This is only a preview of the June 2008 issue of Silicon Chip. You can view 34 of the 104 pages in the full issue, including the advertisments. For full access, purchase the issue for $10.00 or subscribe for access to the latest issues. Items relevant to "DSP Musicolour Light Show":
Items relevant to "USB Power Injector For External Hard Drives":
Items relevant to "Balanced/Unbalanced Converter For Audio Signals":
Purchase a printed copy of this issue for $10.00. |
They go through
the pain – so we can survive
Photo courtesy
Denton ATD Inc.
Crash Test
Dummies
By Peter Holtham
Ever seen those video clips of car crash tests where the dummies
are thrown about like rag dolls? The dummies are highly
engineered to simulate the effect of crashes on human bodies and
they carry lots of instrumentation to record the pain (forces &
deflections) they suffer.
24 Silicon Chip
siliconchip.com.au
W
orldwide, car accidents kill about 400,000 people
and injure 12 million more every year. Despite
these grim statistics our roads are much safer
than they were 50 years ago.
The reason is simple. New cars are repeatedly crashtested by their makers, until they are safe as can be made
for the price. Inside almost every doomed car sits one or
more very expensive and very life-like anthropomorphic
test devices. You and I call them crash test dummies.
Packed inside each dummy are sensors that record the
accelerations, forces and movements felt by its head and
body throughout the crash. This data allows engineers to
see what happens to the driver and passengers millisecond
by millisecond. It enables them to pinpoint how particular
injuries occur.
Such complex capability did not appear overnight. Sixty
years of development has taken crash test dummies from
simple mannequins to today’s complex biomechanical
marvels.
Sierra Engineering built the first
test dummy, for the United States Air
Force, in 1949. Christened ‘Sierra
Sam’, he tested ejection seats in jet
aircraft. Weighing over 90kg, Sam was
not very life-like and so the air force
also used human volunteers. Strapped
into seats on a rocket-powered sled,
the volunteers experienced up to 45Gs
of deceleration while testing harness
designs and seating positions.
In the mid 1950s, General Motors
got to hear about the air force work.
Shortly afterwards, GM started to use
mannequin-like dummies in a simple
crash test program. They soon discovered that their dummies were not able
to survive repeated crashes. Nor would
apparently similar dummies behave
in the same way in similar crashes.
By the 1960s, road safety was becoming a big issue for US politicians.
In 1966 the United States Congress
passed an Act setting minimum safety
standards for new cars.
To help GM meet the standards, the
company started serious development
Hybrid III head and neck (photo
courtesy Denton ATD Inc.)
siliconchip.com.au
of a new crash dummy. They took components from two
crude dummies then commercially available and combined
them to create Hybrid I.
Hybrid I was a 50th percentile male dummy, meaning its
body (1.75m tall) and mass (77kg) represented an average
US adult male. Although much improved compared with
the early mannequins, Hybrid I was still not particularly
life-like.
GM kept up the research and development effort. In
1972 they introduced Hybrid II, with improved shoulders,
spine and knees. By careful calibration they standardised
its design and performance. Then, rather than keeping the
knowledge locked away inside the company, GM made the
drawings and calibration data freely available to anyone.
GM introduced a third generation, the Hybrid III, in 1976.
Now for the first time crash test dummies had a scientific
foundation. Hybrid III’s builders used biomechanics, the
study of how a human body responds mechanically to
impact, to guide their design.
Foot accelerometer and X-axis
potentiometer wiring (source THOR
documentation).
The dummy family
Hybrid III formed the basis for a
whole family of dummies. Next came
a petite 5th percentile female, followed
by her ten, six and three-year old
children. Reflecting the super-sizing
of America, there is also a 1.88m tall,
101kg 95th percentile male.
As with Hybrid II, GM made the
plans and calibration data for Hybrid
III publicly available. Several companies worldwide now manufacture
Hybrid IIIs to the GM specification.
In 1991 the International Standards
Organisation (ISO) adopted Hybrid III
as the standard crash test dummy for
frontal impact testing.
Whatever their size, all Hybrid III
dummies are built in the same way.
Each consists of over 300 component
parts. The skin for the head, arms, and
legs is made from pink coloured vinyl
plastic while the flesh is made from
urethane foam.
The vertebrae of the neck are made
up of rubber and plastic disks sand-
Wire routing and strain relief, THOR
main bundle of instrumentation wires
(source THOR documentation)
June 2008 25
Table 1: Hybrid III Sensors
Location
Type
Amplitude
Channels
Head
Accelerometer
250G
3
Neck
Load cell
Rotation
14kN
290Nm
3
3
Chest
Accelerometer
Deflection
150G
100mm
3
1
Pelvis
Accelerometer
150G
3
Thigh (L & R)
Load cell
20kN
2
Knee (L & R)
Deflection
19mm
2
Lower leg top (L & R)
Load cell
Rotation
12kN
400Nm
4
4
Lower leg bottom (L & R)
Load cell
Rotation
12kN
400Nm
4
4
wiched between steel rings. The consistency of the disks
is carefully controlled so that the structure mimics the
rotation, stretching and bending movements of the human neck.
The upper body has six high-strength steel ribs with
polymer-based damping material. This arrangement simu-
lates how a human chest responds to the crushing forces of
an impact. The lower body has a curved cylindrical rubber
spine; typical of a person slouched in their seat. The pelvis
is an aluminium casting fixed in a sitting position.
A ball-jointed thigh bone mimics human hip to upper
leg movement and rotation. Knee, lower leg and ankle
movements are all reproduced.
Feeling the pain
If you have any empathy at all you will shudder when
you see the impacts that crash test dummies are exposed
to. And they do the feel the pain. Or at least they have
electronic sensors which register the forces which would
cause extreme pain if the dummy was alive.
Dummy manufacturers supply little or no instrumentation themselves, just the spaces where it can be fitted.
Sensors are supplied by specialist instrument companies
and are selected and fitted for a particular crash test.
Hybrid III dummies have four different types of sensor
built in, as shown in Table 1.
Load sensors record the forces on different body parts
during a crash, while rotation sensors measure twisting
moments. Load and rotation sensors are built into the thigh
and shin bones, for example.
Accelerometers are fitted all over the body to measure
acceleration in a particular direction. The head has accelerometers for three directions: front to back, side to side
and up and down.
Female Hybrid III with her two children on the back seat (photo courtesy Denton ATD Inc.)
26 Silicon Chip
siliconchip.com.au
Movement sensors record deflections during a crash. A
linear potentiometer is fitted inside the chest to measure the
amount of compression caused by a seat belt for example.
Another, called a ‘knee-slider’, is used to measure forces
transmitted through the dummy’s knees, particularly if
they hit the lower facia.
No instrumentation is built into the arms. In a head-on
crash the arms flail around uncontrollably but serious
injuries are rare and worthwhile protection is difficult to
achieve.
Table 2: THOR Sensors
Location
Type
Channels
Head
Accelerometer
Tilt sensor
9
1
Face
Load cell
5
Upper neck
Load cell
6
Lower neck
Load cell
6
Front neck
Load cell
1
Recording the data
Rear neck
Load cell
1
Data from the sensors is of no value if it cannot be
recorded for later analysis. In the early days of crash testing, data logging systems were too bulky to fit in the car.
Umbilical cables connected the few instruments in the
test car and the dummy to a remote data recording system.
Data was stored as analog signals on tape for later playback
and analysis.
Carmakers do not want to fill up test cars with bulky or
heavy instrumentation as it might affect the outcome of
the crash. Yet they want all the data they can possibly get
from the crash test dummies, as well as any other sensors
mounted on the car.
Companies specialising in crash test data loggers now
produce on-board units with as many as 96 data channels.
Mass per channel is less than 150g and sampling rates
reach as high as 22000 samples per second. These rugged, battery-powered on-board data loggers amplify, filter,
digitise and store in flash memory all the signals from the
dummy’s sensors. Multiple data acquisition units can be
daisy-chained together when the test car is carrying a family of dummies.
The whole data logging system is cabled to a laptop
Head rotation
Potentiometer
1
Chest
Accelerometer
Deflection
1
12
Upper abdomen
Accelerometer
Deflection
1
1
Lower abdomen
Deflection
6
Spine
Accelerometer
Load cell
Tilt sensor
2
1
4
Pelvis
Accelerometer
Load cell
1
8
Thigh (L & R)
Load cell
2
Knee (L & R)
Deflection
Rotation
2
2
Load cell
Accelerometer
8
2
Load cell
Ankle rotation
Leg accelerometer
Foot accelerometer
12
6
2
6
Lower leg top (L & R)
Lower leg bottom, ankle
and foot (L & R)
Location of sensors in THOR 50th percentile
male dummy (source THOR documentation).
siliconchip.com.au
June 2008 27
computer while it is programmed for a specific test. Once
the test is set up, the cable is removed and the car is ready
to be crashed. Data from the sensors is recorded from the
moment the car starts moving until it comes to rest after
the crash.
Not content with this level of miniaturisation, data loggers are now moving from on-board to in-dummy. The loggers can be connected together by Ethernet through a central
hub. There is a single cable from the dummy for network
communication, trigger, and off board power if necessary.
In-dummy batteries allow the dummy to run completely
cable-free during a test. Each sensor is cabled through the
dummy in small channels in the flesh and spine.
What of the future? Hybrid III is now over 30 years
old and is beginning to show its age. It does not measure
injuries to the abdomen, there is only a single chest deflection measurement, and its leg bones are rigid. It is just
not sensitive enough to crash test modern cars fitted with
seatbelt pre-tensioners, seatbelt load limiters and multistage airbags.
The latest dummy
Development of a new front impact test dummy called
THOR (Test device for Human Occupant Restraint) started
in the late 1990s.
Designing THOR involves the efforts of carmakers, research groups and governments worldwide.
The current version of THOR was released in late
2001.
A spine and pelvis that allows it to sit in different poBelow right: the
WorldSID 50th
percentile male. By
contrast, the CAD
image opposite
is WorldSID 5th
percetile dummy
(images courtesy
of WorldSID Task
Group).
28 Silicon Chip
sitions is just one of THOR’s
many improvements. Its face
has five load sensors to measure facial injuries while its
rib cage measures deflections
in four places compared with
Hybrid III’s one.
Three deflection sensors are
fitted inside the abdomen to assess
soft tissue damage. The legs have
bushings to simulate the elasticity
of real bone. Up to
21 sensors on the
leg bones measure
loads, accelerations
and ankle rotation.
These additional sensors require THOR to
have as many as 134
data channels, four
times as many as
Hybrid III (Table 2).
Despite the millions spent on its
d e s i g n , T H O R ’s
drawings and operating manual can be
downloaded by anyone
from the US Department of Transportation
website. See www-nrd.
nhtsa.dot.gov/departments/nrd-51/thornt/
THORNT.htm
Different dummies
Not all crashes are
siliconchip.com.au
THOR head components (source THOR documentation).
frontal impact. Many are side impacts (T-bones!) caused by
drivers running red lights. Because the injuries are different, the requirements of dummies for side impact tests are
different. Carmakers need to measure the risk of injury to
the ribs, spine and internal organs such as the liver. Head
and neck injuries are also common and carmakers need
crash test data for head airbag development.
Hundreds of scientists and engineers from over 45
organisations worldwide have just spent eight years and
US$14 million designing WorldSID (World Side Impact
Dummy).
Companies from the Netherlands, France and Britain
designed WorldSID’s head, neck and pelvis while companies from the USA developed the rib cage, spine, arms and
legs, as well as the sensors and data loggers.
The result is the most life-like crash test dummy ever
created. WorldSID’s 212 built-in sensors record accelerations of the head, upper and lower spine, shoulder, ribs,
pelvis and arms. It also logs compression of the shoulders
and individual ribs, as well as rotation of the head, torso,
pelvis and ankles.
But whether it’s one of the Hybrid III family, a THOR
or a WorldSID, the procedure for using a dummy remains
identical. International standard test protocols are followed exactly.
Technicians first assemble the dummy, carefully testing
and calibrating each individual ‘body part’. They dress
the dummy in shorts, a short-sleeved shirt and shoes, and
precisely position it in the car.
They stick yellow and black adhesive targets to the sides
of the head to serve as reference points for the crash films.
The eyebrows, nose, chin, knees and lower legs are painted
with patches of different colours. Any contacts with the
car during the crash will then show up as coloured smears.
Once the instrumentation in the dummy and the car has
been checked, the test can start. The test bay is flooded
with light and high-speed film cameras start up. A tow
cable pulls the car towards the crash barrier at 64km/h (40
mph). Just before impact the cable is released and the car
smashes into the barrier.
In just 10 seconds the test is over, although the megabytes
of data recording the dummy’s ‘injuries‘ will take weeks to
SC
analyse. Ultimately, the result is safer cars for us all.
Radio, Television & Hobbies:
ONLY
the COMPLETE
00
$ 62
archive on DVD
&P
+$7 P
• Every issue individually archived, by month and year
• Complete with index for each year
• A must-have for everyone interested in electronics
This remarkable collection of PDFs covers every
issue of R & H, as it was known from the
beginning (April 1939 – price sixpence!)
right through to the final edition of R, TV
& H in March 1965, before it disappeared
forever with the change of name to
Electronics Australia.
For the first time ever, complete and in one
handy DVD, every article and every issue is covered.
If you're an old timer (or even young timer!) into vintage radio, it
doesn't get much more vintage than this. If you're a student of
history, this archive gives an extraordinary insight into the amazing
breakthroughs made in radio and electronics technology following
the war years. And speaking of the war years, R & H had some of the
best propaganda imaginable!
Even if you're just an electronics dabbler, there's something here to
interest you.
NB: Requires a computer with DVD reader to view
– will not work on a standard audio/video DVD player
the handy
handy order
order form
form
Use the
on
page
57
of
this
issue
on page 81 of this issue.
siliconchip.com.au
June 2008 29
|