The EMPF recently conducted a survey on High-G testing methodology. This survey was part of an ongoing program to improve the manufacturing of Inertial Measurement Units (IMUs) that incorporate MEMS sensors. This survey gathered information covering two areas: 1) High-G (~15,000g's) packaging and/or high-G electronics issues and needs, and 2) long term storage testing methods. The High-G testing methodology survey investigated 7 test methods. From the responders, the five most common test approaches were rail gun, air gun, drop test, centrifuge test and pneumatic shock test. The remaining 2 tests in the survey were live fire and "others".

The survey revealed that no single test replicates the complete dynamic environment of the potential end-use. It should be pointed out that the real application, fire launching, generates a shock spike with a High-g and a long duration including balloting acceleration that is difficult to duplicate in a "lab environment". A brief introduction to the main five test methods is given here.

1) Rail gun:
The Rail Gun is a howitzer that fires a projectile into the middle of a rail system that descends gradually into a trough filled with water. The projectile is fitted with a concave "scoop" on the front that impacts the water and "slowly" decelerates the round in the space of about 100 feet. In this test, the test articles are placed in the projectile and launched. Payload is limited to approximately 12-13 kg that is placed in the mid-body of the projectile or similar cargo round, and provides a 15 to 20" long, 5" diameter volume in which the test item sits during the shot. setforward & setback forces tend to be on the high side due to the nature of the test event. While an advantage of the rail gun test is that it is a live-fire gun launch test with realistic setback forces applied to the test articles, its primary disadvantages are the large amount of balloting as the projectile engages the rails, and the uncontrolled nature of the test. Depending on the particular gun, the frequency content, balloting, and setforward forces can vary greatly. The rail gun test is quite expensive, usually about $10,000 per day (up to 4 shots).

2) Air gun:
The air gun fires a cylindrical "piston" by building up pressure around it and then giving it a push from behind at a pressure of about 1000psi. The samples are placed in a test fixture that in turn is loaded into a piston. The piston is then loaded into an air gun breach for firing. The breach can be pressurized to different levels for testing. In order to accomplish setforward acceleration, the test fixture is inverted in the piston. The amount of pressure around the piston is proportional to the g level while the weight of the piston is inversely proportional to the g level. The payload volume is smaller than in the rail gun for this system, but g's are correspondingly higher. The setback level can get up to about 15 kg, but with a shorter duration than the rail gun. setforward is roughly 10% of setback, with balloting usually being very low. Some air guns use a diaphragm method with 2" and 5" diameters. In these guns, pressure is again built up around the piston, while the piston is held in place by a ring with a groove cut into it. The groove depth equates to a breaking point so that when the pressure reaches the right level, the ring breaks at the groove and the piston travels down the tube. Air gun test shots are about $1500 each.

3) Drop test
In our survey, all responders stated that they are using a drop test as a first-level of screening because the method is very handy and economical. Commercial drop test manufacturers claim that they can generate up to a 30,000g acceleration using a shock amplifier (figure 5-3), but the pulse duration is less than 1 millisecond. The free fall drop test is designed to produce classical shock wave shapes. The drop height may be set at pre-determined heights for shock pulse repeatability. Acceleration levels are generated by the drop height while duration time is a function of the programmer provided elastomer pads for half-sine, lead pellets for sawtooth and pneumatic cylinders for square wave pulses. For some accelerated fall drop testers, a bungee cord is available. The equipment is supplied with brakes to prevent repeated impacts, and incorporates a massive steel reaction mass mounted on springs with dampers to isolate the shock from the floor. The shock amplifier consists of a secondary shock table, precision guides fitted with extension springs and elastomer spacer/dampers, and a base. In use, the base is bolted to the top of the main table of a shock test machine, and the test specimen is fixed to the secondary shock table. The main table and the shock amplifier base impact and rebound, while the shock amplifier table continues downward, stretching the spring. The primary advantages of the drop test are low cost, ease of test, and safety. The primary disadvantage is the short duration of the shock.

4) Centrifuge
The centrifuge test can offer many different centrifugal testing parameters. The primary purpose of the centrifuge is to apply an elevated centrifugal force to a test sample and to determine whether it will withstand a given g-force without failure, operate satisfactorily at some g-level to which the device may be subjected in actual use or to accomplish some unique function at a given g-level and to test destruction or failure. The centrifuge is a rotor that is spun at high speeds with test articles mounted at precisely measured radii in the rotor, which allows for the calculation of the g-level applied to the test articles. The primary advantages of the centrifuge include the ability to control the direction and magnitude of the acceleration forces applied to the test articles and has fewer restrictions on the size of the test article. The main disadvantage is that centrifuge testing applies pseudo-static loading, and therefore provides none of higher frequency energy that is present in a dynamic gun-launch environment. The centrifuge is used extensively to verify mechanical models and predicted deflections of all assemblies under high-G loads.

5) Pneumatic Shock
The pneumatic shock test can subject small MEMS devices to mechanical shocks of up to 50,000g. The apparatus uses compressed gas to accelerate a small sled carrying the test articles down a long tube to impact a fixed target. The system is capable of delivering calibrated shocks from 2,000 to 50,000g, and can be instrumented with accelerometers and computer controls. The machine produces a shock pulse in the vertical direction using compressed air to force the carriage to impact on the shock machine base. The basic structure of the machine is a heavy steel weldment filled with reinforced concrete, which will not deteriorate under repeated shocks. The structure contains sufficient mass so that no additional ballast is required. The structure is supported on passive air springs with dampers to isolate the shock from the floor. The test item is mounted to a solid carriage with heavy duty steel inserts in the top surface. The carriage is supported and guided by the lifting and driving piston rod. Friction brakes are used as a rebound brake and as a quick release device. A microprocessor control is available to provide a single point for shock machine set up.

Summation:
The only test that approximates the magnitude and duration of the shock pulse produced by a gun launch is the Rail Gun. However this test is quite expensive and a single test station can only perform up to four tests per day; so it cannot be used as a simple screening test. Besides the centrifuge, the other tests produce shock pulses that are of adequate magnitude, but are not nearly as long in duration as an actual gun launch. The drop test is very commonly used screening test due to its low cost and simplicity. This type of testing is required for any electronics modules that would be used in a high shock environment.