INL Glovebox Tensile Testing System

The goal of this project is to create a small-scale tensile testing system for Idaho National Laboratory with the capability to perform high-temperature tests that can be easily installed in a glovebox.

Background
Idaho National Laboratory is currently researching a Uranium-Plutonium-Zirconium alloy as a potential solid fuel for nuclear reactors. While some testing has already been done, there is still much more to learn, particularly about the mechanical properties of the alloys. Due to its radioactive nature, any test performed on the U-Pu-Zr alloy must be done in a glovebox for safety, hence the need for a glovebox-based tensile tester. Our machine will enable INL to further their knowledge of this alloy and its potential usefulness. There are many models of tensile tester that already exist, but none meet the requirements set by INL. Many of these machines, especially those with the capability to perform high-temperature tests, are much too large and complex to be used in a glovebox. The models that are small enough are designed for much smaller loads than required for this device and cannot heat testing samples.

Deliverables
At the end of this project, we will have a functional tensile testing device that meets the requirements of INL. We will also create detailed instructions for the assembly of the device inside the glovebox, as well as usage instructions for performing tests and operating the heater.

Requirements
In addition to these, there were some more qualitative requirements. The device had to:
 * Provide accurate data regarding sample extension and applied force
 * Be capable of running a test from outside the glovebox after initial setup of sample and heater elements
 * Contain all external electronic components and controls in one location, ideally with a control panel for the ease of the user
 * Be visually appealing. While not a hard constraint, we wanted our machine to look more like a finished product than a rough prototype. We did this through painting and using a box to manage electrical systems

Design Considerations
The main limitation of the glovebox is a size constraint. Not only is there limited space inside the glovebox, but whatever we design has to fit through an 8” diameter hole so that researchers at INL can install it in the glovebox. On top of that, the system must be able to perform high temperature tests up to 700 C. Due to the radiation inside the glovebox, as many of the electronic components need to be outside as possible. The glovebox does have passthroughs that enable this. For the safety of the user, no external surface can exceed 50C, so some form of insulation will be needed around the heating chamber.

Tensile Testing
Tensile testing is used to measure certain mechanical properties of a material. The test is performed by pulling on a sample until it fractures and measuring the sample’s behavior. Measured data from the test is then used to generate a stress-strain curve. There are numerous properties that can be found through tensile testing. The first is a stress-strain curve, which plots the relationship between the stress and strain in a sample during the test. The stress is a measure of the force applied over the cross-sectional area of the sample ($$\sigma=\frac{F}{A}$$). The strain measures how far the sample elongates during the test relative to its original length ($$\epsilon=\frac{\Delta L}{L_0}$$). The linear portion of the graph is known as the elastic region, in which the material will return to its original shape after the load is removed. The slope of this line is known as Young’s Modulus. Past the yield strength, any deformation is permanent. Also of note is the ultimate tensile strength (UTS), which is the maximum stress that material can withstand. Finally, the fracture stress is the point at which the sample breaks.



ASTM Standards
The accepted standards for tensile tests for metals are ASTM A370 and ASTM E8. These specify test methods and specimens for a valid tensile test. While at its most basic, a tensile test is simply applying a pulling force to the sample, these standards lay out practices to ensure accurate testing and acceptable results. First, the force applied must only be axial, otherwise there is a torque applied and the test is not purely tensile. This is ensured by properly aligning the sample in whatever grip or mounting system is used. Second, the speed of the test must be slow enough that accurate data can be collected. This would be done in the code for controlling the motor. ASTM A370 also specifies sample geometry and manufacturing methods. Due to the small scale of our testing machine, we opted to use the smallest sample geometries available, namely the 2.5mm and 4mm diameters. This was eventually changed to only the 2.5mm to ensure that our machine could apply enough force for fracture. The samples that would be used at INL would have to be manufactured in the glovebox. This would be done by initially casting them to size, followed by turning on a manual lathe, then sanding to final size.

U-Pu-Zr Alloy
The existing research on U-Pu-Zr alloys has been summarized in a paper by Janney et al. The paper gives data from previous studies about the thermal and mechanical properties of certain weight percentages of the alloy. While many properties are discussed, the most useful for our purposes was the tensile information. Results for twelve samples were given, including ultimate tensile strength, yield strength, and Young’s Modulus. Unfortunately, the data is minimal and often inconsistent between sources. These samples are in the range of U-(10-20)Pu-(5-15)Zr. For this project, INL asked us to focus on alloys in the U-(20-30wt%)Pu-(10-20wt%)Zr range for which there is no existing data. There is no clear relationship between the weight percentage and any material properties. On top of that, the range of values varies greatly between researchers, making it difficult to estimate what force would be needed for a fracture test.

Maximum Force Calculations


Initial force calculations were performed based on previous tensile data from Janney et al. and a specified sample geometry from ASTM A370. Since we did not know what strengths to expect, we based our design on the smallest and largest ultimate strength of any weight percentage of the alloy that we had available. We figured that this would ensure our device could sufficiently handle a wide range of varying material strengths. The ultimate tensile strength was given in kg/mm2, which had to be multiplied by gravitational acceleration to get units of MPa, which could be multiplied by the cross-sectional area of the sample to find the maximum force that would be required.

During initial calculations, we divided the stress by area instead of multiplying it by the area. These calculations were not checked thoroughly. This meant we dramatically underestimated the maximum force that would be required. The max force ended up being around 1890 lbf on a 4 mm sample, much more than the 12 lbf we believed we were working with initially.

PID Heater Control System
Temperature Control Concept:

When developing the concept of the temperature control for our heating chamber a Proportional Integral Derivative (PID) controller was always the main concept for user control. Finding a PID was the most difficult part, with so many different components and wiring schemes this led to many concepts that would work for our design and needs. Many PID systems, when heating a nickel chromium wire to high temperatures can lead to PID run off which can prove to be an extreme danger to the user. This was brought to our attention during a snapshot check in and during our design review which led us to choosing the following system.

The initial concept was to use two separate PID setups, one for each side of the heater connecting to two different nickel chromium wires. This concept was decided to be the best one among the team which was the basis concept when coming up with the products needed. This led to researching a Solid-State Relay (SSR), a PID, an emergency stop switch, thermocouple, and any accessories or attachments that are required to run the system. Due to the runoff possibility of the nichrome wire we decided to use an emergency stop switch to quickly cut off power to the system, as well as integrating a toggle to cut off power directly to the nichrome wire. When finding the correct nichrome wire to select we had to research different gauge wires to figure out which ones would use the correct current and power based on the length of wire we desired.

Once the temperature control devices have been selected, a nichrome wire needed to be selected. For this project we have started with 18 and 24 AWG wire, determined off Design Evaluation: table #, which allow calculations to begin depending on the resistance per foot [Ω/ft]. For 18 AWG wire it has properties of close to 0.5 Ω/ft which helps determine the lengths from the governing equations below:

P= V^2*R                                    (1)

I= V*R					 	  (2) Length= (Ω/ft)*R                                 (3)

These governing equations help determine key values, that prove extremely valuable for validation and testing. Equation solves for power (P) in the units of watts. This uses Voltage (V), in units of volts, squared divided by the resistance (R), in units of Ω. Equation (2) is used to solve for the current (I) which is equal to the voltage over the resistance. Equation (3) is intricate to our heater design and needs to find a length between 12 [ft] and 20[ft]. If any of those uncoiled wire lengths become too short the wire will not be able to reach both inside sides of the heater. The values determined by the governing equations, (1) and (2), assist in deciding if the heater system is safe to use in standard outlets rated for 1800 [watts] and 15 [amps].

When determining the nichrome wire the calculations needed to be performed as well as performing tests to validate that system can meet the project requirements. These calculations and the validation of the system can be seen in the design evaluation section of this design report.

Temperature Control Selection:

When selecting the temperature control system, we took our initial design as previously stated in the concept consideration section and focused on integrating that system. The next design choice was to decide which products to purchase to start wiring the system. The following is the items purchased for selection:

PID Temperature Controller: CN16D3-S-AC

Solid State Relay: SSRL240DC50

Heat Sink

Fuse: 30 Amp

Nichrome Wire: 18 AWG wire – 0.0403” diameter

Wiring: 16 AWG copper wire

Emergency Stop Switch: POWERTEC 71007

The components above were selected based off safety considerations and ease of understanding to meet our project requirements. The PID was selected for the autotune and alarm features, for the maximum temperature, and for compatibility of k-type thermocouple. After selecting the PID, the SSR was chosen because of the compatibility between the PID and SSR. The SSR was selected for the current it was capable of, 50 [amps]. The nichrome wire was selected from the calculations shown in the appendices: table # for both the 24 AWG and 18 AWG. The calculations show that the 18 AWG nichrome wire can reach the desired maximum temperature of 800 [C], and then through validation show the inside heater temperature can reach the desired temperature of 700 [C] without reaching above a standard outlet’s ratings of 1800 [W] and 15 [A].

Temperature Control Architecture:

The conceptual design for the heaters temperature control system was to make one single heating system instead of two for both sides of the heater. The design is to use one single nichrome wire that will coil between both sides with a section of loose coiled wire to extend to each side. The temperature control system uses the circuit shown on the right in the figure.

PID Circuit Diagram

The diagram above shows how the PID temperature control system will be wired and function. The voltage source is a standard wall outlet connected to the emergency stop switch. From the emergency stop switch the power goes to the PID (Proportional Integral Derivative) connecting at pin 1 and pin 2. The PID has the k-type thermocouple connected to pins 11 and 12 that will be attached to the nichrome wire to read the temperature. On pins 5 and 6 on the PID is where the SSR (Solid State Relay) will be connected. The SSR will connect back to the switch while passing through the safety fuse and then the 1/L1 will connect to the nichrome wire then back to ground on the switch.

Nichrome Wire Validation:
To validate the nichrome wire based on the calculations from equations (1), (2), and (3); testing needed to be performed to validate the calculated values. The calculated values for the 18 AWG and 24 AWG wire from those equations, were solved in excel and are in the Appendix A.1. From those values calculated a graph was created to show how the temperature was affected based off the amount of power being supplied. The 18 AWG wire was selected due to the lengths shown in table A.2 in the appendix. With the 18 AWG wire selected, the Figure (3) below was made to determine the maximum temperature that could be achieved while remaining below the standard outlet’s ratings.

The graph to the right shows the ideal power determined to reach 850 [C] while still maintaining a length of uncoiled wire between 12 to 20 [ft]. This can be shown in the green dot and line which show that all requirements can be met with this 18 AWG nichrome wire. The red dot shows the next resistance of 8 [Ω] lower which resulted in larger power and current value which exceeded standard outlets. The voltage from the outlet, power supply, was known to be 120 [Volts] and the resistance was measure from a multimeter to determine a resistance of 8.4 [Ω]. From the equations mentioned earlier and the graph above, the following values were obtained:

Current: 14.2857 [A]​

Power: 1714.39 [W]​

Wire Length: 16.8 [ft]

The next step was to perform a PID temperature validation test, in which the circuit shown in diagram (#) was completed and the test was ready to be performed. The test was setup to only contain the nichrome wire and the selected PID circuit. The results led to the nichrome wire reaching a temperature of 806 [C] which validates that the nichrome can produce enough heat to bring the temperature of a sample up to 700 [C]. Following this test, the heater would need to be validated to verify the temperature of the inside heating chamber.

Heater Furnace Temperature Validation:
To validate the heating chamber, a heat transfer simulation and temperature test needed to be performed. The SolidWorks simulation, Appendix, validates that the inside chamber temperature can reach the desired maximum temperature of 700 [C] and that the outside temperature will drop as the heat goes through the refractory cement, air gap, and the heaters shell resulting in an outside surface temperature of 40 [C].

To perform the temperature test of the heater, the nichrome wire properties previous stated in the validation were used and integrated into the heater. The coiled nichrome wire was placed into the heating chamber with both ends connecting to two split bolt copper connectors that were submerged into the refractory cement bricks. In the split bolt connectors, the 16 AWG connection wire was wrapped around the nichrome wire ends then the split bolts were tightened. From there the 16 AWG wire was connected to the PID circuit Figure (1). To validate the PID temperature and the inside heating temperature an Arduino gravity digital high temperature sensor (DFR 0558) was used to measure the air temperature. With the test setup complete, tests were ready to be performed.

The temperature data, Appendix A.3, shows the data collected and the input temperature from the PID and the output temperature recorded by the PID. Below is Figure (4) representing the data collected from the Arduino temperature sensor that is measure the inside chamber of the heater per second.



With the temperature [C] on the y-axis and the time [s] on the x-axis this shows how the heater and the PID worked together to reach a desired temperature of 700 [C]. This graph can be split into three sections.
 * The Heat Up: This section shows how long the heater took to heat up to the desired temperature of 700[C]. This took roughly 7 minutes to heat up the specimen chamber.
 * The Soak Time: This section shows the time spent holding the temperature at 700 [C]. With the current PID and heater set up the temperature fluctuates and does not stay constant due to the SSR being an on/off power switch. This fluctuation could be made more accurate which will be discussed in the further work section. Even with the fluctuation the average temperature of the soak time was 706.85 [C] for 8.3 minutes which meets our required maximum temperature.
 * The Cool Down: This section shows the heating chamber cooling down to 100 [C] in about 5 minutes.

With the PID and the heater validated to reach maximum temperature, evaluating if the outside surface temperature of the heater shell was below 50 [C] was done concurrent with the test above. The surface temperature was measured using a laser temperature reader, with the hope to use a temperature sensing camera in further tests. The laser temperature reader found that the shell temperature ranged from 60 – 250 [C], peaking at the extensometer passageway at the 250 [C]. This does not validate our requirement which plans to fix the surface temperature will be shown Further Work section of this report.

Nichrome Wire Validation:
To validate the nichrome wire based on the calculations from equations (1), (2), and (3); testing needed to be performed to validate the calculated values. The calculated values for the 18 AWG and 24 AWG wire from those equations, were solved in excel and are in the Appendix A.1. From those values calculated a graph was created to show how the temperature was affected based off the amount of power being supplied. The 18 AWG wire was selected due to the lengths shown in table A.2 in the appendix. With the 18 AWG wire selected, the Figure (3) below was made to determine the maximum temperature that could be achieved while remaining below the standard outlet’s ratings.

Figure (3)

The graph above shows the ideal power determined to reach 850 [C] while still maintaining a length of uncoiled wire between 12 to 20 [ft]. This can be shown in the green dot and line which show that all requirements can be met with this 18 AWG nichrome wire. The red dot shows the next resistance of 8 [Ω] lower which resulted in larger power and current value which exceeded standard outlets. The voltage from the outlet, power supply, was known to be 120 [Volts] and the resistance was measure from a multimeter to determine a resistance of 8.4 [Ω]. From the equations mentioned earlier and the graph above, the following values were obtained:

Current: 14.2857 [A]​

Power: 1714.39 [W]​

Wire Length: 16.8 [ft]

The next step was to perform a PID temperature validation test, in which the circuit shown in diagram (#) was completed and the test was ready to be performed. The test was setup to only contain the nichrome wire and the selected PID circuit. The results led to the nichrome wire reaching a temperature of 806 [C] which validates that the nichrome can produce enough heat to bring the temperature of a sample up to 700 [C]. Following this test, the heater would need to be validated to verify the temperature of the inside heating chamber.

Heater Furnace Temperature Validation:

To validate the heating chamber, a heat transfer simulation and temperature test needed to be performed. The SolidWorks simulation, Appendix, validates that the inside chamber temperature can reach the desired maximum temperature of 700 [C] and that the outside temperature will drop as the heat goes through the refractory cement, air gap, and the heaters shell resulting in an outside surface temperature of 40 [C].

To perform the temperature test of the heater, the nichrome wire properties previous stated in the validation were used and integrated into the heater. The coiled nichrome wire was placed into the heating chamber with both ends connecting to two split bolt copper connectors that were submerged into the refractory cement bricks. In the split bolt connectors, the 16 AWG connection wire was wrapped around the nichrome wire ends then the split bolts were tightened. From there the 16 AWG wire was connected to the PID circuit Figure (1). To validate the PID temperature and the inside heating temperature an Arduino gravity digital high temperature sensor (DFR 0558) was used to measure the air temperature. With the test setup complete, tests were ready to be performed.

The temperature data, Appendix A.3, shows the data collected and the input temperature from the PID and the output temperature recorded by the PID. Below is Figure (4) representing the data collected from the Arduino temperature sensor that is measure the inside chamber of the heater per second.

Figure (4)

With the temperature [C] on the y-axis and the time [s] on the x-axis this shows how the heater and the PID worked together to reach a desired temperature of 700 [C]. This graph can be split into three sections.

The Heat Up: This section shows how long the heater took to heat up to the desired temperature of 700[C]. This took roughly 7 minutes to heat up the specimen chamber.

The Soak Time: This section shows the time spent holding the temperature at 700 [C]. With the current PID and heater set up the temperature fluctuates and does not stay constant due to the SSR being an on/off power switch. This fluctuation could be made more accurate which will be discussed in the further work section. Even with the fluctuation the average temperature of the soak time was 706.85 [C] for 8.3 minutes which meets our required maximum temperature.

The Cool Down: This section shows the heating chamber cooling down to 100 [C] in about 5 minutes.

With the PID and the heater validated to reach maximum temperature, evaluating if the outside surface temperature of the heater shell was below 50 [C] was done concurrent with the test above. The surface temperature was measured using a laser temperature reader, with the hope to use a temperature sensing camera in further tests. The laser temperature reader found that the shell temperature ranged from 60 – 250 [C], peaking at the extensometer passageway at the 250 [C]. This does not validate our requirement which plans to fix the surface temperature will be shown Further Work section of this report.

Team Members


Team members from left to right: Matt Uptmor Major: Mechanical Engineering Hometown: Meridian, ID Responsibility: Scheduling, Wiki Master Email: uptm3100@vandals.uidaho.edu

Logan Matti Major: Mechanical Engineering Hometown: Redding, CA Responsibility:Budgeting Email: matt3733@vandals.uidaho.edu

James Bradley Major: Mechanical Engineering Hometown: Sandpoint, ID Responsibility: Project Manager Email: brad4056@vandals.uidaho.edu

Jared Gray Major: Mechanical Engineering Hometown: Boise, ID Responsibility: Client Contact Email: gray6722@vandals.uidaho.edu

Additional Documentation
Project Schedule

Fits Like a Glove Gantt Chart

Meeting Minutes

[Fits Like a Glove Meeting Minutes Document]

Presentations