Loading...

Design, model, prototype, test, analyse and evaluate a mechanical human arm (shoulder to wrist)

Bachelor Thesis 2007 122 Pages

Materials Science

Excerpt

VI. Table of Contents

I. Summary

II. Acknowledgements

III. Abbreviations, Terms, Symbols and Units

IV. Table of Figures

V. Table of Charts

V. Table of Charts

1. Introduction
Aims and objectives
Managing the Individual Project

2. Literature Review
2.1. The human arm
2.1.1. General overview
2.1.2. Shoulder joint
2.1.3. Elbow joint
2.1.4. Radioulnar joint
2.1.5. Wrist joint
2.1.6. Range of motion
2.1.7. Size, weight, and force
2.2. History of prostheses
2.2.1. General overview
2.2.2. Prostheses of the arm
2.3. Joints in prostheses
2.4. Robotic human arms
2.4.1. History of robots
2.4.2. Humanoid robots
2.4.3. Application of robotic human arms

3. Requirements for the mechanical human arm
3.1. General requirements
3.2. Degrees of freedom
3.3. Range of motion
3.4. Size, weight and force
3.5. Materials

4. Model of mechanical human arm
4.1. First CAD-model
4.1.1. Parts list and assembly
4.1.2. Angled parts
4.1.3. Description of parts
4.2. Redesign of CAD-model
4.3. Analysis of stresses and strains
4.4. Prototype of physical model
4.4.1. Chosen material
4.4.2. Fastener system
4.4.3. Assembly of the first model
4.4.4. Assembly of the redesigned model
4.5. Materials and serial production

5. Conclusions

6. Recommendations for further work

Bibliography

References

Appendix
Appendix 1 - Parts of CAD-model and physical model
General parts
Shoulder joint
Elbow joint
Wrist joint
Appendix 2 - Test of first CAD-model
Angles of shoulder joint
Angles of elbow joint
Angles of wrist joint
Appendix 3 - Images of Finite Element Analysis with ALGOR
Appendix 4 - Drawings of CAD-model
Appendix 5 - Parts list
Appendix 6 - History of developing the CAD-model
Appendix 7 - Reports of meetings
1st meeting
2nd meeting
3rd meeting
4th meeting
5th meeting
6th meeting
7th meeting
8th meeting
9th meeting
10th meeting
Appendix 8 - Project Plan for Individual Project
1. Outline of my study and the research questions I am trying to answer
2. Series of tasks of study
3. Gantt Chart
4. Main risks in the programme, how I mitigate them and my reordering of sequence

Summary

This study sets out to investigate, model and analyse a mechanical human arm. The study consists of four main steps: the literature research, modelling the mechanical human arm, building the model and finally analysing it.

The mechanical human arm is the same size as the real human arm of a 20-year-old male. The range of motion is also the same.

The investigations cover the functionality of real human arms, the history of prostheses, and applications of mechanical human arms in robotics. Requirements that are based on these information are defined and lead to the first model. This model is tested, rapid-prototyped and evaluated. Weaknesses are shown and an improved model is developed. Analyses of stresses and strains support the design decisions.

The model is designed in such a way that it is possible to add in further investigations components such as motors, pneumatic or hydraulic elements in order to allow the model to be part of a humanoid robot.

“No human investigation can be termed true science if it is not capable of mathematical demonstration. “

Leonardo da Vinci (1452 – 1519)

II. Acknowledgements

Thanks to all those who have been involved in the investigations.

Thanks to Mr. Christopher Hart who helped me with the use of the Rapid-Prototyping machines and other equipment in the CEMS labs.

Thanks to Neil Jones who helped me in producing some parts in the laboratory and finding the right fasteners.

Thanks to Dr. Siamak Noroozi who explained how to use Visual Nastran 4D software.

Thanks to my friend Christian Abraham for his support related to this study.

Thanks to Mr. Rod Veazey whose CAD-course was very helpful to me in learning how to use Solid Edge software.

Thanks to Mrs. Sue Scott for checking parts of the grammar.

Special thanks to my supervisor Dr. Gordon Smith for his help and guidance through this study.

Additionally, I want to thank my brother Nicolas Schröder who gave me some good technical advice for my work.

III. Abbreviations, Terms, Symbols and Units

illustration not visible in this excerpt

IV. Table of Figures

Figure 1: assembly of mechanical human arm

Figure 2: three-dimensional coordinate system

Figure 3: shoulder joint with omoplate and humerus

Figure 4: elbow joint with upper arm, radius and ulna bones

Figure 5: radius and ulna

Figure 6: wrist joint with radius, ulna and digital bones

Figure 7: prosthesis of arm of anonym wearer, 17th century

Figure 8: prosthesis of Götz von Berlichingen

Figure 9: “Sauerbruch-Arm” with bolt

Figure 10: prosthesis of the elbow joint before and after operation

Figure 11: ball and socket joints of prostheses

Figure 12: prostheses of the wrist joint

Figure 13: humanoid robot “Elektro”

Figure 14: humanoid robot “ASIMO”

Figure 15: anthropomorphic muscle robot ZAR 4

Figure 16: robotic arm of ”ARMAR III”, a robot built by students of Karlsruhe University

Figure 17: movements of joints of the arm (excluding finger joints)

Figure 18: not yet finished shoulder joint of first version

Figure 19: upper arm part and forearm part

Figure 20: elbow joint (racked)

Figure 21: elbow joint (inflected)

Figure 22: wrist joint

Figure 23: redesigned CAD-model of shoulder joint (with fasteners)

Figure 24: redesigned CAD-model of elbow joint (with fasteners)

Figure 25: redesigned CAD-model of wrist joint (with fasteners)

Figure 26: Fused Deposition Modelling machine during building process.
The head with the nozzles at the right side builds the part (white material) and supports (brown material)

Figure 27: fastener of the first model that connects the elbow to forearm part to
the medial elbow part

Figure 28: all 70 fasteners of the physical model of the mechanical human arm grouped by joints (shoulder: left bottom; elbow: right top; wrist: left top)

Figure 29: rapid-prototyped parts of the first model for assembly of shoulder joint

Figure 30: first model of shoulder joint with screw contacting shoulder connection
part to omoplate, error marked with red rectangle

Figure 31: first model of shoulder joint with screw contacting shoulder to upper
arm part, error marked with red rectangle

Figure 32: rapid-prototyped parts for assembly of first model of elbow joint

Figure 33: first model of elbow joint with screw contacting elbow to forearm part, error marked with red rectangle

Figure 34: rapid-prototyped parts for assembly of first model of wrist joint

Figure 35: first model of wrist joint with screw contacting wrist to hand part, error marked with red rectangle

Figure 36: all rapid-prototyped parts of the redesigned model (all joints)

Figure 37: rapid-prototyped parts of the redesigned model for assembly of
shoulder joint

Figure 38: rapid-prototyped parts of the redesigned model for assembly of elbow
joint

Figure 39: rapid-prototyped parts of the redesigned model for assembly of wrist
joint

Figure 40: forearm part

Figure 41: upper arm part

Figure 42: central shoulder part of first model

Figure 43: central shoulder part of redesigned model

Figure 44: shoulder angle part of first model

Figure 45: shoulder angle part of redesigned model

Figure 46: shoulder connection part to omoplate of first model

Figure 47: shoulder connection part to omoplate of redesigned model

Figure 48: medial shoulder ring part

Figure 49: shoulder to upper arm part

Figure 50: elbow to upper arm part

Figure 51: medial elbow part of first model

Figure 52: medial elbow part of redesigned model

Figure 53: elbow to forearm part of first model

Figure 54: elbow to forearm part of redesigned model

Figure 55: wrist to forearm part

Figure 56: wrist angle part

Figure 57: medial wrist ring part

Figure 58: wrist to hand part of first model

Figure 59: wrist to hand part of redesigned model

Figure 60: shoulder joint: 0° extension, 60° lateral rotation, medial shoulder ring
part and shoulder to upper arm contact shoulder angle part, side view

Figure 61: shoulder joint: 0° extension, 60° lateral rotation, medial shoulder ring
part and shoulder to upper arm contact shoulder angle part, top view

Figure 62: shoulder joint: 0° extension, 70° medial rotation, medial shoulder ring
part and shoulder to upper arm contact shoulder angle part, side view

Figure 63: shoulder joint: 0° extension, 70° medial rotation, medial shoulder ring
part and shoulder to upper arm contact shoulder angle part, top view

Figure 64: shoulder joint: 0° flexion/extension, 0° lateral/medial rotation, neutral position, side view

Figure 65: shoulder joint: 65° extension, 0° lateral/medial rotation, shoulder to
upper arm part and shoulder connection part to omoplate contact each other,
side view

Figure 66: shoulder joint: 90° flexion, 60° lateral rotation, shoulder to upper arm
part and shoulder connection part to omoplate contact each other, medial
shoulder ring part and shoulder angle part contact each other, side view

Figure 67: shoulder joint: 90° flexion, 60° lateral rotation, shoulder to upper arm
part and shoulder connection part to omoplate contact each other, medial
shoulder ring part and shoulder angle part contact each other, top view

Figure 68: shoulder joint: 90° flexion, 70° medial rotation, shoulder to upper arm
part and shoulder connection part to omoplate contact each other, medial
shoulder ring part and shoulder angle part contact each other, side view

Figure 69: shoulder joint: 90° flexion, 70° medial rotation, shoulder to upper arm
part and shoulder connection part to omoplate contact each other, medial
shoulder ring part and shoulder angle part contact each other, top view

Figure 70: shoulder joint: 165° flexion, 0° lateral/medial rotation, shoulder to upper arm part and shoulder connection part to omoplate contact each other, side view

Figure 71: elbow joint: 0° extension, 0° pronation/supination, neutral position,
elbow to upper arm part and medial elbow part contact each other, side view

Figure 72: elbow joint: 0° extension, 0° pronation/supination, neutral position,
elbow to upper arm part and medial elbow part contact each other, top view

Figure 73: elbow joint: 0° extension, 95° pronation, elbow to upper arm part and medial elbow part contact each other, elbow to forearm part and medial elbow
part contact each other, side view

Figure 74: elbow joint0° extension, 95° pronation, elbow to upper arm part and medial elbow part contact each other, elbow to forearm part and medial elbow
part contact each other, top view

Figure 75: elbow joint: 150° flexion, 0° pronation, elbow to upper arm part and
medial elbow part contact each other, side view

Figure 76: elbow joint: 150° flexion, 95° pronation, elbow to upper arm part and medial elbow part contact each other, elbow to forearm part and medial elbow
part contact each other, side view

Figure 77: elbow joint: 150° flexion, 95° pronation, elbow to upper arm part and medial elbow part contact each other, elbow to forearm part and medial elbow
part contact each other, top view

Figure 78: wrist joint: 0° flexion/extension, 0° abduction/adduction, neutral
position, side view

Figure 79: wrist joint: 0° flexion/extension, 0° abduction/adduction, neutral
position, top view

Figure 80: wrist joint: 0° flexion/extension, 20° abduction, medial wrist ring part
and wrist angle parts contact each other, side view

Figure 81: wrist joint: 0° flexion/extension, 20° abduction, medial wrist ring part
and wrist angle parts contact each other, top view

Figure 82: wrist joint: 0° flexion/extension, 36° adduction, medial wrist ring part
and wrist angle parts contact each other, side view

Figure 83: wrist joint: 0° flexion/extension, 36° adduction, medial wrist ring part
and wrist angle parts contact each other, top view

Figure 84: wrist joint: 73° extension, 0° abduction/adduction, wrist to hand part
and medial wrist ring part contact each other, side view

Figure 85: wrist joint: 73° extension, 0° abduction/adduction, wrist to hand part
and medial wrist ring part contact each other, top view

Figure 86: wrist joint: 75° flexion, 0° abduction/adduction, wrist to hand part and medial wrist ring part contact each other, bottom view

Figure 87: wrist joint: 75° flexion, 0° abduction/adduction, wrist to hand part and medial wrist ring part contact each other, side view

Figure 88: central shoulder part, both shoulder angle parts and shoulder
connection part to omoplate; force (50N, z-direction) applying to the surfaces of
the shoulder angle parts that limit lateral rotation

Figure 89: elbow to forearm part; force (100N, y-direction) applying to the surface
that limits supination/pronation

Figure 90: elbow to upper arm part; force (100N, -y-direction) applying to the
surfaces that limit flexion

Figure 91: medial elbow part; force (100N, y-direction) applying to the surface that limits supination/pronation

Figure 92: medial shoulder ring part; force (100N, -x-direction) applying to the surfaces that limit lateral/medial rotation

Figure 93: medial wrist ring part; force (25N, y-direction) applying to the surfaces
that limit flexion

Figure 94: shoulder connection part to omoplate; force (300N, -y-direction)
applying to the surfaces that limit flexion

Figure 95: shoulder to upper arm part; force (100N, x-direction) applying to the surfaces that limit flexion/extension

Figure 96: wrist to fore arm part and both wrist angle parts; force (100N,
-y-direction) applying to the surfaces that limit adduction

Figure 97: wrist to hand part; force (100N, -z-direction) applying to surfaces that
limit flexion/extension

Figure 98: shoulder joint with central shoulder part and medial shoulder ring part.
The central shoulder part already has an angled surface for limiting flexion and extension

Figure 99: central shoulder part with lugs in order to limit flexion, extension,
abduction and adduction all in one. This solution is not feasible

Figure 100: central shoulder part close to the final chosen design. The large lug should limit flexion, but this solution is also not feasible

Figure 101: shoulder joint with central shoulder part, medial shoulder ring part
and shoulder to upper arm part

Figure 102: elbow joint with all three main parts (inflected)

Figure 103: medial wrist ring part in early design stage

Figure 104: Gantt Chart of Individual Project

V. Table of Charts

Chart 1: movements of the joints of the arm and their explanations

Chart 2: ranges of motion of all joints of the human arm with minimum and
maximum values

Chart 3: lengths, circumferences and diameter of the human arm (50 percentile)

Chart 4: design requirements in order of importance

Chart 5: decision matrix between ball and socket joint and cardan joint

Chart 6: range of motion of the mechanical human arm, based on Chart 2
(Chapter 2.1.6.)

Chart 7: parts list with main parts, their file names, and joints where they belong to

Chart 8: assembly files with descriptions

Chart 9: parts that limit the range of motion

Chart 10: calculation of distances and lengths of upper arm part and forearm part

Chart 11: available Rapid-Prototyping methods at UWE with their advantages and disadvantages

Chart 12: parts of the physical model that are produced with Rapid-Prototyping

Chart 13: fasteners of CAD-model

Chart 14: parts list

Chart 15: series of tasks of Individual Project. The bold and italic events are accumulative events. 105

1. Introduction

The idea of replacing deformed or mutilated parts of the body has its origin since birth of mankind. Since the mediaeval times, it has been possible to manufacture flexible prostheses of extremities such as arms or legs. The early prostheses were made out of wooden parts, mostly one single wooden part. The first known prostheses were by the Egyptians in 2100 B. C. Since mediaeval times they were also produced out of iron parts. The wooden parts were still common, but there were also prostheses made out of iron and wood.

The development of modern prostheses as they are known today has its beginning in the First and Second World Wars. Surgeons such as Ferdinand Sauerbruch or Konrad Biesalski invented prostheses that allowed the wearer also more complex range of motions. These prostheses allowed the wearer simple movements within the integrated joints. Nowadays prostheses are assembled out of modern materials. They often include microprocessors paired with tactile sensors, surfaces that imitate the look and feeling of real skin and they provide almost all the functionality of real human limbs.

Additionally, there are many applications of mechanical human arms in robotics.

Aims and objectives

The topic of this Individual Project is “design, model, prototype, test, analyse and evaluate a mechanical human arm (shoulder to wrist)”. The objectives are:

- to create a mechanical human arm which is as realistic in behaviour, capacity and functionality as a real human arm. It will be created as a CAD-model and as a physical model,
- to produce models of the most complex CAD-parts with Rapid-Prototyping,
- to analyse the stresses and strains of the model.

Managing the Individual Project

Information about how the Individual Project was managed can be found in the Appendix (Chapters “Appendix 7 - Reports of meetings” and “Appendix 8 - Project Plan for Individual Project”).

2. Literature Review

2.1. The human arm

2.1.1. General overview

The term “arm” is an anatomical term. The arm is also called the upper limb. The human arm consists of three main parts, the upper arm between shoulder and elbow, the forearm between elbow and wrist, and the hand. The term arm is often used to describe the whole upper limb in colloquial speech (Wikipedia, 2006a), but the hand is not part of it.

In evolutionary terms, it is a further development of the forefoot of animals to a gripping tool for humans, based on the same vertebrate ancestor (O’Neil, 2007). Besides this, it is responsible for balancing the centre of mass of the upright walk with its commuting movements.

There are in fact four joints which are part of the human arm (Moeslund and Granum, 2001): the shoulder joint; the elbow joint; the wrist joint; and the radioulnar joint. They are all different kinds of joints which have different degrees of freedom.

Besides the bones, the arm consists mainly of sinews, nerves, muscles, blood vessels and veins. (Hillman et al, 2000).

Movement is possible in a maximum of three dimensions, depending on the anatomy of the joint. The range of movement of human joints is based on a model with a neutral position. The neutral position of a human arm is as follows:

- Shoulder: The upper arm hangs directly downwards with an angle of 0° to the trunk. The inner side of the elbow is turned to the trunk.
- Elbow: Upper arm and forearm form an angle of 180°.
- Wrist: The extension of the hand and the forearm form an angle of

180°. The inner surface of the hand is turned to the trunk.

The whole arm has seven degrees of freedom[1] (Kyoto University, 2004) and can change its ambient position via five parts of the body: the shoulder girdle, the upper arm, the forearm, the hand and the fingers. For this report, the movements of the shoulder girdle, the hand (except the wrist joint) and the fingers are not considered.

The human arm from shoulder to wrist contains 3 bones, the humerus of the upper arm, and the radius and the ulna of the forearm (The Columbia Encyclopedia, Sixth Edition, 2006). As said on the website of Norman (1999), the adjoining hand contains 27 bones; the carpus contains 8 of them which make up the wrist.

When the omoplate and the 8 bones of the wrist are added, there are 12 bones which form the arm.

There are three planes that describe the kinematics of the human body (Lenarčič and Umek, 1994). The horizontal x-axis goes through both shoulder joints, and the y-axis is the vertical axis. The z-axis forms together with the x-axis a horizontal plane. The three-dimensional coordinate system in Figure 2 is used for all descriptions within this report. There are alternative technical terms for the planes:

- x-y-plane: frontal plane (showed in red)
- x-z-plane: transverse plane (showed in blue, also called horizontal plane)
- y-z-plane: sagittal plane (showed in green)

illustration not visible in this excerpt

Figure 2: three-dimensional coordinate system

(source: self-made)

2.1.2. Shoulder joint

The shoulder joint as shown in Figure 3 is a ball and socket joint. It is also called the glenohumeral joint (Smith et al, 2004). The upper arm has a ball attached to one end, which has no perfect round shape. There is one single bone that forms the upper arm, the humerus, and it ends in a hinged joint, the elbow. Beside the thighbone it is one of the strongest bones of the human skeleton.

illustration not visible in this excerpt

Figure 3: shoulder joint with omoplate and humerus

(source: http://www.orthopaedie-gewerbepark.de/pictures/schulter.jpg)

The shoulder joint has the largest range of motion of the joints of the upper limb (Carlson, 2003) and of all socket and ball joints of the human body. The shoulder blade is one single bone which contains the socket, and the bone of the upper arm is ball shaped at the shoulder end. The ball is almost hemispherical in shape. This is the basis for movements in each direction. The ball and socket touch each other in any combination just at 1/3 of the area of the ball (Soames, 2003). The shoulder allows a movement around 3 axes.

The movements of the upper arm in the y-z-plane are called flexion and extension. Flexion is the movement towards the front of the body whereas extension describes moving the arm away from behind the body. The range of flexion is 165°, the range of extension 65°, in total 230°. Although there is not the full 360°, it is possible to circle the arms 360°, because the arms do not circulate in a round circle lying on one plane around the axis, it is more like an oval ellipse.

Raising the upper arm sideways in the x-y-plane is called abduction with a range of motion of 90°. To drop it again is called adduction, which has a range of motion of 30°[2]. Both are 120° in total.

The movements in the x-z-plane are called lateral and medial rotation. Medial rotation means moving towards the chest, in this direction at an angle of 70°. Lateral rotation is the opposite and allows a range of motion of 60°.

All ranges of motion of the shoulder are larger if the support of the shoulder girdle is also counted as defined by Berg (1999). That is the reason why the arm can be laterally raised up to 180° whereas the movement within the shoulder joint counts just 90° of this movement. Without the supporting movements of the shoulder girdle, the upper arm loses much of its range of motion.

2.1.3. Elbow joint

At the other end of the upper arm is the elbow joint. The elbow joint is often regarded as a joint which allows movements around two axes, but this assumption is incorrect. Only in combination with the radioulnar joint more complex movements are possible (Gray, 2005). The elbow joint itself is a hinge joint with possible movement around one axis. At the other side of the elbow joint, there is the forearm. The forearm consists of two bones, the radius and the ulna which are both connected to the elbow joint. Therefore, the elbow joint has two distinct articulations as Figure 4 shows. They are called humeroulnar and humeroradial joints (Soames, 2003).

illustration not visible in this excerpt

Figure 4: elbow joint with upper arm, radius and ulna bones

(source: http://www.orthopaedie-gewerbepark.de/pictures/ellbogen.jpg)

The elbow of a female person allows a total range of movements of 160°, divided into 150° for flexion and 10° for extension. Corney (1971) shows that 85% of females are able to reach an angle of extension, but males normally can not. Both movements take place in the y-z-plane. When the elbow is elongated in its neutral position, the angle between upper arm and forearm is 180°.

2.1.4. Radioulnar joint

The two bones in the forearm (Figure 5), the radius and ulna, form another joint in the arm, the radioulnar joint. The radius rotates around the ulna, and both bones end in the wrist. The ulna is the longer and thicker of the two bones (Wikipedia, 2006b); it lies at the side of the small fingers.

illustration not visible in this excerpt

Figure 5: radius and ulna

(source: http://www.uwyo.edu/RealLearning/injuries/pictures/HR188%20Bent%20Ulna/dscn3855.JPG)

The bones of the forearm are not in contact with each other at all. They are both connected to the wrist and elbow joints. Because of this, a radial movement around the y-axis is possible. But this movement is not part of the elbow joint itself. It can be regarded as a fourth joint which exists because of the unique arrangement of the bones between elbow and wrist. It provides the arm with another axis of movement and is important for flexible positioning of the hand and its work carried out (Elbow Pain Info, 2007).

There are two joints which allow the movement of radius and ulna, the superior radioulnar joint, located towards the elbow joint, and the inferior radioulnar joint located towards the wrist joint. Radius and ulna each have two joints. Although they are directly positioned in the elbow joint in a similar way to the wrist joint, they are not the same as the elbow joint or the wrist joint.

The radioulnar joint has a range of movement of 95° in each direction, or a total movement of 190°. The movements are called pronation and supination. In supination, the bones cross each other whereas they are parallel to each other in pronation.

2.1.5. Wrist joint

The wrist joint consists of a double row of four small short bones each (Marshall, 2007) as shown in Figure 6. The bones which are next to the forearm form the radiocarpal joint. The second row of bones next to the bones of the fingers forms the intercarpal joints. The intercarpal joints are not considered any further in the investigations of this report, because they are not needed for the functionality of the wrist joint leading to the forearm.

illustration not visible in this excerpt

Figure 6: wrist joint with radius, ulna and digital bones

(source: http://www.ruhr-uni-bochum.de/radiologie-josefhospital/download/3d_handgelenk.jpg)

The radiocarpal joint of the wrist joint is an ellipsoid synovial joint (Soames, 2003). It allows two degrees of freedom, but it has limits in comparison to the socket and ball joint of the shoulder joint because of the oval shape of the ball. Turning movements around the elongated axis of the forearm are not possible. But because of the ability to turn the forearm about 95° in the radioulnar joint, the limit of the wrist joint is nullified.

The wrist allows movements in two planes. The movements of the hand up and down are called extension (up) and flexion (down) and take place in the y-z-plane. The second pair of movements, moving the hand to the left and to the right, is called abduction (away from the body) and adduction (towards the body). In the model of neutral positions, these two movements take place in the x-y-plane.

The range of motion for flexion and extension decreases with age in the wrist joint more than in other joints. For a 20-year-old person the flexion is 75° and the extension 73°, in total 148°. Abduction is 20° whereas adduction is 36°, in total 56°.

2.1.6. Range of motion

Gender and age are the two main reasons for different ranges of motion between individual humans (Bell and Hoshizaki, 1981). Additionally, heredity and individual circumstance such as fitness also influence the ranges of motion as cited by Vlach, 2007.

Generally, female persons are more agile than male ones, but there is only a little difference and in some joints there is no relevant difference. The older the person, the less agility there is and the movement is more and more confined. In neonates the range of motion is the largest and it decreases throughout life. Children lose the largest relative percentage of their range of motion during their first five years of life. Other reasons for differences in the range of motions have genetic causes. Chart 1 shows the terms of the different possible movements of the human arm from shoulder joint to wrist joint. These movements limit the ranges of motion.

illustration not visible in this excerpt

Chart 1: movements of the joints of the arm and their explanations

(source: self-made)

Chart 2 shows an overview of the minimum and maximum values of the range of motion. The maximum values are those of neonates and the minimum values are those of 60 to 70-year-old persons. Measurements have been taken from both female and male persons. These values are of average nature and exceptions are possible.

illustration not visible in this excerpt

Chart 2: ranges of motion of all joints of the human arm with minimum and maximum values

(source: self-made)

2.1.7. Size, weight, and force

Size, weight and force of the human arm are also dependent on gender, age, heredity and individual circumstances.

The science of measuring lengths, circumferences, diameters and distances of the human body and other measurements e. g. volume or mass is called anthropometry (NASA, 2006). Because of the varying individual lengths and distances, there is a system necessary that groups average lengths together. Due to this, the percentile is an important term. The measurements made on a preferably large group of humans are divided with the help of the percentile. There are three main figures related to the percentile:

- the 5 percentile (P5)
- the 50 percentile (P50)
- the 95 percentile (P95)

When it is necessary to ensure that 90% of the population are in a range of measurements, the difference between the 95 percentile and the 5 percentile is taken. The 50 percentile shows the average value. In Chart 3, the basic values of the human arm are shown for the 50 percentile.

illustration not visible in this excerpt

Chart 3: lengths, circumferences and diameter of the human arm (50 percentile)

(source: Dreyfuss Associates, 2002)

Both arms have a weight of 10% to 12% of the total weight of the human body as defined by Croney (1971b). This means for a person that weighs 80 kg that one arm weighs from 4 kg to 4.8 kg. The percentage allotment of the bones on the total weight of the human body is about 12% to 14%. This results in a weight of 0.48 kg to 0.672 kg for the bones of one arm[4]. These figures are important for the choice of materials for the mechanical human arm.

Force is not as easy to determine as weight or size. It is heavily dependent on the individual fitness. Because of these huge biological differences, the force that a human arm can produce is difficult to ascertain. It is not further covered in this chapter, but later discussed in the requirements.

2.2. History of prostheses

2.2.1. General overview

The word prosthesis has a Greek origin; it is called “protithenai” and means to hypothesise or to put in addition (MedicineNet, 2007). Prosthesis can be divided into three groups: extremities, organs and parts of organs.

The first simple prostheses for extremities were invented by the Egyptians in

2100 B. C. and consisted mainly of wood (Nerlich et al, 2001). From the mediaeval times onwards, prostheses were made out of wood or iron.

Chronicles show that it was possible to produce prosthesis from the elbow to the fingers with high technical skills in the 17th Century, as shown in Figure 7. All the prostheses up to this point in time had in common that they could not be moved by the muscles of the adjoined body parts. Either the prosthesis itself was one part and not flexible, or it was flexible but it had to be moved with the healthy hand. These prostheses are called passive ones. The other ones that can be controlled by the wearer with the help of his muscles are called active ones.

illustration not visible in this excerpt

Figure 7: prosthesis of arm of anonym wearer, 17th century

(source: http://www.chess.at/geschichte/kempelen-Dateien/image008.jpg)

The prostheses for extremities can be divided into two other groups: the exoprostheses which are outside the body and the endoprostheses which are inside the body and covered by tissue. Prosthesis for the upper limb can be parts of both groups. Another division can be made between pretty prostheses and functional ones.

2.2.2. Prostheses of the arm

Before the 20th century, the replacement of a single joint was impossible because of the lack of medical knowledge. The first prostheses substituted the wrist and the hand or the forearm with wrist and hand. There was never a single joint, it was always an extremity. Afterwards, prostheses with joints were invented.

In the year 1504 the German knight Götz von Berlichingen lost his right hand during a military campaign. He got a prosthesis that was very complex for those days and was considered as a mechanical marvel. It consisted of approximately 200 mainly steel parts (Figure 8). It was possible to change the positions of the fingers and the thumb (Lovasz, 2003) with gears located inside the arm. This was an individual prosthesis and not part of mass production.

illustration not visible in this excerpt

Figure 8: prosthesis of Götz von Berlichingen

(source: www.bvmed.de/stepone/data/downloads/4d/9a/00/GeschichteMedTechnik2.pdf)

During the First World War the surgeon Ferdinand Sauerbruch invented the “Sauerbruch-Arm” as shown in Figure 9, a prosthesis for victims of the war. The special feature of this prosthesis is the potentiality to move the prosthesis with muscularity (The National Academic Press, 1997). This was done using a wooden bolt which was inserted into a channel of the intact muscles. This bolt was linked to metal wires which controlled the grip of the fingers. The wearer of the prosthesis was able to close or open the fingers of the hand with his intact muscles of the upper arm. The price for this prosthesis was very high that few people could afford it.

illustration not visible in this excerpt

Figure 9: “Sauerbruch-Arm” with bolt

(source: http://www2.hu-berlin.de/hzk/kabinette/large/SauerbruchArmprothese.jpg)

There is one main disadvantage of the prostheses that are controlled via mechanical wires placed into the intact muscles. The stronger the grip of the fingers, the more force the wearer has to exert. So there is a limit to the force.

Another idea for generating more force is the application of the myoelectric technique (Zhou et al, 2005). The movement of muscles causes a changing electric potential on the surface of the skin. These signals are very low and hence to be amplified before they can be used to control the prosthesis. In these prostheses there are microcontrollers and other electronic parts that control the motors opening and closing the fingers or performing other movements, e.g. flexion of the wrist. The disadvantage of these prostheses is that they are usually heavy because of the motors and hence electric supply units.

The aim in developing modern prostheses is to reduce the weight but ensure as much functionality as the real limb. Modern prostheses are lighter than the older ones. New materials are used to decrease weight, to improve firmness and to afford more movements. A hand out of fluid copes with these requirements best. The fluid materials are controlled by electric fields and are very light. Another advantage is that the prosthesis has tactile sensors which allow it to be forceful but also with fine motor skills. This is due to the grasp force of the gripper which can be adjusted and controlled (Zhi-Zeng et al, 2005).

Data about weight and other technical information from manufacturers of prostheses was not available in order to support the investigations of this report.

2.3. Joints in prostheses

The joints are important parts of endoprosthesis and exoprosthesis. For prostheses, the functionality of the radioulnar joint and the elbow joint are grouped together in prostheses as shown in Figure 10. They are more easily realised in mechanical form than within the human body. The joints themselves are small, but capable of coping with the forces needed by the wearer. Because of the divided layout of the joints in some prostheses, there are two centres of rotation that deviate from the biological arrangement of the human elbow.

illustration not visible in this excerpt

Figure 10: prosthesis of the elbow joint before and after operation

(source: http://www.linz.at/images/GSBProtheseseitl(1).jpg)

The shoulder joint is mostly realised with ball and socket joints as shown in Figure 11. These joints are very small, accurate and maintenance-free. They fit perfectly into the human body as their size is the same as the real human shoulder joint. They are connected to the bones via bolts that are stuck in drilled holes in the bones.

illustration not visible in this excerpt

Figure 11: ball and socket joints of prostheses

(source: http://www.zimmergermany.de/html/pronews/PN0303.pdf)

The prostheses of the shoulder joint have one centre of rotation for all three degrees of freedom. Because of the round shape of the inner ball and the bearing, a higher range of motion is possible than the real human shoulder joint normally provides (OhioHealth, 2006). Theoretically, the wearer of the prosthesis could turn the upper arm around the y-axis in circles, but this is limited because of sinews and muscles and also because of the design of the bearing. In this connexion, prostheses with these joints allow the wearer to be even more flexible, but they are an inexact copy of the capacity of the human shoulder joint.

Prostheses of the wrist joint are the ones of the arm that are the most similar to the biological ideal. This is due to the ellipsoid shape of the ball and the socket. The structure of the radiocarpal bones and the radius and ulna bones are similar to this. The contacting surfaces of ball and socket automatically limit the ranges of motion. A typical modern prosthesis is shown in Figure 12.

illustration not visible in this excerpt

Figure 12: prostheses of the wrist joint

(source: http://www.argomedical.ch/images/universal2.jpg)

The most important property of joints in prostheses is, besides the easy movement within the range of motion, the capability to work within the human body. Therefore, the joints need mountings or bolts to be fixed to the bones and their size must allow the surrounding sinews and muscles to fulfil their task.

2.4. Robotic human arms

2.4.1. History of robots

The term “robot” first appeared in the novel “R. U. R. (Rossum’s Universal Robots)“ by Karel Čapek in 1921. The word is Czech in origin and means work or forced labour. Earlier, in about 320 B. C., Aristotle philosophised about tools working on their own (History of robots, 2007). Other authors followed who wrote about robots or robotics. Isaac Asimov is one of them, and his novel “Runaround” was, after the one by Karel Čapek, the next most famous one that brought the topic to a wider publicity.

Jacques de Vaucanson constructed a programmable loom in 1740. It is regarded as a robot, but it worked without electricity and was not moveable. Later, robots were often parts of films, but for a long time technical knowledge was not sufficient to manufacture robots that could make decisions. This changed in 1949, when one of the first mobile robots was developed. Its name was “Elise” and it was developed at Bristol University. It had the shape of a turtle and was able to follow a light source (About machines and humans, 2002).

The first commercially available robot was the industrial robot called “Unimate” (University of Texas at Austin, 2007). This was in 1962 and it was mostly used by General motors for dangerous work at conveyor-belts.

Important disciplines that are needed for robotics are mechanical engineering, information technology, and electrical engineering. The development of robots was overcome in the last two decades more and more limitations, leading to robots that can walk like humans, the so-called humanoids.

2.4.2. Humanoid robots

Humanoid robots belong to the group of walking robots. They fulfil two elementary requirements: They are able to walk upright and they can appreciate their environment. The first humanoid robot (Figure 13) was developed between 1937 and 1938 by the company Westinghouse as described by Weeks (2005). It was called “Elektro” and had some basic functions, like walking or raising the arm that could be remote-controlled[5].

illustration not visible in this excerpt

Figure 13: humanoid robot “Elektro”

(source: Weeks, 2005)

Later, there have been several parallel developments in this research area. The first stable walking humanoid robot was developed in 2000 by Sony and called “Qrio”. Other programmes have been tracked by Toyota, Mitsubishi, Fujitsu and Honda. Honda started the “Humanoid Robot Research and Development Program” in 1986 which finally led to “ASIMO” (Figure 14), the humanoid robot with the actual highest state of development. First, these robots learned how to walk, later came developments on the application of the whole body and how to use the arm.

illustration not visible in this excerpt

Figure 14: humanoid robot “ASIMO”

(Source: HONDA, 2001)

2.4.3. Application of robotic human arms

This topic considers arms of humanoid robots or robotic arms that are similar to human arms, but they are distinct from arms that are applied in industry such as SCARA[6] applications for example. Those robotic arms have shapes and functionalities that differ partially strongly from a human arm.

Due to the optimal physiology of the given degrees of freedom and the specific arrangement of the joints, the shape and functionality of a human arm are often copied in a technical way. It is the same reason for the whole humanoid robot as for the robotic human arm – they are designed to operate in an environment that is arranged for humans, especially the human living space (HONDA, 2001). Systems that deviate from the shape of the human arm are not as easy to implement in a human environment as those that articulate the most human-like.

illustration not visible in this excerpt

Figure 15: anthropomorphic muscle robot ZAR 4

(source: Boblan, I., Bannasch, R., Schulz, A., Schwenk, H., 2007)

Almost all robotic arms belong to a robotic system with at least a body. Most of them are developed by Universities in order to do research on robotics, artificial intelligence, new materials (Figure 15) or other fields of research. Figure 16 shows that there is not always a unique centre of rotation for all degrees of freedom of one joint. In order to keep the functionality simple, the joint is divided into sub-joints of which each has got its own actuation in order to fulfil the ranges of motion.

illustration not visible in this excerpt

Figure 16: robotic arm of ”ARMAR III”, a robot built by students of Karlsruhe University

(Source: IPEK, 2006)

Moving the upper arm and the forearm as a real human arm is technically solved. But in order to get the full capacity of a human arm, there are sensors needed that help the fingers to recognise when an object is fully-gripped with enough force to hold it and less to destroy it. Research concentrates on this.

There are several applications for robotic human arms in different areas. For example, there are special robotic arms for industry, especially at assembly-lines, and they are used for other tasks such as missions in space (TOSHIBA CORPORATION, 1999), entertainment, research on the above described fields or prostheses. In robotic arms, the ranges of motion are limited by sensors, but there are normally no parts or surfaces that contact each other in the largest possible angle. Robotic arms and prostheses of the human arm often have some key features in common and research overlaps. Nevertheless, developments in the robotic field are often the cutting-edges for modern prostheses.

3. Requirements for the mechanical human arm

3.1. General requirements

The design of the mechanical human arm as a mixture of prosthesis and a robotic arm requires most of the information of Chapter 2. Some parts like the history are less important. The mechanical human arm is neither a real prosthesis, nor a real fully working robotic arm. This is the reason why the information on prostheses and robotics is combined in the following requirement discussion.

There is no need for the mechanical human arm to consist of the same number of parts as the bones in the human arm. One reason for this is the simplification, for example by combining the elbow joint and the radioulnar joint into one single joint. In this way, two bars representing the radius and ulna are not necessary, and weight is also saved. Additionally, there are just three instead of four single joints. It is not possible to build the joints of the mechanical human arm in the same way as prostheses are built. The joints of prostheses often consist of two parts, for example the wrist joint, and are often covered by tissue consisting of muscles and sinews. These support the locations of the parts of the prosthesis to each other because they surround them. It is different with the mechanical human arm, because there is no surrounding tissue. Hence, fasteners are required to fix the parts to each other.

Building the mechanical human arm that it can be used as a prosthesis causes a challenge due to a lack of individuality. Each wearer has different ranges of motion. Hence, there are small angles that can be exchanged. These angles provide the model with the required range of motion. The whole arm can be adjusted for any age and both genders with the help of these angles. The forearm part and the upper arm part can also be exchanged for longer or shorter ones. This solution provides the mechanical human arm with flexibility. Furthermore, by reversing the angles, the mechanical human arm can be adapted to be a right or a left arm.

illustration not visible in this excerpt

Chart 4: design requirements in order of importance

(source: self-made)

As shown in Chart 4, there are several main requirements. They are in the order of importance with the highest importance at the top. Some of them do not support each other. The range of motion is the most important requirement as it is one of the aims and objectives. All the other requirements, except for one, support this requirement. The only one that does not contribute to this aim is the firmness of the model. Complex shapes are necessary, especially within the shoulder joint, to allow the joint the full range of motion. These complex shapes are not very firm, for example the lugs of the shoulder connection part to omoplate. In reverse, a smaller range of motion would allow stronger features like lugs and therefore enhance firmness.

The firmness of the model is in opposition to three other requirements. The first one is the amount of material used for the model. The less material used, the lower the weight. This is eminently important for the application of the mechanical human arm as prosthesis. But the difficulty is to find the right balance between firmness and weight. Between these two, firmness is the more important one. Nevertheless, the chosen materials are not too heavy and therefore the requirements are not in opposition to each other in a strong way.

There are higher stresses and strains at the maximum range of motion between the parts than somewhere in the middle area of the range of motion, because there, some surfaces contact each other. The model becomes firmer, as the number of surfaces in contact with each other increases and as larger the contacting surface area is. Of course this is a large advantage, but there are also disadvantages. For example, if not only the shoulder angle parts, but also some surfaces of the central shoulder part limit the lateral and medial rotation, then there would be an unwanted redundancy. Reassembling the shoulder angle parts the other way round would cause the problem that the medial shoulder ring part could no longer perform the full range of motion. So there should only be two surfaces that limit the range of motion in order to keep the highest flexibility.

The firmness is also influenced by the diameter and the circumference of a joint. It is strongly dependent on the design, where the forces and the stresses and strains are carried through the material. In order to test the angles of the joints properly, it is of importance to locate the surfaces that limit the range of motion as far away from the centre of rotation as possible. These surfaces have to hold high stresses and strains in the maximum range of motion. But this leads partially to a chosen design where, for example, some lugs have a weak shape. In this case, the firmness does not support to the requirement of carrying out the forces as far away from the centre as possible.

Designing and modelling a mechanical human arm means not only to create a CAD-model that looks good, behaves as a real human arm and can cope with the same capacitance. It must also be feasible to manufacture it. There is a broad range of complex shapes that can be rapid-prototyped, but the parts should be feasible to be manufactured, for example by casting. Therefore, the shapes of the parts should follow some common rules for casting and other manufacturing processes. A less important requirement says that the model should consist of as few parts as possible. These two requirements do not agree with each other as generally applies that the less parts there are, the more complicated they will be. The requirement of ease of manufacture is the more important one of these two.

A further requirement is that the parts should have the same size as a real human arm. This requirement does not affect the other requirements. It is on the fourth place in the order of importance, because if the size of the mechanical human arm does not fit to a real human arm it is worthless as prosthesis.

The requirement that the model should consist of as few parts as possible is the one that affects the most of the other requirements. Having as few parts as possible facilitates assembly and minimises the complexity of the mechanical human arm. So, it has several advantages. On the other hand, it has some main disadvantages that should not be underestimated in their impact on a well designed solution. The impact on the ease of manufacture has already been discussed earlier. The second affected requirement is the application of a right and left arm without any additional parts. It is useful to build the mechanical human arm as a right or left arm by just disassembling four angle parts and reassembling them the other way round. But this requires more parts than just building a model for one side of the body.

Using standard fastener parts such as screws and nuts supports the requirement of easy assembly. But it does not agree to the requirement that the model should consist of as few parts as possible. On the one hand, using a solution with screws or bolts requires at least one bolt and one screw. This is quite a lot if regarded from the point of view of the total number of parts. But on the other hand, there are no other appropriate solutions possible in order to reduce the number of parts. Clipping the parts together would be possible, but this demands more complex shapes of the parts. Gluing parts together is not appropriate because once glued together there is no flexibility of easy exchange of parts left. However, standard fasteners do not weigh too much, they are available in different variations and are therefore appropriate.

The redundancies of joints within joints means that, for example, the wrist to forearm part and the medial wrist ring part are not only connected at only one place. This is important to allow the joint an accurate movement and smooth running. But this requirement affects the number of parts as most main parts are connected at two places and therefore require double the number of fasteners. The requirement of the redundancy is almost at the bottom of the list of requirements, because it is partial already covered by the requirements of the full range of motion and the survivability of high stresses and strains. Those would be affected in a negative way without the requirement of redundancy of joints within joints.

Applying other ranges of motion by easy exchange of angled parts is the last requirement in order of importance because this is no part of the aims and objectives, but it enhances the flexibility and usability of the mechanical human arm and is therefore adding value to it. This is the large advantage, but it does not agree with the requirement of consisting of as few parts as possible. The requirement that parts should be designed out of as little material as possible to lower weight does partially not agree to the requirement that just two surfaces should contact to limit the range of motion. The more contacting surfaces there are, the more material will be necessary. To set these requirements not in opposite of each other actually means that there should be just one surface that limits the range of motion. But this would minimise firmness.

A minor conflict between two requirements is the fact that the model should be easy to assemble and the redundancy of joints within joints. Assembly time takes longer as there are more parts, but it does not always affect the ease of assembly. For example at the assembly of the screw-nuts to the medial wrist ring part, there is just little place for assembly, this would not have occurred without redundancy of joints within joints.

As the whole discussion of the requirement shows, there are many conflicts which have to be evaluated in order of importance. The order of importance is driven by the aims and objectives of the study, followed by applied technical standards and rules. Deciding for each part as an individual component and for the arm as a whole system which requirements are the most important and which may be disregarded is a difficult process. These decisions can be found in the chapter with the description of parts (Chapter 4.1.3.), but of course not every single feature of each part described in detail.

Nevertheless, there are also many requirements that support each other. This makes the design decision in many ways easier. Additionally, the requirements discussed above lead to a high functionality, maintainability and dependability. This is the reason why these three points are no single requirements that are additional mentioned in Chart 4. They result from the other requirements automatically.

One design decision of the whole arm could have been to model the complete upper arm as one single part and the complete forearm also as one single part. This layout would have had the advantage that there are fewer parts, but it was not realised for several reasons. Firstly, large parts like the complete upper arm with some specific shapes for the joints at both ends would be difficult to manufacture. A further reason is that there is less flexibility in response to changing requirements. For example, if the upper arm should be 10mm less long for a different model, in this case a complete large and complex part has to be remanufactured. On the other hand, the chosen design with several smaller parts fulfils changing requirements because smaller and simpler parts can easily be remanufactured.

Two possible designs for the shoulder joint are considered, a ball and socket joint or a cardan joint. Ball and socket joints are common for shoulder joints of a prosthesis of the arm. Chart 5 shows the allocation of some arguments. The matrix gives an overview of the strengths and weaknesses of both solutions. Because of the clearly better results for the cardan joint (63.25%), the ball and socket joint is not a good solution. Nevertheless, a cardan joint is not applicable to a prosthesis because it does not cope with the muscles and sinews that surround it. It would only work if there was a cover around it. The decision matrix shows this inapplicability, but the argument does not count too much to the total allocation.

illustration not visible in this excerpt

Chart 5: decision matrix between ball and socket joint and cardan joint

(source: self-made)

[...]


[1] Finger joints and shoulder girdle are not included in this figure.

[2] This figure includes the range of motion of the shoulder girdle.

[3] No further values found in literature. Nevertheless, this range of motion also decreases at higher age as all others do.

[4] Calculation of boundaries: 4 kg * 0.12 = 0.48 kg; 4.8 kg * 0.14 = 0.672 kg

[5] Technically, it is no robot because it is remote-controlled.

[6] SCARA stands for “Selective Compliance Assembly Robot Arm” or “Selective Compliant Articulated Robot Arm”

Details

Pages
122
Year
2007
ISBN (eBook)
9783640097302
ISBN (Book)
9783640119356
File size
4.4 MB
Language
English
Catalog Number
v94662
Institution / College
University of the West of England, Bristol
Grade
1,0
Tags
Design Individual Project

Author

Previous

Title: Design, model, prototype, test, analyse and evaluate a mechanical human arm (shoulder to wrist)