1
INTERNATIONAL CONFERENCE ON ENGINEERING DESIGN
ICED 03 STOCKHOLM, AUGUST 19-21, 2003
HUMAN POWER; COMFORTABLE ONE-HAND CRANKING
Arjen Jansen, Peter Slob
Abstract
Research into ergonomics is one of the aspects in the research for human-powered energy
systems. In this specific field, data on maximum force exertion and endurance can be found in
a large number of publications, mainly originating from sport or military related research.
Data on comfortable or sustainable force exertion however prove not to be available. In this
research project we attempted to measure comfortable/sustainable force exertion. We mapped
one specific movement (one-handed cranking) using the Critical Power test. This test is based
on the assumed linear relation between maximal work and time to exhaustion (Morton’s 3-
parameter critical power model). The experimental set-up consisted of an altered cycle-
ergometer which was adjustable in height. We measured the subjects' (eight young males)
maximum power output and the time to exhaustion at different power levels. The research
showed a sustained power output from cranking to be: 54 ± 14 Watt (mean ± SD). In the
paper we will present the research project and its results and link them to literature in the field
of comfort.
Keywords: critical power test, comfort, human power
1. Introduction
1.1 Why human-powered products?
Nowadays we see a growing number of portable electric consumer products, mainly powered
by batteries. Examples are; audiovisual, communication and information products, in which
the electronics provide the main functionality, but also an increasing number of products that
deliver mechanical work at their output. Considering the clear advantages of rechargeable
batteries (high energy density, wide availability and international standardization), they will
remain the main source of power in the forthcoming period. Nevertheless, the use of batteries
can be cumbersome as well. Batteries run out of energy when you need them most, they’re
not always available, they have to be replaced or charged in a troublesome way and in the
long run batteries turn out to be a rather expensive power source. Moreover, due to the
increasing number of battery-powered portable products, the environmental impact of battery
use might increase as well. Driven by consumer perception and environmental concern, the
Personal Energy System (PES) group at DUT aims at finding alternatives for the increased use
of batteries in portable energy products. In this scope the PES-group focuses on the application
of renewable energy sources in consumer products. Special emphasis is given to low power
2
energy sources such as; human power (i.e. the use of human work for energy generation),
direct methanol fuel cells, and photovoltaic solar cells.
In this paper we will focus on human power.
In our research project we defined human power as; using the human body as an energy
source for electric products. The main advantages of human powered products are; they
operate independent of energy infrastructures, have a long “shelf-life” and can be
environmentally beneficial in the long run. Some examples of human powered products on
the market nowadays can be seen in the next figure.
Figure 1. Some examples of human powered product. From left to right; the Seiko Kinetic watch, the Coleman
radio, the Philips AE 1000 radio and the Aladdin Power hand generator
In the research for human power we identified three areas of scientific interest: the
environmental impact of human-powered vs. battery-powered products, the engineering of
small -but efficient- generators and human factors of power input. This paper will discuss the
latest area, specifically one-hand cranking.
1.2 Comfort and measuring comfort
Acceptance of human-powered products, i.e. prolonged and repetitive use of the product, is
only possible when discomfort –inevitable associated with the use of these products- can be
limited. Definitions of (dis)comfort vary from ”comfort is the subjective positive perception
of the nature and intensity of the load, resulting from using or operating an object” [1],
“discomfort is the result of bodily pains, arising as a result of the postures and effort
involved” [2], “discomfort is a phenomenon of perception, related to pain, fatigue, and
perceived exertion”. Discomfort can be divided into short term and long term discomfort [3].
Short term discomfort (our focus) can be measured by rating or ranking the subjects feelings
[4] and observing/registering of body posture, movement and force (OWAS and RULA method)
[5]. From literature we also concluded that discomfort is associated with pain and fatigue.
Kroemer [6] describes a number of methods to determine fatigue. Out of these methods,
measuring the output is the most practical way to assess (strain) and fatigue, defined as “…the
inability to maintain power output…”[7]. In our research we found one method aimed at
measuring sustained force exertion, the Critical Power test. It has been used for the
determination of the power output sustained for several hours in synergic muscle groups [8],
total body work [9] and several other purposes. The CP-test (critical power test) has been
validated by comparing it to the ventilatory threshold and the physical work capacity at
fatigue threshold [9].
3
The CP-test has been used in earlier research [10] on cranking with three trained paraplegic
subjects (mean ± SD = 36 ± 9 yrs). In this test, the shoulder was in line with the axis of the
ergometer and the wheelchair was placed so the subject’s arm was fully extended when the
crank handle was at its greatest distance. We suspect this was one-hand cranking although this
is not clear from the literature. The power outputs used were 25, 37.5 and 50 W. The test
ended immediately when the subject was unable to maintain a cranking rate of 50 RPM. Rest
periods between each period of exercise continued until the heart rate returned to a range
within 10 BPM of the subject's resting heart rate. Each test was repeated three times in
different sessions. Linear regression was expressed by the following equation;
W
lim
= AWC + CP⋅t
lim
(1)
W
lim
work value [Joule] AWC anaerobic work capacity [Joule]
t
lim
time to exhaustion [seconds] CP critical power [watt]
For subject 1: W
lim
= 5905 + 22⋅t
lim
, Subject 2:W
lim
= 6384 + 24⋅t
lim
and subject 3:W
lim
= 5722
+ 21⋅t
lim
This means that subject 1 has a Critical Power of 22 W , subject 2 of 24 W and
subject 3 of 21 W [10].
2. The Critical Power Test
2.1 Problem statement and research question
In the field of ergonomics, data on (maximum) static and dynamic force exertion and
endurance can be found in a large number of publications, mainly originating from sport or
military related research. In our search we could not find specific data on ‘comfortable’ or
‘sustainable force exertion’, nor standardized measurement methods to determine these
values. We assume the critical power test to be the best available alternative. Most estimations
on long-term static force exertion are based upon percentages, varying from 15 to 20%, of the
maximum strength [6] [11]. We assumed an identical relation between P
max
and CP in order to
estimate the time to exhaustion (t
lim
).
This leads to the definition of the following research question; is it possible to quantify the
“sustainable-comfortable power output” from one-hand cranking, and if so; what is its value
and variation for a specified population? [12]
2.2 Materials and methods
Subjects; the subjects in the pilot study (n=2) and the main study (n=8) were healthy young
men, age from 19 to 26 years. Materials; we used an adapted bicycle ergometer (Lode RH30)
(see figure 2) in which the pedal was replaced by a crank handle (diam. 25 mm, length
95 mm). The crank arm length was 175 mm (fixed). The ergometer was mounted on a
hydraulic lift in order to adjust the crank height to fit the subjects’ anthropometric
measurements. The resistance (power) of the ergometer could be varied in between 5 to
250 watt. The ergometer featured a speedometer (analogue dial in RPM) and an analogue
output. The analogue output was connected to a writing recorder (Kipp, BD 41) in order to
log the cranking rate.
4
A photoelectric pulse plethysmograph (Boucke, Infraton Kardio PF100) was used to measure
the heart rate (in BPM) every ten seconds during the warm-up, test period and cool-down. We
used a video camera to record the tests. Time was measured using a hand-held stopwatch.
Figure 2. Measurement set-up (altered ergometer) and relevant anthropometric values (NB no research subject)
Test procedure; The arm cranking ergometer (see figure 2.) was secured to a hydraulic lift in
order to keep the heart in line with the crank axis. The subject was seated in a normal chair so
the subject's arm was almost fully extended when the crank handle was at its greatest distance.
A speedometer was observed by the subject to maintain the prescribed cranking rate (60 RPM)
throughout the test. Cranking rate was recorded continuously and the heart rate was measured
at 10-s intervals using a photoelectric pulse plethysmograph. The main test was preceded by a
pilot study in order to gain more insight in a number of variables. Here we learned the initial
instruction “stop cranking if you feel pain or when cranking becomes uncomfortable” proved
to be to vague. It was altered into “keep cranking until you’re unable to maintain the cranking
rate of 60 RPM”. All tests were preceded by a 2-min warm-up at 5 W, with increasing cranking
rate until 60 RPM and followed by a 2-min cool-down at 5 W. Not more than two tests on one
day, with a rest period of at least three hours between tests.
2.3 Measurements protocols
Protocol CP-test; The subjects performed three tests in which the power output remained
constant and led to the onset of muscle fatigue. The appropriate power output was set within
2-3 s. The moment the power output was set, the stopwatch and recorder were activated. The
test was ended when it lasted longer than 30 minutes or in case the constant power output
level could no longer be sustained (i.e. cranking rate drops below 55 rpm, determined by the
written output from the recorder.
Protocol for maximal power output (P
max
); The initial power output was set to 5 W, after 10
seconds an increase of 5 W and subsequent increases of 10 W every 10 seconds. The test
ended when the constant power output level could no longer be sustained, i.e. a drop in
cranking rate below 55 rpm (determined by the written output from the recorder). The test
was followed by a 2-min cool-down at 5 W. The P
max
-test was done twice: one before and one
after the Critical Power Test.
90
64
23
5
Protocol comfort test; The CP was validated by a comfort test. The appropriate power output
was set within 2-3 s after starting to crank, then the stopwatch and recorder were activated.
Every two minutes the subject gave his rating for the perceived exertion. The test was ended
after 30 minutes; or when the constant power output level could no longer be sustained, i.e. a
drop in cranking rate below 55 rpm. The comfort test was done once.
2.4 Results main study
The results of the study consist of; P
max
, P
max
at re-test, W
lim
(according to Morton’s 3-
parameter critical power model [13]) and the comfort test (rating perceived exertion at Borg
scale [4]).
Table 1. Results of main study
Subject P
max
[watt] P
max
re-test
[watt] W
lim
(=AWC + CP.t
lim
) [Joule]
A 118 129 (+9%) 4655 + 35 t
lim
B 119 157 (+32%) 5053 + 55 t
lim
C 142 141 (1%) 3257 + 78 t
lim
D 109 120 (+10%) 8249 + 50 t
lim
E 139 159 (+14%) 12.596 + 54 t
lim
F 128 168 (+31%) 5096 + 51 t
lim
Main 126 146 (+16%) -
0
20
40
60
80
100
120
140
160
180
200
0 300 600 900 1200 1500 1800
Endurance (s)
Power (W)
Subject A Subject B Subject C
Subject D Subject E Subject F
Figure 3. Endurance