Ergonomics

ISSN: 0014-0139 (Print) 1366-5847 (Online) Journal homepage: http://www.tandfonline.com/loi/terg20

Measurement of prehensile grasp capabilities

by a force and moment wrench: Methodological
development and assessment of manual workers

Michael Greig & Richard Wells

To cite this article: Michael Greig & Richard Wells (2004) Measurement of prehensile grasp
capabilities by a force and moment wrench: Methodological development and assessment of
manual workers, Ergonomics, 47:1, 41-58, DOI: 10.1080/00140130310001611107

To link to this article: http://dx.doi.org/10.1080/00140130310001611107

Published online: 20 Feb 2007.

Submit your article to this journal

Article views: 166

View related articles

Citing articles: 15 View citing articles

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=terg20

Download by: [Punjab Engineering College] Date: 19 December 2015, At: 03:58
 ERGONOMICS, 15 JANUARY, 2004, VOL. 47, NO. 1, 41 58

Measurement of prehensile grasp capabilities by a force and

moment wrench: Methodological development and assessment
of manual workers

MICHAEL GREIG* and RICHARD WELLS

Department of Kinesiology, University of Waterloo, ON, Canada

Keywords: Force; Moment; Wrench; Instrumentation; Strength; Hand.

Downloaded by [Punjab Engineering College] at 03:58 19 December 2015

Prehensile grasp capability is typically quantied by pinch and grasp forces. This
work was undertaken to develop a methodology to assess complex, multi-axis
hand exertions through the measurement of forces and moments exerted by the
hand along and about three orthogonal axes originating at the grip centre; termed
an external wrench. Instrumentation consisting of a modied pinch/grip
dynamometer axed to a 6 df force cube was developed to simultaneously
measure three forces, three moments and the pinch/grip force about the centre of
the grip. Twenty right hand dominant manual workers (10 male and 10 female),
free of hand or wrist disorders, completed a variety of maximal strength tasks.
The randomized block design involved three separate grips power grip, lateral
pinch and pulp pinch. Randomized within each block were three non-concurrent
repetitions of isolated maximal force and moment generations along and about
the three principle orthogonal axes and a maximal grip force exertion. Trials were
completed while standing, with the arm abducted and elbow exed to 908 with a
wrist posture near neutral. Where comparable protocols existed in the literature,
forces and moments exerted were found to be of similar magnitude to those
reported previously. Female and male grip strengths on a Jamar dynamometer
were 302.6 N and 450.5 N, respectively. Moment exertions in a power grip
(female and male) were 4.7 Nm and 8.1 Nm for pronator, 4.9 Nm and 8.0 Nm for
supinator, 6.2 Nm and 10.3 Nm for radial deviator, 7.7 Nm and 13.0 Nm for
ulnar deviator, 6.2 Nm and 8.2 Nm for extensor, and 7.1 Nm and 9.3 Nm for
exor moments. Correlations with and between maximal force and moment
exertions were only moderate. This paper describes instrumentation that allows
comprehensive characterization of prehensile force and moment capability.

1. Introduction
Characterization of an individuals prehensile capacities has predominantly
examined single axis, or uni-component, exertions such as grip force or pinch force.
These values are often used as a normalization method to classify relative task
demands (e.g., dividing grip force requirement by maximal grip force attainable or
population average). This methodology is applicable for some task demand
predictions but not for the majority of tasks since they are multi-axial (Wells and
Greig 2001). Developing methods to characterize the force and moment magnitudes

*Author for correspondence. E-mail: magreig@ahsmail.uwaterloo.ca

Ergonomics ISSN 0014-0139 print/ISSN 1366-5847 online # 2004 Taylor & Francis Ltd
http://www.tandf.co.uk/journals
DOI: 10.1080/00140130310001611107
 42 M. Greig and R. Wells

encountered in occupational settings and activities of daily living is important since

no task can be completed without some type of force generation. In addition, force
has a key role as a causative mechanism in work-related musculoskeletal disorders
(Silverstein et al. 1987, Moore et al. 1991, Bernard 1997, NRC 2001).
Force magnitudes can be directly or indirectly determined during occupational
tasks or activities of daily living, with various degrees of precision and invasiveness,
by knowledge of object mass (Buchholz 1988), estimation (Armstrong 1982,
Speilholz et al. 2001), force matching in an unpublished study by one of the
authors, electromyography (Buchanan et al. 1993, Duque et al. 1995), or
instrumentation of the tool being used (Stoy and Aspen 1999).
Normative population data for grip strength capabilities has been determined
using a myriad of instruments. These have ranged from handgrip dynamometers
Downloaded by [Punjab Engineering College] at 03:58 19 December 2015

(Mathiowetz et al. 1985, Martin vigrometers (Desrosier et al. 1995), and strain gauge
dynamometers (An et al. 1986) to the more complex transducer based arrangements
that can measure the contributions of individual digits to the total grip force (Chao
et al. 1989, Li 2002). Similarly, the measurements of force or moment creation
magnitudes have utilized unidirectional instrumentation, multi-axis transducers to
measure exertions in more than one motion (Buchanan et al. 1993) and specially
created jigs (An et al. 1986, Deivanayagam and Sethi 1993, Snook et al. 1997). The
authors are not aware of work that has been done combining all of these apparatuses
to collect grip force, force and moment data simultaneously.
The majority of demand estimations (stress placed on the internal structures of
the forearm and hand) created by prehensile tasks are almost always determined
from the normalization to grip strength. Although this may be appropriate in
selected tasks, such as gripping an object, tasks with increasing complexity (for
example gripping and manipulating an object) are not well characterized (Wells
and Greig 2001). If gripping an object or squeezing a pair of pliers, grip force as a
percentage of the maximum force generation possible would estimate that demand
accurately but, in a task such as holding a hammer, operating a screwdriver, or
turning a key in a lock, the role of the grip has changed. The grip function then is
to eectively stabilize the object and provide a means for force and moment
transmission from proximal segments, with capacities depending on the grip
chosen. One approach to address this issue is by determining the functional
strength capabilities on a range of common objects (Peebles and Norris 2003). The
authors of the present study recently proposed that characterizing the external task
by a force and moment wrench, which also includes the internal grip force
generation, would better characterize strength during functional tasks (Wells and
Greig 2001). Wrench classication describes the forces and moments as a 6 6 1
vector applied at a point (gure 1). If this information were to be coupled with
muscle activation measurements of the forearm and hand, a more comprehensive
understanding of the demands placed on the individual in prehensile tasks would
be possible.
This paper reports on the development of instrumentation to measure the force
and moment wrench in three distinct grasps and, as a demonstration of its
applicability, also documents the strength capacities of manual workers. It will
therefore build on the information gathered by other researchers regarding
individual strength capacities, providing insight into the magnitudes of forces and
moments an individual can generate both singly and in combination. Correlating
these force and moment magnitudes to those measured on power and pinch grip
 Measurement of prehensile grasp capabilities 43
Downloaded by [Punjab Engineering College] at 03:58 19 December 2015

Figure 1. Depiction of wrench classication at the grip centre (Wells and Greig 2001), where:
X (+) are palmar and dorsal forces, Y (+) are pull and push forces, Z (+) are radial and
ulnar forces, Mx (+) are radial and ulnar deviator moments, My (+) are pronator and
supinator moments, and Mz (+) are exor and extensor moments.

dynamometers will illustrate the applicability of using these grip strengths to predict
task demands for an individual.
A full characterization of the prehensile strength capacities of the working
population will enable increased comprehension of the acceptable ranges within
which individuals could perform manual work for both the design and evaluation of
work.

2. Methods
2.1. Force and moment collection methods
Grip forces were obtained from a strain gauge dynamometer and amplier (Model
ST-GRIP, Medical Research Ltd., Leeds, UK). For the power grasp, two half pieces
 44 M. Greig and R. Wells

of plastic pipe were added to the forks of the grip dynamometer, making the
diameter (62 mm), similar to that found in many tools. For the power grasp, the long
axis of the dynamometer was aligned vertically (gures 2 and 3). For the lateral and
pulp pinch grips, metal tabs 20 6 50 mm were added to the end of the forks. The
dynamometer orientation for the lateral pinch (gripping with thumb opposed by side
of rst and second ngers) was horizontal. The setup for the pulp pinch (gripping
with thumb pad in opposition to pads of rst and second ngers) was chosen to
maintain the same wrist and forearm posture for all three grips. To accomplish this,
the long axis of the dynamometer was in the horizontal plane but aligned at a 458
angle (gures 2 and 3). Grip dimensions were 28 mm and 27 mm respectively for the
Downloaded by [Punjab Engineering College] at 03:58 19 December 2015

Figure 2. Schematic of cube and grip dynamometer arrangement and orientation for all
three grips.
 Measurement of prehensile grasp capabilities 45
Downloaded by [Punjab Engineering College] at 03:58 19 December 2015

Figure 3. Three grips used in the current study: power, lateral pinch and pulp pinch.
 46 M. Greig and R. Wells

lateral and pulp pinch. The contact surfaces of the dynamometers were covered by
new white athletic tape (Sportstape, Athletics Products Inc., Schomberg, ON,
Canada) for each participant to maintain surface consistency within and between
participants.
A 6 df multi-axis force and torque sensor (cube) (AMTI MC3A-6-250,
Watertown, MA, USA) with amplier (AMTI SCA-6-4, Watertown, MA,
USA) was used (gure 2). This allowed the measurement of three orthogonal
forces and moments. Calibration of the force and moment output was done using
a matrix provided by the manufacturer to minimize cross-talk. Verication of the
matrix occurred before the study commenced. Seven channels of information were
collected from the transducers at a sampling rate of 1024 Hz, A/D converted and
stored for further processing. A shunt calibration was used to convert the grip
Downloaded by [Punjab Engineering College] at 03:58 19 December 2015

force data into Newtons and was veried prior to study commencement. In each
of the grips, forces acting on the grip dynamometer created a moment about the
centre of the cube (gure 2). To measure moments about the grip centre, a
moment correction was required.
Since the grip dynamometer is rigidly mounted to the force cube and has
instrumentation on only one of its forks, a misrepresentation of the grip force could
be created. For example, in the power grasp setup, the instrumented fork would
register a force if a proximally directed or pulling force was created even when no
force was exerted across the two forks of the grip dynamometer. As a result the grip
force would be overestimated. This was corrected by taking the lesser of the two
values (registered force on the instrumented fork vs the calculated magnitude (from
the cube) on the non-instrumented fork) which was chosen to represent the internal
grip force since it was assumed that the larger value was augmented by an external
force application to the grip dynamometer.
To handle the collection and feedback of all data channels, an interactive software
program (LabVIEW 5.1, National Instruments Corporation, Austin, TX, USA) was
created. The program was driven by a le containing the randomized single and
multi-axial force exertions. Information regarding the upcoming task was displayed
to give a description to the participant. Force and moment data were collected,
zeroed, and calibrated in real-time as described and displayed as bar charts on the
computer monitor for participant feedback.

2.2. Participant population

Twenty individuals (table 1) were recruited from an industrial temporary employ-
ment agency. All had manual work experience, were right hand dominant and free of
forearm, hand and wrist disability. Individuals who had incurred trauma to the
forearm or hand region within the previous year were excluded from the study. Past
work experience included assembly line (automotive, barbeque, packaging jobs),
brick laying, tree planting, service industry work and warehouse work. Informed
consent was obtained from all individuals prior to study commencement using a
procedure approved by the Research Board of Ethics, University of Waterloo.

2.3. Collection protocol

Anthropometric information about the participant included three maximal grip
strength values collected on a commercial grip dynamometer (Jamar, Model 2A).
Grip span was set at the second notch to compare with the study population
reported by Mathiowetz et al. (1985).
 Measurement of prehensile grasp capabilities 47

Table 1. Participant anthropometric information.

Height Weight Age Grip Strength
(m) (kg) (years) (N)
Females (n = 10) mean 1.59 70.6 33.4 302.6
SD 0.07 24.4 11.9 42.8
min 1.50 41.1 19.0 210.9
max 1.72 115.5 57.0 372.8

Males (n = 10) mean 1.78 82.3 32.7 450.5

SD 0.07 15.2 11.3 65.8
min 1.72 66.3 19.0 348.3
max 1.94 106.2 58.0 564.1
Downloaded by [Punjab Engineering College] at 03:58 19 December 2015

Grip strength reported as maximum exertion on Jamar dynamometer.

As this was a subset of a larger protocol, the following explanation focuses purely
on those trials that were included to examine maximal strength capacities. The grips
examined were the power grasp, lateral pinch and pulp pinch. The order of
measurement of the three grips was randomized within a subjects trials and
balanced between the male and female participants. Randomized within each grip
were 13 possible actions. These included force and moment exertions along and
about the three principal orthogonal axes in both the positive and negative directions
(see gure 1 for clarication of actions and their axes). Maximal grip or pinch force
exertions were exerted as well. Specically, the force actions examined were maximal
push and pull, palmar and dorsal and radial and ulnar directed forces, and pronator
and supinator, exor and extensor, and radial and ulnar deviator moments.
Following description of the task, the participant was instructed to gradually increase
their level of exertion up to their maximum and maintain a constant exertion for 3 s.
Visual feedback was given of all forces and moments produced. The participant was
instructed to minimize the non requested actions as much as possible. Throughout the
active portion of the collection, the individual was standing with their feet comfortably
spaced approximately shoulder width apart. Their orientation with respect to the cube
and dynamometer setup allowed the participant to have their arm adducted against their
side with elbow exed to a 908 degree angle and their wrist positioned midway between
pronation and supination with slight extension. During the inactive portion of the
protocol, the participant was allowed to move around or be seated if they desired.
Post collection, ve males and ve females were recruited as volunteers to examine
the frictional properties of the interacting surfaces. Three trials to determine the
coecient of friction were completed. In these trials the participant attempted to
maintain a 40 N pulp pinch grip while pulling horizontally in line with the long axis
of the grip dynamometer until their hand slipped free (a modication of the protocol
of Buchholz et al. 1988).

2.4. Manipulation of data

Voltage output from the ampliers was A/D converted and stored as a raw le at the
end of collection. Simultaneously, all data had the bias removed, was calibrated and
stored as a converted le. Channel biases and the dynamometer calibration value
were determined from the average of separate one second collections. Maximum
values were determined from the largest magnitude of a one second moving average
window within the three second collection period.
 48 M. Greig and R. Wells

For the trials to determine the coecient of friction, the point of slipping was
determined from a manual inspection of the converted grip force and pull force data.
Coecient of friction was predicted using the equation

m L=2  F

where m = coecient of friction, L = pulling force (N), F = grip force (N) (Bucholz
et al. 1988).
Converting the data into spherical coordinates for the forces and moments
allowed the assessment of the isolation of the exertion. The ideal relationship was 1.0
meaning the exertion matched the requested force or moment action.
Downloaded by [Punjab Engineering College] at 03:58 19 December 2015

3. Results
The gain and amplication settings within the setup for this study allowed a
resolution of 1.2 N grip force, 0.2 N for horizontal forces (Fx and Fy), and 0.75 N
for vertical forces. Moment resolution was under 0.015 Nm for the x and y axes and
approximately 0.010 Nm for moments about the z-axis. Application of the white
athletic tape to the grip dynamometer surface created a surface coecient of friction
between the interacting surfaces averaging 0.72 (SD + 0.13).
Strength magnitudes were greatest in the power grasp for all actions in males and
for 12 of 13 actions for females (tables 2 and 3). With the exception of female lateral
pinch exertions (n = 9), all exertions had the potential of 10 male and female
participants. For dorsal and palmar force directions in lateral and pulp pinch and for
radial and ulnar force directions in lateral pinch, the maximal forces exceeded the
instrumentation range. As a result, these actions were capped at a magnitude to
prevent damage to the instrumentation. Therefore maximal force exertions in these
directions are not reported in this paper.

3.1. Gender dierences

Female strength was 66% of that of the males (288.1 N vs 436.3 N) in the
commercial hand grip dynamometer grip strength test preceding the start of the
protocol. Female strength was 67.3% of males across all grips with the two
populations closest in magnitudes for lateral pinch extension (88.4% 0.8 Nm vs
0.9 Nm). Pulling force (81.3% 92.4 N vs 113.6 N) using a power grip was the only
other action with a ratio over 80%. The lowest ratios were seen in the isolated power
grip force exertion (55.9% 162.3 N vs 290.1 N) and the lateral pinch pronator
moment (49.1% 1.9 N.m vs 3.8 Nm). Across an action for all grips, the pronator
moment had the lowest ratio (58.1%, 49.1%, 58.6%, power grip, lateral pinch, pulp
pinch, respectively).

3.2. Isolation of force and moment direction

The best isolated action, meaning that minimal to no force or moment was generated
on any axis other than that requested, was found in radial force for the lateral pinch
(99% of resultant magnitude from desired action; shown in table 4). Best isolation in
power grasp and pulp pinch was found in ulnar deviator moment (98%) and ulnar,
radial, or pull force (94%), respectively. Conversely the lateral pinch had on average
the most poorly isolated actions of all the grips in the maximal exor moment task
(50%) with only slight improvement in extensor moments (53%). In the pulp pinch,
radial deviator and supinator moments were the poorest isolated actions (63%),
Downloaded by [Punjab Engineering College] at 03:58 19 December 2015

Table 2. Average maximal force generated in isolated force exertions and the female to male strength ratio for three dierent grip types in 10 males and
10 females.
Force (N + SD)

Measurement of prehensile grasp capabilities

Grip Push Pull Dorsal Palmar Radial Ulnar
Power grasp
Female 162.3 (43.6) 70.6 (24.9) 92.4 (35.2) 59.0 (22.4) 76.7 (32.7) 118.8 (32.3) 101.4 (22.3)
n 10 10 10 10 10 10 10
Male 290.1 (63.3) 113.6 (31.6) 113.6 (19.3) 74.4 (17.2) 87.1 (15.0) 194.6 (49.1) 161.2 (48.3)
n 10 9 9 9 9 10 10
Female/Male 0.56 0.62 0.82 0.76 0.83 0.61 0.63
Lateral pinch
Female 64.8 (13.2) 73.6 (21.0) 69.5 (18.0) 40.8 (9.1) 47.3 (4.2) 107.5 (8.9) 62.2 (10.8)
n 9 9 9 6 4 6 9
Male 89.1 (13.3) 104.9 (21.1) 96.6 (21.2) 56.0 57.8 NA 70.9 (12.3)
n 10 10 10 1 1 0 5
Female/Male 0.73 0.70 0.72 0.75 0.84 0.70 0.71
Pulp pinch
Female 85.0 (24.0) 73.1 (21.4) 77.0 (24.5) 39.9 (8.5) 42.4 (6.4) 63.7 (12.1) 56.5 (12.2)
n 10 10 10 10 4 10 10
Male 107.3 (27.6) 96.4 (19.0) 100.2 (23.9) 44.0 (10.3) 42.2 (1.7) 101.9 (11.7) 75.1 (17.4)
n 10 10 10 4 2 7 9
Female/Male 0.79 0.76 0.59 0.68 0.85 0.58 0.71
**Note: Shaded areas indicate deviations from full population due to transducer limits, participant withdrawal or technical diculties.

49
 50 M. Greig and R. Wells

Table 3. Average maximal isolated moment exertions generated about the centre of the grip
and the female to male strength ratio for three dierent grip types in 10 males and 10 females.
Moment (Nm + SD)
Radial Ulnar
Pronator Supinator Deviator Deviator Extensor Flexor
Power grasp
Female 4.7 (1.7) 4.9 (1.6) 6.2 (2.7) 7.7 (1.9) 6.2 (1.8) 7.1 (2.3)
n 10 10 10 10 10 10
Male 8.1 (2.3) 8.0 (1.9) 10.3 (3.5) 13.0 (4.2) 8.2 (1.7) 9.3 (2.0)
n 10 10 9 9 8 8
Female/Male 0.58 0.61 0.59 0.57 0.68 0.69
Downloaded by [Punjab Engineering College] at 03:58 19 December 2015

Lateral pinch
Female 1.9 (0.6) 2.2 (1.0) 2.0 (0.6) 1.1 (0.2) 0.8 (0.3) 0.9 (0.4)
n 9 9 8 8 8 8
Male 3.8 (1.1) 3.5 (0.8) 3.0 (1.1) 1.7 (0.5) 0.9 (0.3) 1.3 (0.6)
n 10 10 9 9 9 9
Female/Male 0.49 0.62 0.70 0.55 0.78 0.69

Pulp pinch
Female 1.3 (0.4) 1.9 (0.7) 1.1 (0.6) 2.0 (0.9) 1.8 (0.7) 1.5 (0.5)
n 10 10 9 9 10 10
Male 2.2 (0.4) 2.4 (0.8) 1.7 (0.3) 2.6 (0.9) 2.6 (0.8) 2.1 (0.6)
n 9 10 9 8 10 10
Female/Male 0.59 0.77 0.64 0.75 0.69 0.64

Table 4. Ratio of resultant exertion magnitude to target force/moment.

Power Grip Lateral Pinch Pulp Pinch
Action Average Maximum Average Maximum Average Maximum
Force
Dorsal 0.87 1.00 0.81 0.98 0.71 1.00
Volar 0.85 0.99 0.91 1.00 0.73 0.99
Push 0.72 0.94 0.78 0.94 0.87 0.99
Pull 0.95 1.00 0.94 1.00 0.94 1.00
Radial 0.97 1.00 0.99 1.00 0.94 1.00
Ulnar 0.97 1.00 0.97 1.00 0.94 1.00
Moment
Pronator 0.74 0.99 0.93 1.00 0.83 1.00
Supinator 0.86 1.00 0.91 0.99 0.63 0.85
Radial Deviator 0.94 1.00 0.90 1.00 0.63 0.98
Ulnar Deviator 0.98 1.00 0.76 0.99 0.82 0.98
Flexor 0.93 1.00 0.50 0.98 0.91 1.00
Extensor 0.81 1.00 0.53 0.95 0.81 1.00

whereas pronator moment (74%) and push force (72%) were lowest in the power
grasp.

3.3. Correlations between forces and moments

Table 5 shows the correlations between the maximum force and moment magnitudes
measured. The correlation between the commercial hand grip dynamometer and the
 Measurement of prehensile grasp capabilities 51

power grip on the measurement device was 0.83 in the power grip. The relationship
between the grip force from the commercial hand grip dynamometer and maximal
lateral and pulp pinch was lower (0.59 and 0.73, respectively). The power grip had
higher correlations (table 5; seven of 12 actions 4 0.70) between the force or moment
exertions and the grip strength from the commercial dynamometer. Correlations
were inconsistent for a given action between the force measurement tools.

4. Discussion
The advantage of the equipment used within this study is that it simultaneously
collected all forces and moments exerted by the hand during a task. This
methodology has not been reported in the literature, nor has a complete data set
with these three grips and this number of actions for one person been published.
Downloaded by [Punjab Engineering College] at 03:58 19 December 2015

Numerous uniaxial analyses have occurred, using a variety of equipment

arrangements and participant limb orientations. These have focused most frequently
on the power grip (for example An et al. 1986). In general, moments are less well
reported. Within the present study the axes were dened at the centre of the grip. In
most studies within the literature, the origin is placed at the wrist axis. This would
not create dierences between the force magnitudes, but would often result in larger
moment values at the wrist due to the dierence between the point of force
application and the wrist axis of rotation.
The collection of maximum exertions, both moments and forces, allowed the
identication of which actions are capable of the greatest exertion magnitude. The
rank order of the magnitudes showed a similar trend across all forces, with radial
exertions consistently near the top whereas volar and dorsal exertions were
consistently the lowest. Conversely, the strength hierarchy in moment exertions is
more dependent on the grip utilized, with moderate consistencies between
genders.
It was expected, based upon the literature, that the strength magnitudes for all
actions would be greatest in the power grasp and lowest in the pulp pinch. However,
the two pinch grips were very close in magnitude to each other. The magnitude of
push force was nearly identical for all grips. This was attributed to two factors. First,
pushing is the action least dependent on the frictional interaction of the grip
apparatus and hand. Secondly, the action of force is in line with the long axis of the
forearm, running close to the wrist axis of rotation meaning that a low moment
about the wrist would be generated. This would require less wrist muscle activation
than is necessary for other actions to both create a force and maintain the wrist
position. Therefore the magnitude generated would be limited by factors external to
the hand and forearm complex such as shoulder strength or balance restrictions.
The reported strength magnitudes are the average of the highest individual
maximums as opposed to the average of all magnitudes within an exertion direction.
The rationale for this was based on the assumption that this would report the most
isolated action. Also, since these measures were a subset of a quite extensive
protocol, the potential for reporting values that had been adversely aected by
fatigue would be minimized.
Coecient of friction values were higher than magnitudes found by other
researchers. Buchholz et al. (1988) found adhesive (sports) tape to have coecients
of friction of 0.41 and 0.66 for dry and moist conditions, respectively, compared to
0.72 for this study. The work by Buchholz et al. (1988) used a constant grip force
created by the thumb and index nger, with increasing load magnitude, whereas the
 52
Downloaded by [Punjab Engineering College] at 03:58 19 December 2015

Table 5. Correlation between maximum force or moment exertion magnitudes, grip force or maximum exertion on a Jamar dynamometer for the power
grip (18 to 20 samples per cell).
Forces Moments

Jamar Power Grip Radial Ulnar

Dynamometer Force Push Pull Dorsal Volar Radial Ulnar Pronator Supinator Deviator Deviator Flexor Extensor
Jamar Dynamometer 1.00 0.83 0.81 0.57 0.51 0.38 0.86 0.87 0.73 0.85 0.72 0.85 0.46 0.66

M. Greig and R. Wells

Power Grip Force 1.00 0.83 0.55 0.55 0.38 0.91 0.82 0.86 0.85 0.70 0.80 0.70 0.57

Forces
Push 1.00 0.81 0.79 0.77 0.88 0.84 0.83 0.86 0.77 0.85 0.69 0.81
Pull 1.00 0.86 0.94 0.61 0.67 0.55 0.60 0.59 0.65 0.46 0.75
Dorsal 1.00 0.87 0.66 0.64 0.62 0.60 0.78 0.75 0.52 0.70
Volar 1.00 0.44 0.52 0.45 0.45 0.56 0.53 0.42 0.66
Radial 1.00 0.86 0.91 0.89 0.81 0.90 0.73 0.73
Ulnar 1.00 0.78 0.78 0.76 0.86 0.67 0.75
Moments
Pronator 1.00 0.81 0.82 0.87 0.72 0.58
Supinator 1.00 0.78 0.90 0.70 0.65
Radial Deviator 1.00 0.89 0.65 0.66
Ulnar Deviator 1.00 0.63 0.78
Flexor 1.00 0.74
Extensor 1.00
 Measurement of prehensile grasp capabilities 53

participants in this study created the force by pulling on the dynamometer. A slightly
dierent style of grip was also used. The higher coecient within this study could
also be due to slight bleeding of the tape adhesive through to the grip surface.
Nevertheless, within this study, the surface characteristics were consistent between
tasks and participants.
Female grip force magnitudes from the pretest collection were almost identical to
the age matched average found by Mathiowetz et al. (1985) for the power grip
(300.2 N vs 308.5 N). Conversely male values of this study were 87.4% of the
magnitude reported by Mathiowetz et al. (450.5 N vs 515.4 N). Participant body
positions were also dierent between the protocols. Those involved in the
Mathiowetz et al. study were seated, whereas those within this study were standing.
These posture dierences have been shown to cause strength dierences for a given
Downloaded by [Punjab Engineering College] at 03:58 19 December 2015

individual (Balogun et al. 1991) but unlike the relationship between the work of
Mathiowetz et al. (1985) and this study, grip strength values found by Balogun et al.
(1991) were 5.2% lower in the seated posture.
Power grip exertion magnitudes collected with the cylindrical dynamometer
surface were lower than hand grip dynamometer values seen elsewhere (Mathiowetz
et al. 1985, An et al. 1986, Peebles and Norris 2003). The grip dynamometer within
this study was modied to reect the shape and size found on power grasp tools, with
a grip diameter of 62 mm. Grip force has been demonstrated to have an optimum
diameter between 50 and 60 mm (Oh and Radwin 1993, Blackwell et al. 1999).
However, when examined for maximum total normal nger forces (Amis 1987) or
EMG and fatigue based optimal cylinder size (Ayoub and Lo Presti 1971), optimum
diameters are seen between 30 and 40 mm. Peebles and Norris (2003) used a grip
force measuring device similar to, but narrower than, the strain gauge dynamometer
used in the present study. At widths of 50 and 70 mm, their results from a similar age
range were much higher than the values from the hand grip dynamometer study, but
were quite similar to the values obtained by the Jamar dynamometer.
For the lateral pinch force, magnitudes were within the range of values seen by
other researchers. Swanson et al. (1987 as cited in Kamal et al. 1992) found values
that averaged 10 N lower for both males and females compared to values found
within this study. In contrast, the majority of authors found magnitudes greater than
100 N for males and approximately 75 N for females (Mathiowetz et al. 1985
(averaged for same age range), An et al. 1986, Imrhan and Alhaery 1994).
Mathiowetz et al. (1985) and Kamal et al. (1992) found female strength magnitudes
similar to the majority of authors, but their average results for six males was nearly
identical to the values in the present study.
Finding pinch force comparisons for the pulp pinch used within this study was
more dicult due to a lack of identical work and, at times, inconsistent terminology
in describing the specics of the grip tested. The pinch force magnitudes in this study
consistently exceed those of other researchers (An et al. 1986, Imrhan and Loo 1989)
but were within 7 N and 4 N for females and males, respectively, when compared to
the age matched population in the work of Mathiowetz et al. (1985). A pinch span of
27 mm was used within the present study compared to approximately 17 mm within
the study of Mathiowetz et al. (1985) or the span of under 16 mm of Imrhan and Loo
(1989). Pinch spans of 20 to 140 mm have been examined for the lateral pinch and
chuck pinch by Imrhan and Rahman (1995). The exertion magnitude ratio (chuck/
lateral) for the closest width tested (32 mm) was 1.2. In the present study, the ratio
for males was 1.2 and that for females 1.3.
 54 M. Greig and R. Wells

Flexor and extensor moments about the wrist during a power grip are generally
higher in the literature (for example Vanswearingen (1983) and An et al. (1986)).
Flexor magnitudes within this study were 10 to 14 Nm lower for males and 3 to
6 Nm lower for females. Dierences in extensor magnitudes were less extreme. The
average magnitude found by An et al. (1986) was 1 Nm less (5.2 Nm versus 6.2 Nm).
Female values for pulp pinch (Snook et al. 1995, 1999) were substantially higher
(6.13 Nm and 3.92 Nm vs 1.5 Nm and 1.8 Nm, for exor and extensor, respectively)
and showed greater dierences between exor and extensor than those found with
this protocol. Again this is likely a by-product of dierent arrangements and axis
denitions. References were not found for the lateral pinch exor and extensor
moment magnitudes.
Pronator and supinator moments are well reported in the literature. Male
Downloaded by [Punjab Engineering College] at 03:58 19 December 2015

supinator magnitudes excepted (An et al., 1986), previously cited strength

magnitudes were slightly lower (approximately 1 Nm) than pronator and supinator
magnitudes within the present study (Salter and Darcus 1952, An et al. 1986).
Pronator moments were very similar to supinator moments in all three grips.
Pronator values have been shown to be 90% of the supinator magnitude (Snell-
Massie et al. 1997) for three-jaw chuck and lateral pinch. Salter and Darcus (1952)
found that pronator and supinator values become equal at 88 of supination in the
power grip.
Radial and ulnar deviator moment magnitudes were highest in the power grip.
Snook et al. (1997) have examined ulnar deviator moments in the power grip for
females. Their values were slightly lower than the results in the present study.
Vanswearingen (1983) has also investigated radial and ulnar deviator strengths in the
power grip for males and females using a modied Cybex II machine. Contrary to
the ndings in this study, Vanswearingen (1983) found that radial deviator values
were greater than ulnar deviator values for males and females. Magnitudes collected
were also substantially higher and are likely reective of the dierences in protocol.
Unlike Vanswearingen (1983), An et al. (1986) found radial deviator magnitudes to
be 86% of ulnar deviator values for males and females, a result more similar to the
near 80% relationship between males and females in the population of the present
study.
Comparison to other studies examining the strength magnitudes created by
pushing or pulling shows a great dierence with the magnitudes seen in this study.
Most studies in the literature used substantially dierent protocols, usually tending
towards two-handed power grasps and extensively braced postures (Hennion et al.
1989, Kumar, 1995). MacKinnon (1998) has investigated the one hand maximal pull
exertion in xed standing, nding magnitudes of 154 N in a comparable posture.
Within the present study foot placement was constrained as a comfortable position,
approximately shoulder width apart with arm and elbow angles stipulated. In
MacKinnon (1998), the feet were placed at a constant width with no constraint on
the arm position. Peebles and Norris (2003), with participants in a free posture while
pulling with the dominant hand on a vertical bar at elbow height, observed pull
strength magnitudes 250 320 N greater than the age matched female and male
values collected in this study. Similarly, push strength magnitudes were 250 350 N
lower in the participant population of this study and are likely attributable to
postural constraints and equipment dierences. In pinch grips, Imrhan and
Sundararajan (1992) and Imrhan and Alhaery (1994) have made an extensive
examination of the maximal pull magnitude that can be created. Pulling strength
 Measurement of prehensile grasp capabilities 55

using a lateral pinch grip was similar to values within the present study (99.8 N vs
96.6 N). Using the grip they dened as chuck pinch, which appeared most similar to
the two nger pulp pinch in this study, gave values substantially lower (67 N vs
100 N). The pulp pinch values of this study were likely higher due to the oblique
angle of contact between the hand and grip dynamometer as opposed to the line of
force being in line with the long axis of the gauge. The actions used in the tests by
Imrhan and Sundararajan (1992) and Imrhan and Alhaery (1994) were also created
perpendicular to the line of action used here (i.e. pulling across the body versus fore
and aft). Chuck pinch pull strengths collected by Peebles and Norris (2003) were
nearly identical (76.8 N) for females in the 31 50 year age range at a width of
20 mm. Male values in this study were 8.3 N less in the same age range.
Downloaded by [Punjab Engineering College] at 03:58 19 December 2015

4.1. Correlations
Only moderate correlation magnitudes were found between an individuals
maximum force and moment generation and the grip force generated within the
grip on a grip force dynamometer. This gives little evidence for the existence of a
general hand strength construct. Therefore the grip force alone may not capture all
of the capabilities within the hand/forearm complex to generate forces and moments.
Frequently, the hand grip dynamometer has been used as a prediction of prehensile
strength, without consideration given to the directions of forces and moments
involved in the task. This would be an acceptable method if the strength capabilities
in the six axial forces and six axial moments of an individual correlated well with
their grip strength and, more specically, their strength measured via a grip
dynamometer (e.g., A Jamar Dynamometer). It is understandable that this
relationship is poorer between pinch strength and hand grip strength as the grip
used in a hand grip dynamometer is fundamentally dierent. However, relationships
within the power grip might be expected to be better.
Past literature examining the construct of whole body strength measures has
shown that correlations between strength measures are also moderate (Borchardt
1968, Baumgartner and Zuidema, 1972, Zuidema and Baumgartner 1974). These
studies have shown that a single strength measure is usually insucient to
characterize strength capacity in dierent types of exertion. The present study
supports the assertion that a grip strength measure only is insucient to characterize
strength of the forearm and hand.

4.2. Limitations
Although this study provided a comprehensive examination of the strength
capacities of experienced manual workers, limitations in the application and
interpretation of the results must be considered. A limitation was imposed by the
equipment for a small number of actions. Only participants who did not exceed this
level are reported in correlations and the maximum values are not compared to the
literature. Examination of the relative demand incurred on the hand as determined
through muscle activation and rating of perceived exertion for sub-maximal
exertions will be examined in a subsequent study.
Throughout the active portion of the protocol the individual was standing, with
their arm adducted against their side, elbow at 908 and feet comfortably spaced
apart. The strength values can then be considered reective of the potential
limitations in the proximal joint strength (e.g., shoulder) and balance aorded in this
posture.
 56 M. Greig and R. Wells

To ensure that the participants isolated strength capability for a desired action
was obtained, visual feedback was given of all channels that were collected. Within
the data collection process repeat attempts were given if the isolation of the action
was poor. Should these actions be considered functionally similar to motions of the
low back, a perfectly isolated trial would not be possible, since o-axis exertions are
present and increase with fatigue (Parianpour et al. 1988).

5. Conclusion
The methodology employed within this study allows exertion magnitudes along and
about the axes of an orthogonal system at the grip centre to be collected
simultaneously. Values are consistent with ndings of previous researchers where
sucient similarity in protocol exists, but are specic to the setup with the grip
Downloaded by [Punjab Engineering College] at 03:58 19 December 2015

coupling and proximal joints limiting magnitude generation in some tasks. Further
application of the methodology will enable the determination of the response to the
generation of sub-maximal external wrench magnitudes typical of a variety of grasp
types and tasks. Simultaneous collection of muscle activation levels and participant
perception of eort will then allow a more complete understanding of the demand
required in manual tasks. The presentation of this demand analysis will occur in a
future paper.

Acknowledgements
The authors would like to acknowledge the support of the Ontario Workplace Safety
and Insurance Board, Grants #98 0008 and #01 041. However the views and opinions
expressed are those of the authors.

References
AMIS, A. A. 1987, Variations of nger forces in maximal isometric grasp tests on a range of
cylinder diameters, Journal of Biomedical Engineering, 9, 313 320.
AN, K. N., ASKEW, L. J. and CHAO, E. Y. 1986, Biomechanics and functional assessment of
upper extremities, in W. Karwowski (ed), Trends in Ergonomics/Human Factors III,
(Amsterdam: Elsevier), 573 580.
ARMSTRONG, T. J. 1982, Development of a biomechanical hand model for study of manual
activities, in R. Easterby, K.H.E. Kroemer and D.B. Chan (eds), Anthropometry and
Biomechanics: Theory and Application (New York: Plenum Publishing Corporation),
183 192.
AYOUB, M. M. and LO PRESTI, P. 1971, The determination of an optimum size cylindrical
handle by use of electromyography, Ergonomics, 14, 509 518.
BALOGUN, J. A., AKOMOLAFE, C. T. and AMUSA, L. O. 1991, Grip strength: eects of testing
posture and elbow position, Archives of Physical Medicine and Rehabilitation, 72, 280
283.
BAUMGARTNER, T. A. and ZUIDEMA, M. A. 1972, Factor analysis of physical tness tests, The
Research Quarterly, 43(4), 443 450.
BERNARD, B. 1997, A Critical Review of Epidemiological Evidence for Work-Related
Musculoskeletal Disorders of the Neck, Upper Extremity, and Low Back, US Department
of Health and Human Services, National Institute of Occupational Safety and Health,
Cincinnati.
BLACKWELL, J. R ., KOMATZ, K. W. and HEATH, E. M. 1999, Eect of grip span on maximal grip
force and fatigue of exor digitorum supercialis, Applied Ergonomics, 30, 401 405.
BORCHARDT, J. W. 1968, A cluster analysis of static strength tests. The Research Quarterly,
39(2), 258 264.
 Measurement of prehensile grasp capabilities 57

BUCHANAN, T. S., MONIZ, M. J., DEWALD, J. P. and ZEV RYMER, W. 1993, Estimation of muscle
forces about the wrist joint during isometric tasks using an EMG coecient method,
Journal of Biomechanics, 26, 547 560.
BUCHOLZ, B., FREDERICK, L. J. and ARMSTRONG, T. J. 1988, An investigation of human palmar
skin friction and the eects of materials, pinch force and moisture, Ergonomics, 31,
317 325.
CHAO, E. Y., AN, K. N., COONEY, III W. P. and LINSCHEID, R. 1989, Hand functional strength
assessment and its clinical application, in Biomechanics of the Hand: A basic research
study (Singapore: World Scientic Publishing Co. Pte. Ltd), 97 129.
DEIVANAYAGAM, S. and SETHI, N. 1993, Eects of wrist deviation on hand torque strength and
the potential for CTS, in W. S. Marras., W. Karwowski, J. L. Smith and L. Pacholski
(eds), The Ergonomics of Manual Work (London: Taylor & Francis), 265 268.
DESROSIER, J., HEBERT, R., BRAVO, G. and DUTIL, E. 1995, Comparison of the Jamar
Dynamometer and the Martin Vigorimeter for grip strength measurements in a healthy
Downloaded by [Punjab Engineering College] at 03:58 19 December 2015

elderly population, Scandinvian Journal of Rehabilitation Medicine, 27, 137 143.

DUQUE, J., MASSET, D. and MALCHAIRE, J. 1995, Evaluation of handgrip force from EMG
measurements, Applied Ergonomics, 26(1), 61 66.
HENNION, P. Y., MOLLARD, R. and COBLENTZ, A. 1989, Human force exertion: The signicance
of the measure? In: Proceedings of the Human Factors Society 33rd Annual Meeting,
(Santa Monica: The Human Factors Society), 1201 1205.
IMRHAN, S. N. and ALHAERY, M. 1994, Finger pinch-pull strengths large sample statistics, in
F. Aghazadeh (ed.), Advances in Industrial Ergonomics and Safety VI, (London: Taylor
and Francis), 595 599.
IMRHAN, S. N. and LOO, C. H. 1989, Trends in nger pinch strength in children, adults, and the
elderly, Human Factors, 31(6), 689 701.
IMRHAN, S. N. and RAHMAN, R. 1995, The eects of pinch width on pinch strengths of adult
males using realistic pinch-handle coupling, International Journal of Industrial
Ergonomics, 16, 123 134.
IMRHAN, S. N. and SUNDARARAJAN, K. 1992, An investigation of nger pull strengths,
Ergonomics, 35(3), 289 299.
KAMAL, A. H., MOORE, B. J. and HALLBECK, M. S. 1992, The eects of wrist position/glove type
on peak lateral pinch force, in S. Kumar (ed), Advances in Industrial Ergonomics and
Safety IV, (London: Taylor and Francis), 701 708.
KUMAR, S. 1995, Upper body push-pull strength of normal young adults in sagittal plane at
three heights, International Journal of Industrial Ergonomics, 15, 427 436.
LI, Z. M. 2002, Inter-digit co-ordination and object-digit interaction when holding an object
with ve digits, Ergonomics, 45(6), 425 440.
MACKINNON, S. N. 1998, Isometric pull forces in the sagittal plane, Applied Ergonomics, 29(5),
319 324.
MATHIOWETZ, V., KASHMAN, N., VOLLAND, G., WEBER, K., DOWE, M. and ROGERS, S. 1985, Grip
and pinch strength: normative data for adults, Archives of Physical Medicine and
Rehabilitation, 66, 69 74.
MOORE, A., WELLS, R. and RANNEY, D. 1991, Quantifying exposure in occupational manual
tasks with cumulative trauma disorder potential, Ergonomics, 34(12), 1433 1453.
NRC (NATIONAL RESEARCH COUNCIL 2001, Musculoskeletal Disorders and the Workplace,
(Washington, DC: National Academy Press).
OH, S. and RADWIN, R. G. 1993, Pistol grip power tool handle and trigger size eects on grip
exertions and operator preference, Human Factors, 35, 551 569.
PARIANPOUR, M., NORDIN, M., KAHANOVITZ, N. and FRANKEL, V. 1988, The triaxial coupling of
torque generation of trunk muscles during isometric exertions and the eect of fatiguing
isoinertial movements on the motor output patterns, Spine, 13(9), 982 992.
PEEBLES, L and NORRIS, B. 2003, Filling gaps in strength data for design, Applied Ergonomics,
34, 73 88.
SALTER, N. and DARCUS, H. D. 1952, The eect of the degree of elbow exion on the maximum
torques developed in pronation and supination of the right hand, Journal of Anatomy,
86, 197 207.
SILVERSTEIN, B. A., FINE, L. J. and ARMSTRONG, T. J. 1987, Occupational factors and carpal
tunnel syndrome, American Journal of Industrial Medicine, 11, 343 358.
 58 M. Greig and R. Wells

SNELL-MASSIE, S., BARBER, M., PAZDERKA, M., WILHELM, G. and HALLBECK, M. S. 1997,
Interaction of static pinch and forearm torque, in Proceedings of the Human Factors and
Ergonomics Society 41st Annual Meeting, (Santa Monica: The Human Factors and
Ergonomics Society), 688 691.
SNOOK, S. H., CIRIELLO, V. M. and WEBSTER, B. S. 1999, Maximum acceptable forces for
repetitive wrist extension with a pinch grip, International Journal of Industrial
Ergonomics, 24, 579 590.
SNOOK, S. H., VAILANCOURT, D. R. and CIRIELLO, V. M. 1995, Psychophysical studies of
repetitive wrist exion and extension, Ergonomics, 38(7), 1488 1507.
SNOOK, S. H., VAILANCOURT, D. R., CIRIELLO, V. M. and WEBSTER, B. S. 1997, Maximum
acceptable force for repetitive ulnar deviation of the wrist, American Industrial Hygiene
Association Journal, 58, 509 517.
SPEILHOLZ, P. S., SILVERSTEIN, B. S., MORGAN, M. CHECKOWAY, H. and KAUFMAN, J. 2001,
Comparison of self-report, video observation and direct measurement methods for
Downloaded by [Punjab Engineering College] at 03:58 19 December 2015

upper extremity musculoskeletal disorder physical risk factors, Ergonomics, 44(6), 588
613.
STOY, D. V. and ASPEN, J. 1999, Force and repetition measurement of ham boning:
Relationship to musculoskeletal symptoms, American Association of Occupational
Health Nurses, 47(6), 254 260.
SWANSON, A. B., GOSAN HAGERT, C. and SWANSON, G. 1987, Evaluation of impairment in the
upper extremity, Journal of Hand Surgery, 12A, 896 923.
VANSWEARINGEN, J. M. 1983, Measuring wrist muscle strength, The Journal of Orthopaedic and
Sports Physical Therapy, 4(4), 217 228.
WELLS, R. and GREIG, M. 2001, Characterising human hand prehensile capabilities by force
and moment wrench, Ergonomics, 44(15), 1392 1402.
ZUIDEMA, M. A. and BAUMGARTNER, T. A. 1974, Second factor analysis of physical tness tests,
The Research Quarterly, 45(3), 247 256.