Part 22 of the vibration and shock calibration standard covers “Shock calibration by comparison to a reference transducer”. Compared to the general accelerometer calibration techniques discussed last month, shock calibration is a
specialty method. As such, there are a number of accepted apparatus and methods depending on the desired shock acceleration magnitude and pulse width.
Uncertainty reference conditions for secondary shock calibration (from ISO 16063-22:2005(E))
|Shock apparatus||Accel peak magnitudea (km/s2)d||Minimum pulse durationa,b (ms)||Uncertainty limit (%)|
|Hop Bar velocity||100c||0.050||10|
|Hop Bar accel||100c||0.050||6|
|Split Hop Bar force||100c||0.050||10|
a Variations in peak values and duration = +/-10% b Pulse duration is measured at 10% of the peak value c Larger accelerations (peak values) and shorter pulse durations are possible but without reference
to primary methodologies. d 100 km/s2 = ~10,000 g
The standard explains the basic mechanism of each apparatus and the various data acquisition and processing methods.
The most common embodiment for commercial shock calibration in the broad useful range of a few 0.1 km/s2 (~10 g) to 100 km/s2 (~10,000 g) is the pneumatically excited apparatus. The apparatus consists of pneumatically pressurized
reservoir coupled to a barrel housing a captive projectile which is fired against a selectable anvil placed in the other end of the barrel. Both the mass of the anvil and the cushioning material at the impact surface can be tailored along with
varying the reservoir pressure to develop shock pulses of various durations and amplitudes. The data acquisition is conventionally handled with high speed time data capture although some systems are now moving to a broadband FFT approach. The
benefits of the more mainstream pneumatic technique are: broad useful amplitude range, traceability to primary method, safe simplified operation, ability to easily tailor the impulse duration/magnitude, and reasonable repeatability. Modern implementations of the pneumatic shock calibration integrate seamlessly to standard accelerometer calibration workstations allowing for consolidated accelerometer calibration databases and easy use of standard functions, like print
calibration certificate, save, recall, trending, etc...
A Hopkinson bar is used for extremely high shock levels of up to 1000 km/s2 (~100,000 g). This is accomplished by either a split Hopkinson bar with an integral piezoelectric force sensor, or a single ended Hopkinson bar with strain gages
mounted in the middle or with either velocity or acceleration reference at the free end.
In a shock calibration via a split bar, a rigid piezoelectric force sensor replaces the material test specimen. Provided the impact is sufficiently long with the second bar being sufficiently short, the response of the second bar may be approximated
as rigid body motion. Thus its acceleration is calculated using ’s second law F=ma, where the force, F, is measured and m is the mass of the fly away transducer under test.
In the single ended embodiment of a Hopkinson bar accelerometer shock calibration, the test transducer may be calibrated in terms of velocity by comparing the integrated output of the transducer with either strain gauges or a laser Doppler vibrometer.
The test transducer may also be calibrated in terms of acceleration by comparing the output of the transducer with the derivative of the output of either strain gauges or a laser Doppler vibrometer. A single ended apparatus reduces the complexity
of the test system by eliminating the need for the integral force sensor and vacuum coupling to fixture the second bar of split apparatus. Data acquisition is normally a high speed time data acquisition system with peak value or polynomial curve
Recapping, shock testing via the Hopkinson bar is commonly used for generating extremely high acceleration levels of 1 km/s2 (~100 g) to 2000 km/s2 (~200,000 g) and is an extremely well documented method in the literature due to its long
history in mechanical specimen testing.
The dropball method basically consists of a magnetically mounted anvil carrying the test and reference sensor, which is then placed inside a tube to guide the drop ball. By varying the diameter and mass of the ball, as well the material
at the impact interface, the shock pulse can be customized in terms of amplitude and duration. Data acquisition is typically accomplished with high speed time data capture and analyzed with the peak amplitude method. Ranges in the 1 km/s2 (~100
g) to 10 km/s2 (~1000 g) are commonly tested with this method as in automotive crash applications. Dropball can be used to 100 km/s2 (~10,000 g). However, alignment is critical with the dropball technique;
if the ball hits off center large rotations occur.
Lastly, the pendulum method basically consists of a rigid frame, a hammer pendulum and an anvil pendulum carrying both the test and reference shock accelerometers. The hammer pendulum is lifted to a predetermined angle and then falls to
strike the anvil pendulum through a chosen rubber pad at the contact surface, creating a haversine pulse shape. Special considerations are made to locate the transducers at node points of the pendulums first axial mode of vibration, as well
ensuring that the center of gravity of the seismic mass of the test transducer is aligned with the sensitive axis of the reference transducer. The pendulous shock test is normally accomplished with a digital data acquisition system for time capture
and implemented with either a maximum value or polynomial time data fit as the analysis method. The pendulous shock calibration method is sometimes found in the automotive market.
By purchasing the ISO 16063-22 standard and thorough search of the internet, one can gain a very comprehensive understanding of the methodology. However, we know standards can sometimes be confusing and/or intimidating, but they don’t have
If you have any questions about accelerometer calibration or the standards please don’t hesitate to contact us at The Modal Shop, part of the PCB Group of companies. We are glad to help.