Paper Number



Dynamics Division

Dynamic Environmental Testing of Test Pallet

Jeffrey T. Wyant

Copyright © 2006 Department of Mechanical Engineering, Colorado State University

ABSTRACT

THIS PAPER COVERS THE DYNAMIC ENVIRONMENTAL TESTING THAT WILL BE PERFORMED ON THE TEST PALLET FOR THE JAPANESE AEROSPACE EXPLORATION AGENCY BARE TETHER EXPERIMENT. THIS PAPER WILL FOCUS ON THE TESTS TO BE PERFORMED IN ORDER TO ENSURE THE TEST PALLET WILL SURVIVE THE HARSH ENVIRONMENT OF LAUNCH ON AN S-310 SOUNDING ROCKET. THE TESTS TO BE PERFORMED INCLUDE A SINE SWEEP, RANDOM VIBRATION, QUASI-STATIC, SHOCK TESTING AND ACOUSTIC VIBRATION TESTING. THIS PAPER COVERS RESEARCH INVOLVED IN PREPARING THE TEST, HOW THE TESTS WILL BE PERFORMED AND HOW THE TESTS WILL BE ANALYZED AND APPLIED TO OUR FINAL TEST PALLET DESIGN.

INTRODUCTION

BACKGROUND

During launch of the S-310 sounding rocket the test pallet is expected to experience harsh dynamic conditions. They include vibration from the rocket and its engines, significant G accelerations from propulsion and from the rocket spinning, and shock from different parts of the rocket separating. In order to ensure the pallet will function properly throughout the entire mission, it must be tested for these conditions.

In the aerospace industry there are five major dynamic tests that are performed on flight or flight-like hardware to ensure they will survive the harsh conditions of launch. These tests include a sine sweep, random vibration, quasi-static, shock testing and acoustic vibration tests. The basics of each test are described below.

The sine sweep test is a test that runs the hardware through a range of frequencies, with the amplitude kept constant, to determine the natural frequencies of the tested item. By comparing the acceleration, usually measured in g’s, to frequency one can determine the natural frequencies of the test item. This is done by placing accelerometers on the test item so the acceleration can be measured. The test item is then run through the range of frequencies, usually done on a shaker table, while the accelerations from the accelerometers are recorded. When the acceleration is plotted against the frequency one might observe large peaks at certain frequencies. The frequencies where these peaks occur are the natural frequencies. If any of these frequencies are expected to be experienced during launch, design consideration will have to be made. One would need to either re-design to change the natural frequencies of the test item or design the tested item to withstand the increased accelerations due to experiencing resonance.

The random vibration test is a test that puts the test item through a wide range of frequencies that might be experienced during launch. Instead of limiting the vibration to one frequency at a time, the test item is subjected to a range of frequencies applied at the same time. The frequencies can be applied at the same time by stacking the frequencies on top of each other, similar to how a radio signal works. The frequencies are applied on a random basis and natural frequencies are usually notched out so the test item will not experience resonance during the test. The test is defined with a random vibration profile in which the Power Spectral Density (PSD), measured in g2/Hz, is plotted against the frequency. The random vibration profile is usually defined by the item inducing the frequency; in our case, it’s the S-310 sounding rocket during launch. A simple way to get the Power Spectral Density from acceleration and frequency is as follows:

[pic]

Courtesy of Motorolla

Another important part of the random vibration profile to be noted is the area under the curve. The square root of the area under the curve is the Grms, or the root mean squared of the g accelerations, of the vibrations. This is shown as follows:

[pic]

The test is performed on a shaker table which induces the random set of vibrations to the test item. Once completed, thorough examination and tests are conducted to expose any failures that may have occurred. These weaknesses are then corrected and the test is re-run until no failures are observed.

Quasi-static testing is a test that puts the test item through high accelerations to ensure it is structurally sound and can withstand large g forces involved with vibrations during launch. This is done by holding the frequency constant, usually from 3-10 Hz, and increasing the amplitude of the vibration, usually until failure occurs. This causes large g force to be imposed on the test item and exposes any structural weaknesses in the design.

Shock testing is done to ensure the test item can withstand instantaneous vibrations, usually due to the discharge of ordinance on the rocket, which will be experienced during launch. This test is analogous to indirectly hitting the test item with a hammer. This test is similar to the random vibration because a plot of Power Spectral Density to frequency is used to define the test parameters. This test is usually done by mounting the test item to a plate or structure that will induce the “shock” to the test item. The structure is then subjected to the instantaneous vibration, or shock, through a hammer device. The hammer device is usually a heavy piece of metal that is dropped along a guide and hits the structure inducing the vibrations. The vibration is controlled by changing the way the hammer strikes the structure through height released and dampening mechanisms.

Acoustic vibration tests subject the test item to large amplitude sound waves which then induce vibration on the test item. These vibrations are usually caused by large portions of the rocket, such as the faring, vibrating. This, in turn, fills the rocket with high-amplitude sound waves and can damage the flight hardware. To simulate this environment, the test item is placed in a room with a large speaker. The speaker then subjects the test item to the desired sound waves and frequencies. After the test, the item is inspected and tested to ensure no failures occurred. In our case, acoustic vibrations are not going to be as important as in a larger-scale rocket because there a not large portions of the rocket around or pallet that will induce high amplitude sound waves.

Since specific testing of the hollow cathode is discussed in this paper, it is important to go over what the hollow cathode is and how it works. For the purpose of this experiment, the hollow cathode’s function is to ground the test pallet to the space plasma in the earth’s outer atmosphere. This is done by ionizing an inert gas, in this case xenon gas, which produces free electrons that can carry current from the hollow cathode to the space plasma. This is achieved by first inducing a flow of about 4-6 SCCM of inert gas through the hollow cathode and heating the cathode up to 1100 degrees Celsius via the heater coils wrapped around the outside of the cathode. Once heated, a porous tungsten insert inside the hollow cathode is then able to boil electrons off the inert gas and ionize it. A keeper just in front of the hollow cathode creates a voltage bias which in turn accelerates the electrons into the atmosphere. This flow of electrons is what allows current to flow from the pallet to the space plasma.

[pic]

Courtesy of Dr. John Williams

Figure 1: How the Hollow Cathode Works

The hollow cathode is an extremely important aspect of this project because if it fails, the entire Japanese Aerospace Exploration Agency bare tether experiment will not be able to be performed. This is why it is crucial that the hollow cathode be tested extensively to ensure its functionality though out the entire mission.

Enviornmental Testing

TESTING

Analytical Analysis

In order to verify that the tests results are close to what is expected, the tests will be first conducted analytically. In the following calculations the natural frequency for the hollow cathode, without the heater coils, will be calculated. For this calculation the hollow cathode can be treated as a simple cantilever beam and fundamental natural frequency equations can be used. Using the results from these equations it will give a good idea of what the natural frequencies should be. The fundamental natural frequency equation is as follows:

[pic]

Where:

[pic] (static deflection)

Since:

[pic] (distributed load)

[pic] (distributed mass)

This can be simplified to:

[pic]

Where:

[pic] (moment of inertia)

W = weight

L = length

E = modulus of elasticity

m = mass

ro and ri = cross-sectional outer and inner radii

wo = weight per unity length

mo = mass per unit length

A = cross-sectional area

Using these equations on the hollow cathode: [pic]

E = 186 GPa for tantalum

L = 4.13 cm

m = 7.51 g

mo = .177 kg/m

[pic]

This result shows a natural frequency around 1500Hz should be expected.

Computer Analysis

Pro-Engineer static and modal analyses were also run on the hollow cathode and heater coil. A model of the hollow cathode and heater where developed in Pro-Engineer as shown below.

[pic]

Figure 2: Pro-E Hollow Cathode and Heater coil model

[pic]

Figure 3: Close up of Heater Coil

In order for the program to be able to mesh this part the part had to be modified with some less complex features. This includes making the 10 heater coil wraps around the hollow cathode into one long cylinder and changing the outer cross section of the heater coil from a circle into a square as shown below. This allowed for an easier and less time consuming analysis process.

[pic]

Figure 4: Modified Heater Coil and Hollow Cathode Model

[pic]

Figure 5: Close up of Modified Heater Coil

The first analysis performed was a static analysis to determine how the hollow cathode would react under accelerations due to the spinning and propulsion of the rocket. According to the European space agency, the maximum acceleration of a rocket similar to ours was ten g’s, or 3862.2 in/sec2 (note: S-310 information could not be found so European Space Agency information from a similar sounding rocket is being used for now). For the spin acceleration, a spin rate of 2.8 rev/sec was used to calculate a tangential acceleration of 1547.6 in/sec2. When applied to the Pro-Engineer model it produced the following results.

[pic]

Figure 6: Displacement Results

[pic]

Figure 7: Stress Results

[pic]

Figure 8: Maximum Stress

|Max Von Mises Stress |881.2 psi |

|Max Displacement |.000126 inches |

Both the stress and displacement results are within an acceptable level such that structural failure is not expected.

In the modal analysis a frequency range of 5Hz to 2000Hz was used since this is the sine sweep range suggested by the European Space Agency. The analysis resulted in two modes in this frequency range, 1061.3Hz and 1176.54Hz. The two mode results are shown below.

[pic]

Figure 9: Mode 1: 1061.3Hz

[pic]

Figure 10: Mode 2: 1176.54Hz

When compared to the computational analysis done previously, the results are fairly similar. This leads to confidence that the resonance frequency of the hollow cathode will be in the 1 to 1.5kHz range.

Performing test verification analytically for the random vibration, quasi-static and shock test will be too difficult since we are looking for component failure in these tests. Since I am unable to produce similar tests in Pro-Engineer, computer analysis is also not an option at this point. These tests will probably be run under specified conditions, which will be checked for accuracy, without preliminary analysis.

Test Schedule

To make these tests more successful, a certain testing schedule will be used in order to preserve the testing item as long as possible and minimize costs.

The first test that will be performed is the sine sweep test. Since the random vibration will notch out the resonance frequencies, one will need the results from the sine sweep to perform the random vibration test. Also, sine sweep is not designed to cause failure, therefore the pallet would, hopefully, survive without being damaged.

The next test to be performed will be the shock test. This test is the least likely to cause failure due to the fact that the shock from the separation of the mother and daughter of the rocket is not expected to be significant enough to cause failure. This will, hopefully, result in little to no repair after this test.

The random vibration test will be the next test to be performed. Since this test is designed to expose weaknesses due to failure, there may be significant repair and redesign required after this test is performed. If significant changes are made to the pallet, another sine sweep may be required because the natural frequencies could have changed during the alterations. Another shock test will probably not be needed if it passed the first time; but this will be determined as the situation arises.

The final test performed will be the quasi-static test. This test is designed to cause catastrophic structural failure due to the high-magnitude vibrations that occur during this test. In some cases, the test item can be destroyed so this will be the last test performed in order to first get results from the previous tests.

It is not unusual to run the sine sweep again after all the other tests have been performed and changes to the pallet have been made to ensure the resonance frequencies have not changed a significant amount.

Preliminary work

Before each test is performed, preliminary work must be done as far as defining the scope of each test, control boundaries, and if there is any pallet set up need for the test.

For the sine sweep test the range of frequency, amplitude and rate of increase of frequencies are needed to fully define the test. The European Space Agency recommends for sine sweep testing for its sounding rockets a range from 5Hz to 2000Hz, amplitude of .25 g’s and a sweep of 2 octave/minute. The sine sweep information for the S-310 rocket is not yet available, so these test parameters will be used until we receive the S-310 rocket information. A fixture to mount the pallet to the test machine will also need to be designed and manufactured in order to secure the pallet to the testing platform.

For the random vibration test the following profile was developed from the European space agency test recommendations. In this profile, for a frequency range of 20Hz to 1000Hz the PSD will increase from 0.01 g2/Hz to 0.1 g2/Hz at a rate of 1.8 dB/octave and for a frequency range of 1000Hz to 2000Hz the PSD will remain constant at 0.1 g2/Hz.

[pic]

Figure 11: Random Vibration Profile

As shown earlier the GRMS for this test can be calculated using the area under the profile line.

[pic]

[pic]

Where A1 is the area under the sloped part of the profile and A2 is the area under the flat part of the profile.

[pic]

[pic]

The specific test conditions for the quasi static test and the shock test have not yet been defined. Information from the Japanese Aerospace Exploration Agency (JAXA) is still need before these tests can be defined.

Facilities

For most vibration tests a machine called a sliptable is used to run the tests. These are usually fairly large pieces of equipment that are able to shake between 5Hz and 3000Hz. They use a method similar to speakers to shake the table on which the test item is mounted.

[pic]

Courtesy of Thermotron

Figure 12: Standard Sliptable Setup

A table like this one, but much smaller in size, will be needed to run the sine sweep, random vibration test, and quasi static test.

For the shock test the test set-up does not have to be as sophisticated a machine as the sliptable. At Northrop Grumman Space Technology (NGST) a structure created with I-beams and a guided solid-steel cylinder, also called the hammer, is used. Paper and cloth circular cut outs are used to dampen the impact of the “hammer” to achieve the required results. These types of testing equipment tend to not be consistent. An environment test conductor at NGST states that “anything ranging from the time of day to a fly landing on it seems to affect the output of this thing.” The structure is then fitted with the required measurement devices to ensure the test matches the desired test profile and is within acceptable accuracy. The exact machine that will be used on the test pallet must be determined.

Form of Results

The results for the sine sweep will be in the form of modes. A mode is the frequency at which the test item resonates. These modes occur at different frequencies, so our result will be in the form of frequencies at which the item resonates.

For the other three tests, random vibration, shock and quasi-static, the results will be in the form of pass or fail. After the test each Item will be put through a series of functionality as well as structural tests to ensure it is still in working order. It will either pass or fail these tests, i.e. it will either work or it will not.

Failure criteria

In the sine sweep test the pallet will be considered a bad design if it experiences resonance at too many frequencies within the range of the expected frequencies during launch.

For the random vibration, quasi-static and shock tests the pallet will have failed if it no longer functions properly after the test or there is a significant amount structural damage to the pallet.

analysis

Analysis of results

For the sine sweep the results will be used to determine the natural frequency of the pallet. Since resonance can cause catastrophic problems through increased g’s. The pallet will either be designed to avoid having natural frequencies that have the same frequencies of those experienced during launch or be designed to withstand resonance.

In order to avoid experiencing resonance for a significant amount of time one can either try to lower or raise the natural frequency of the pallet. This is usually done by altering the stiffness of the pallet through material selection. This can be seen in the fundamental natural frequency equation.

[pic]

As the equation shows, the natural frequency is inversely proportional to the static deflection. Therefore, if a stiffer material is used, the static deflection will be decreased and the natural frequency will be increased. Alternately, if the stiffness is decreased the deflection will be increased and the natural frequency will be decreased. Having a material with a high loss coefficient will also help to dampen the vibrations experienced by the pallet.

As shown in previous analysis the hollow cathode can expect to experience resonance within the sine sweep range. This is not a concern for the cathode itself in that it has a high enough cross sectional area to length ratio and is made of tantalum, a sufficiently stiff material that is expected to be able to withstand resonance. The concern is with the fragile heater coil that surrounds the cathode and is responsible for heating it to 1100 degrees Celsius. If this heater coil is damaged the entire experiment may fail. It is important to note that the coil does have some amount of give in that it is not permanently attached to the hollow cathode. This will help prevent damage to the heater coil at contact with the cathode. The point of largest concern is where the heater coil attaches to the mounting plate of the cathode. It will be rigidly supported at this point and too much movement could cause the heater coil to fail. This attachment point will need to be designed so the heater coil will not break at this point or the cathode may have to be stiffened to ensure there is limited motion in the heater coil.

For the other three tests, random vibration, shock and quasi-static, the analysis will be conducted through testing the functionally of the pallet after the test as well as visually inspecting the pallet to ensure there are no major structural flaws. The electronics, gas feed system and hollow cathode are the major points of concern and will each be tested extensively. After each test the functionality of the electronics will be tested by the real-time system team member. A run through of all electronic operations will be performed. The thermo/fluids team member will then test to see if any leaks or obstructions have developed in the gas flow system as well as test to make sure the cathode is still heating properly. Then there will be a visual test conducted by everyone to ensure there are no cracks or distortions in the structure or anywhere else on the pallet. Then the pallet will go through a total functionality test in which it will be run as if it were the actual experiment.

If it fails any of these tests, the pallet will be redesigned and re-tested until it does pass each test. This will ensure the quality of our final product and the completion of the test pallet’s experiment.

Conclusion

WHEN LAUNCHING ANY EQUIPMENT INTO SPACE VIA A ROCKET THERE ARE CERTAIN DYNAMIC ENVIRONMENTAL TESTS THAT MUST BE PERFORMED IN ORDER TO ENSURE THE EQUIPMENT WILL FUNCTION PROPERLY AFTER LAUNCH. THESE TESTS ARE A SINE SWEEP, RANDOM VIBRATION, QUASI-STATIC, SHOCK AND ACOUSTIC VIBRATION TESTS. THESE TESTS ARE PERFORMED BY EVERY AEROSPACE COMPANY IN THE UNITED STATES, IF NOT WORLDWIDE, BEFORE THEY PUT ANY PAYLOAD ON A ROCKET AND LAUNCH IT INTO SPACE. PERFORMING THESE TESTS HAS BEEN A STANDARD IN THE AEROSPACE INDUSTRY FOR SOME TIME AND UNDERSTANDING THESE TESTS AND WHY THEY ARE NEEDED IS CRUCIAL.

The test pallet team is no exception. The test pallet will be put through each of these tests just like in industry. But before performing each of these tests there are certain steps that had to be taken before it could be tested. These include developing the appropriate parameters and profiles for each test, most of which are defined by the S-310 sounding rocket, as well as performing preliminary analytical and computer analysis to gain confidence that the tests results are valid. A schedule of testing and knowledge of what is needed for each test is also necessary for a smooth and successful testing process. After testing has been completed, a full analysis of the pallet will be performed and, if any problems arise, the appropriate changes will be made and the pallet will be retested. This cycle will continue until the pallet passes every test and there is full confidence in its design.

For anyone planning on putting any type of equipment on a rocket these tests are required. Without performing these tests one can expect failure. The preliminary analytical and computer analysis are not required but are still helpful for a first timer to obtain a good idea of what to expect from their sine sweep test and obtain a good understanding of what the sine sweep is designed to achieve. Overall, all these tests will prove to be useful in the design of the test pallet and are a must for anyone doing aerospace testing.

References

1. VIBRATION ANALYSIS FOR ELECTRONIC EQUIPMENT BY DAVE S. STEINBERG

2. Vibration Analysis for Electronic Equipment:

3. Department of Defense Test Method Standard for Environmental Engineering Considerations and Laboratory Tests:



4. Random Vibration:



5. European Users Guide to Low Gravity Platforms:



6. Fundamentals of Electrodynamic Vibration Testing Handbook:



CONTACT

I AM AN UNDERGRADUATE MECHANICAL ENGINEERING STUDENT AT COLORADO STATE UNIVERSITY. I SPENT TWO MONTHS WORKING AT NORTHROP GRUMMAN SPACE TECHNOLOGY. SOME TIME WAS SPENT IN THE ENVIRONMENTAL TESTING LAB WORKING ON SHOCK AND THERMAL TESTING. I AM CURRENTLY WORKING AS THE DYNAMICS PERSON ON THE TEST PALLET TEAM AT COLORADO STATE UNIVERSITY.

e-mail: jeffwyan@engr.colostate.edu

Definitions, Acronyms, Abbreviations

A: CROSS-SECTIONAL AREA

E: modulus of elasticity

G: Acceleration due to gravity or 9.81 m/s2

JAXA: Japanese Aerospace Exploration Agency

L: length

m: mass

mo: mass per unit length

PSD: Power Spectral Density

ro and ri: cross-sectional outer and inner radii

SCCM: Standard Cubic Centimeter per Minute

W: weight

wo: weight per unit length

-----------------------

Heater coil

Hollow cathode

Tantalum wire

Magnesium oxide insert

Tantalum housing

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download