Lecture 14 – Human-System Interfaces

1
 Fundamental Truth
• Almost every design you will be involved
with at one point or another will interface
to people
• It is the one part of your design over which
you have the minimum amount of control
• In most cases, you have to design to
accommodate the human component
because you can’t change it
• As such, it is best to understand the nature
of humans as component in a system
2
 Human Engineering Issues
• Anthroporometry – how big are the humans you need to work with
– Classic problems – seat design in aircraft, cars, trains, buses, etc…
– Spacesuit design, other worn equipment design
– Task design
• Limits –
– what kind of loads can humans sustain without injury
• Example Touch limit –
• What kind of loads can humans impart to systems
• Support – what kinds of support do humans need – how much oxygen, water, food, what kinds of
wastes do they generate
• Interfaces – what are good and bad interface practices in workstation and habitability design
• NASA STD 3000 – Human Systems Interface Standards contains a wealth of knowledge about
this topic… it is located at
• http://msis.jsc.nasa.gov/
• It is also located….
http://standards.nasa.gov/public/public_detail.taf?Documents_uid1=1609&doc_name=NASA-ST
D-3000%20VOL%20I
• Mil-Std-1472F is the DOD standard on Human Engineering and it is also a great resource . It is
located at
• http://www.weibull.com/knowledge/milhdbk.htm
• There is much commonality between Mil-Std-1472f and NASA STD-3000
– The Mil-Std is more general and the NASA STD-3000 contains more spacecraft specific
knowledge
• The FAA also has a human factors design standards for commercial and general aviation aircraft
– http://hf.tc.faa.gov/hfds/
• The Society of Automotive Engineers also maintains human factors requirements
– 200 standards as of July 2010
3
Example – Touch
Temperature

4
Example of
Anthroporometry

5
6
7
8
9
10
 So you know the 95% american male dimensions, is
this enough ?

– No – if you build a model human with all 95%

dimensions , it would look like Frankenstein’s
monster because no one is 95% in all
dimensions
– What you have to do is look at each design
problem and determine the parameters which
are critical to the design and ensure that it can
support 95% percentile in those critical
parameters
– When you need an entire model person , for
example for a crew occupancy study, you build
a composite model from all of the critical
components that results in a body solution that
is within the 95% height
• May actually need several models – a long
legged, a short legged long torso, etc…

11
X-38 cabin
with
reference
95%
american
male
models

12
X-38 cabin with
reference 95%
american male
models
13
Crash test dummies with a 95% american male, a 5% american
female and ten, six and three year old dummies. From 14
wikipedia.com
Reach and Visibility Limits

15
 John Paul Stapp and determining human G limits

When he began his research in 1947, the aerospace conventional wisdom was a man would suffer fatally
around 18 g. Stapp shattered this barrier in the process of his progressive work, experiencing more
"peak" g-forces than any other human. Stapp suffered repeated and various injuries including broken
limbs, ribs, and miscellaneous traumas… In one of his final rocket-propelled rides, Stapp was subjected
to 46.2 times the force of gravity (wikipedia)

One of the bravest officers in the history of the United States Air Force. His work convinced the industry
to put crash provisions in the aircraft to protect occupants up to demonstrated survivable limits
16
He also championed automobile safety and is one of the primary people responsible for seat belts in cars
today
Acceleration Limits

17
 Redout and Blackout During Accelerated Flight

http://www.wellnessoptions.ca/html/1/011/1_i11_2_main.html 18
Reference coordinate system
X axis – through center of chest and back
Y axis - through right and left arm
Z axis – feet to toes

19
Rate limits

Impact
Acceleration
Limits

20
I couldn’t resist – Vince and Larry – from wikipedia

Seriously – crash test dummies are a vital part of testing in

aerospace engineering to measure rates and accelerations
in crash or impact scenarios. Their use is slowly being
replaced by computer modeling but they are still used for
verification tests. – picture of dummy and computer model
from First Technology
http://www.ftss.com/pcat/products.cfm?pcat=h3-
95m&class=1&x=500&y=905 21
 In most vehicle impacts

• There are actually three impacts

– The impact of the vehicle to an obstruction
– The impact of the human occupant to the
internals of the vehicle
– The impact of the humans organs (heart lung,
kidneys, spleen, brain etc… )to the internal
walls of the body such as the rib cage or skull

22
23
 Fourier’s Theorem and impact loads
• In mathematics, a Fourier series decomposes a periodic function or
periodic signal into a sum of simple oscillating functions, namely sines and
cosines (or complex exponentials).
• Fourier series were introduced by Joseph Fourier (1768–1830) for the
purpose of solving the heat equation in a metal plate.
• A single impact, like a hammer strike, can be modeled as the sum of a
series of sine waves
– Thus an impact can be viewed as having high and low frequency components
– an object being struck receives vibrations across a wide set of frequencies
– An object being struck will ring if its own resonant frequency is excited by the
frequencies resident in the impact
– This is how a bell rings !
• The natural frequency of the bell is determined by the bell shape and determines the
tone of the bell
• Human beings ring like a bell when struck by an impact also
• The ringing can cause structures to tear away from each other and cause
fatal damage
• This is why people with apparently little exterior impact damage can
actually have fatal internal damage

24
 Fourier square wave – How a pulse can be made from a sum of sine waves

N=0 N=3

N=1 N=4

N=2 N=20

25
Human
Body
Resonant
frequencies

26
 Brinkley Model

• The “gold standard” in aerospace engineering

for evaluating impact and dynamic events on
humans is the Brinkley model
• Developed at Wright Patterson Air Force Base it
is a dynamic response model which takes a time
history of acceleration data and determines
human organ response and compares it to limits
for tissue damage
• Based on a comprehensive study of human
survival with ejection seats and animal test data
• For NASA, we add conservatism to the model to
account for the deconditioning that occurs due to
weightlessness

27
 X-38 Subscale Landing – 100 lbs

X-38 Full Scale Flare and Landing – 25,000 lbs 28

 X-38 Human Tolerance Design Analysis
• Primary objective of the analysis: to assess the effect of landing impact on a deconditioned
crew.
• Analysis Math Model:
– sanctioned by the NASA-STD-3000 documentation.
– developed by Wright Patterson AFB for ejection seat human tolerance analysis.
– extensive discussions held with model developers to certify usage for landing impact
analysis.
– inputs are filtered high impact accelerometer data from a hard-mounted instrument not
taking into consideration any couch damping effects.
– currently assumes the crew sits in the vehicle with feet forward in the vehicle nose and
back against the bottom of the vehicle.
– Injury-Risk Criteria = function of dynamic response compared to limits based on compiled
medical data for a deconditioned crew.
– If IRC is greater than one, then the risk category is the next level:
• low risk level (less than 0.5% chance of incapacitating injury)
• moderate risk level (0.5% to 5% chance of incapacitating injury)
• high risk level (5% to 50% chance of incapacitating injury)
 X-38 Human Tolerance Design Analysis:
Example FF1 Impact Data

• Peak filtered g’s in the vertical and horizontal: 9 and 8 respectively

• Landing rating: LOW RISK or less than 0.5% of incapacitating injury to a
deconditioned crew.
• Implies there is less than 0.5% chance of incapacitating injury.
• FF1 was the first flight test with a ranking of low risk.
-Z Port IST Data for IMPACT
(Vehicle Axes)
10

Raw data
Landing Impact Results Based on Filtered Port IST Data
8
Filtered w/cutoff freq = 100 Hz Time of Time of
Description Maximum Maximum Minimum Minimum Risk
Measured Linear Acceleration (G)
6 X Axis 8.9225 1.083 -0.9953 1.503
Y Axis 1.273 1.084 -0.5341 1.009
-Z Vehicle Axis Acceleration (g)

Z Axis 7.941 1.091 -0.4921 1.001

4 Resultant 11.0168 1.084 0.0047 0.888
Dynamic Response
X Axis 8.3578 1.118 -0.8678 1.464 low
2 Y Axis 0.8449 1.114 -0.2796 1.403 low
Z Axis 8.1785 1.129 -0.5199 2.87 low
Injury-Risk Criteria
0 Low Risk 0.5854 1.127 0 0
Moderate Risk 0.4968 1.127 0 0
High Risk 0.3987 1.127 0 0
-2
0 0.5 1 1.5 2 2.5 3
Duration (sec)
Duration 0 sec =~14955.34 sec ops time
 X-38 Human Tolerance Design Analysis
Brinkley Model Inputs: deconditioned crew; IST data filtered as per Brinkley specs; crew backs parallel to pallet, feet
D ro p
forward.
Im p a ct Im p a ct V e l. (fp s) P e a k g 's (filte re d ) D yn a m ic R e sp o n se g 's ** In ju ry R isk C rite ria *

T e st T e ch n iq u e V e rt. H o riz .*** V e rt. H o riz . V e rt. H o riz . H e a lth y Ill/In ju re d

P 2D 4 part. flare 23 30 27 14 32 15 m oderate high
P 2D 6 N o flaps 26 65 51 18 46 17 > high > high
P 2D 8a 50% flaps 25 16 40 9 40 9 high > high
P 2D 11 50% flaps 24 16 31 10 38 12 high > high
part. flare (33%
P 2D 13 flaps ) w ith trees 21 73 18 15 20 16 m oderate high
L0/R 20; 5th S tage
P 2D 14 F ailed. 27 65 41 16 47 17 > high > high
P 2D 15 L49/R 45% flaps 21 51 21 17 21 18 high > high
P 2D 16 L0/R 28% flaps Unknow n Unknow n 47 17 55 17 > high > high
P 2D 17 50% flaps 22 62 51 20 43 15 > high > high
P 2D 20 55% F laps 25 37 46 13 38 7 high > high
P 3D 1 L70/R 60% 17 56 9 8 9 8 low low
P 2D 21 98% F laps 21 50 32 14 30 7 m oderate high
P 2D 22 F lare 14 54 11 9 12 8 low low
P 2D 23 ~ L80/R 70% flaps 23 46 25 8 29 7 m oderate m oderate
P 2D 24 F lare 15 39 11 9 10 8 low low
P 2D 25 ~ L0/R 50% flaps 17 58 16 10 19 9 low low
P 2D 26 ~ L0/R 50% flaps 15 70 13 9 8 9 low low
P 3D 2 E arly F lare 18 25 6 6 6 7 low low
P 3D 3 F lare 10 41 12 4 5 2 low low
P 3D 4 F lare 8 59 12 2 9 1 low low
P 2D 28 E arly F lare 18 39 22 8 23 7 low low
aft IS T: 35 8 31 7 low m oderate
P 3D 5 E arly F lare 15 39 21 7 12 6 low low
* high risk level (5% to 50% chance of incapacitating injury)
moderate risk level (0.5% to 5% chance of incapacitating injury)
low risk level (less than 0.5% chance of incapacitating injury)
** peak acceleration of the human body response to impact loads (human body modeled as a spring-mass-damper system)
*** wrt ground
32
Vibration limits

33
34
Avoiding Cut Hazards

35
Electrical Shocks

36
37
38
39
 Alarm Classes
• In general, three classes of alarms are used in aerospace systems to
notify the crew of hazardous or anomalous conditions
• Class 1 – Emergency – reserved for life threatening situations that
require immediate attention
– Classic examples – loss of cabin pressure, fire/smoke, ground
proximity, toxic atmosphere, impending collision (TCAS), stall
warning
– Each class 1 alarm typically has a different sound as well as visual
indications
• Class 2 – Warning – conditions that require immediate correction to
avoid loss of mission or crew
– Classic examples – hardware malfunctions, critical system
operating limits, critical consumable limits (fuel)
– Typically all warnings have a common aural tone and may have
individual visual annunciations
• Class 3 – Caution – conditions of a less time critical nature but that
require crew attention to avoid degradation into a life threatening
situation
– Typically all cautions have a common aural tone and they may
have a common visual indicator with a requirement for the crew to
look up additional information to determine the exact cause of the
alarm
40
Any questions ?

41