
Therefore, maximum acceleration is equal to .
5.1.1 Actuator Flexure / Spider
It is important to accurately center the coil to the field assembly of the actuator for proper actuation. It is also necessary to provide a means by which to support the coil when the actuator is not powered. Taking from speaker technology, a poly-cotton diaphragm flexure, better know as a speaker spider was selected to perform both jobs. When used within their design limits, poly-cotton spiders have a near infinite fatigue life. Actual selection and fitting of the diaphragm will be accomplished once NiCoForm, Inc. receives the actuator. Replacement surrounds are commercially available at low cost in the event the surround fails or is damaged at some point in the future.
The PLC Expansion Base being utilized is the D2-04B-1. The unit is manufactured by Automation Direct which is a company specializing in PLC hardware of a large variety. The unit is a four-slot base with an internal 110/220-volt power supply. One of the main reasons for choosing this unit is that NiCoForm’s employees are already familiar with the unit. This unit has been used on some of NiCoForm’s previous projects with much success. Their familiarity also opens new avenues for the design team to seek assistance when implementing and debugging the fatigue life tester. The current employees have some knowledge as to how the unit works and could become a critical resource next quarter during construction. A second reason for choosing this unit is the availability of accessories and add-ons that may need to be utilized to make the project a success. The expansion base has specific mounting specifications, which need to be taken into consideration when choosing the electrical enclosure.
Automation Direct also manufactures the CPU to be included with the expansion base to control a majority of the system’s functions. The CPU module chosen was the D2-250-1. As with the expansion base, the CPU module was chosen primarily due to the existing familiarity of the customer. Secondary factors taken into consideration were compatibility, availability, and price of both the unit itself and add-on modules. The CPU module will be programmed using a ladder logic design using DirectSoft32 software, which will be supplied by NiCoForm and was originally purchased from Automation Direct.
The operator interface chosen for this application is yet another product offered by Automation Direct. The EZText 220P is a two line by twenty characters display that also offers a keypad for inputting the numerical data necessary for conducting the fatigue life test of each bellows. The HMI will control the PLC’s CPU directly before, during, and after the test stage. It will allow the user running the test to input criteria such as frequency and pressurization level based on, and calculated from, properties of the individual bellows being tested at that time. During the test stage the two-line by twenty-character display will read the number of cycles ran up to that point in time, and a pressure reading of the internal pressure of the closed system for the purpose of detecting failure in the bellows.
The motion controller decision was based mainly on one main criterion: which linear actuator is to be utilized. The actuator chosen, which is discussed in section 5.1, also comes from BEI Technologies Inc. This limited the design team’s choices to controllers offered by BEI. They only offer one controller for all of their linear voice coil actuators, so once the actuator was chosen the choice for controller was set.
5.2.5 Pressure Switch
Since it is necessary to detect the failure of a bellows for a test to come to an end, the team decided that pressurizing the inside of the bellows and monitoring the pressure would be a viable solution. Initial thoughts centered on using a pressure transducer to send signal to the PLC. Code would be written to compare the output signal of the pressure transducer with a set value. When the pressure transducer output fell below the preset value, signaling a failure, the test would be stopped and the number of cycle to failure counted. Background research and pricing revealed the cost of pressure transducers to be relatively high, so a decision was made to look into a less expensive method of detecting a bellows failure.
More thorough background research led the team to decide on the use of a low cost pressure switch. The solenoid will open at a preset pressure and close once the pressure falls below a set level again, signaling the PLC to stop the test and count the fatigue cycles. The decision to utilize a pressure switch instead of a pressure transducer simplifies the design because there will be no need for calibration of the switch -- there would have to be if the transducer were used.
An adjustable pressure switch was selected based on rough calculations of an assumed pressure to which the bellows will be pressurized. The user will have the ability to adjust the set pressure for different sized bellows if necessary.
The solenoid valve was selected based on the pressure range, budget considerations, and on the amount of power available from the PLC. The types of connections available were also a consideration. The # 7994K57 Miniature Solenoid Valve from McMaster-Carr fit all of our design requirements and was among the lowest priced valves we found. The valve will receive a DC signal to open and pressurize the bellows to approximately 10 psi at the onset of the test. Once the bellows reaches this pressure, the PLC will cause the solenoid to close and remain closed until the pressure switch is thrown due to a drop in pressure below its set lower limit.
The bellows will need to be attached to the testing apparatus at two locations. The first location will be held in place, the second will be connected to the actuator. For each differing bellows diameter an adapter was designed. The design analysis is shown below.
The pressure-side bellows adapter needs to secure the metal bellows to the head. The design considerations for this were incorporation of a through hole to allow for the pressurization of the bellows and the ability to withstand the forces the adapter will see without failing. Our original system design is tailored to test a 0.5” diameter bellows, hence that is what was analyzed. A finite element analysis using GiD was conducted to determine if the geometry of the adapter has any high stress concentrations to take into consideration. As can be seen from the amount of force the adapter encounters, the stress is well below the material endurance limit (See Figure 5.1). The maximum stress seen by the adapter is 1.1892e6 Pa, or about 172.5 psi.
Figure 5. 1 Adapter in Compression - 20lbf
The material used to manufacture the adapter has an endurance limit of 16.6-ksi. The design will have no problem holding up to the stresses involved in a test run.
The drive-side bellows adapter will see stresses that are dependent upon the spring rate of the bellows and the acceleration of the adapter as it changes direction during a test. Again, the finite element analysis was done using GiD. This time the adapter was constrained at the four screw locations in the X, Y, and Z directions, as the screw locations were the main areas of concern. A “self-weight load” was applied to the adapter that would simulate an acceleration of 50g’s. For the mass of the adapter, this acceleration creates a 61.00 Newton force, or 13.71 pounds force. This amount of force applied to the adapter reveals the stress concentrations seen in Figure 5.2.
Figure 5. 2 Stress Concentration at 50G Acceleration
The maximum stress calculated was 3.3084e6 Pa, or 479.84 psi. Once again, the model shows stress concentrations well below the 16.6ksi endurance limit. Based on the Finite Element Analysis, the design should work to the expectation of the design team.
There are an extensive variety of bellows designs and a variety of testing scenarios. The adapters will need to be manufactured to suit each of these situations. The basic design will remain the same; only the diameters to suit the bellows will change. In the future, testing will need to be done for each individual adapter design.
5.4.1 Main Support Rods
From the beginning of the design process, it was clear that the support rods should be fabricated using a ferrous metal because of the ease of accurate prediction of an endurance limit. It is also clear the forces generated will be on the order of tens of pounds, and surely less than 100 pounds. The initial design has the actuator located at the top of the platform. If this design were implemented, the support rods would be subjected to alternating compressive and tensile loads due to the acceleration of the mass of the coil, bellows adapter, and bellows being tested. There would also be a spring force generated due to the compression and extension of the bellows themselves. This force would be equal to the spring constant of the bellows multiplied by the distance of compression or extension. In changing the design, the actuator was relocated at the base of the testing platform. This means that the only forces the support rods will now see are the weight of the pressure-side components and the relatively small spring force generated from the compression of the bellows. Under these conditions, the main consideration in the selection of stock support rods is that the stress be kept below the endurance limit of the selected material.
A spreadsheet was generated (see Appendix B) that will calculate the endurance limit necessary for our material selection as well as the axial stress for different rod diameters as a function of applied load. From the spreadsheet, it is apparent that the support rods can be fabricated using small diameter rod and still survive the abuse inherent in fatigue testing. In addition, it is also important that the support rods allow smooth movement of the tester top without jamming. Smaller diameter rod may have a tendency to bind the tester top when it’s loosened and relocated.
While the calculations show a small diameter rod can be used it is still important to keep the mass of the test platform reasonably high in an effort to ensure a minimal amount of vibration during operation. Precision ground rod stock that is one inch in diameter and composed of hardened 304 stainless steel was selected. Stainless steel, because of its chromium content, is resistant to oxidation is which is import to ensure no surface rust will develop and impede the movement of the head in its vertical travel.
5.4.2 Base Plate
|Fatigue Life Tester for Electroformed Bellows Team 04035 Zach Eastman||James P. Chambers1,*, Bernard P. Arulanandam1, Leann L. Matta2, Alex Weis3, and James J. Valdes4|
|Chaire «James McGill» d’étude du discours social|
«James McGill» d’étude du discours social (The James McGill Professorship of Social Discourse Theory) pour deux mandats consécutifs,...
|The Project Gutenberg ebook of Ulysses, by James Joyce #4 in our series by James Joyce|
|Reading on gathering by Charles Eastman||James M. Noble md, ms|
|James p. Mc culley, md, Chair||James P. McCulley, md, Chair|
|James p. Mc culley, md, Chair||20: 04: 19 James Jacobs joins Main|