Lopez de Lacalle L.N., Lamikiz A. Machine Tools for High Performance Machining

Подождите немного. Документ загружается.

4 New Developments in Drives and Tables 153

tained by the oil lamina is provided by an external pump. In this manner, the op-

eration is guaranteed without wear and tear, and without the possibility of the

stick-slip phenomenon appearing, due to the fact that the bearing does not present

any static friction [2].

One of the surfaces is provided with cavities or cells, known as “oil cells”,

which are provided with pressurised oil from the outside. Around these cells is the

area through which the oil is released as it loses pressure. This area is known as

“land”. The distance between the land and the surface which slides over it is the

“oil gap”. Usually, it is 10 to 40 microns and generates a certain resistance to the

passage of the fluid. The pressure difference between the cavity and the atmos-

pheric pressure which acts on the outside is known as “cell pressure”.

The oil flow resistance along the gap can be estimated as follows:

⋅

⋅⋅

(4.8)

The pressure gradient along the length of the land can be assumed to be linear

in an initial approximation, such that for purposes of force it can be assumed that

the pressure acts on up to one half of the length of the land. The area on which it is

assumed that the complete pressure acts is known as the effective area (A

eff

The hydrostatic guides are designed with several cells, such that they can sup-

port off-centre forces and moments. Each cell should be supplied at a different

pressure, in order that it can withstand a different force according to the operating

conditions. For this, usually a single pump is used, with restrictors that allow to

supply each cell at the appropriate pressure.

In this manner the pressure in a cell will be:

⋅=

(4.9)

Usually, the restrictors are built in the form of capillaries so that in this manner

the resistance depends on the oil viscosity, as occurs in the cells. The resistance to

the oil flow of a capillary can be calculated as follows:

⋅

⋅⋅

(4.10)

Different types of capillaries are used. The short restrictors should have a very

small diameter to provide the necessary resistance. This diameter is limited by the

size of suspended particles in the oil which can block the capillary. The use of short

capillaries is also limited because the design is very sensitive to the diameter of

these; the resistance depends on the fourth power of the diameter. Capillaries with

larger diameters and a longer length are also used, which tend to be in spiral form.

The state of equilibrium of a cell vs a specific force can be determined. From

this state of equilibrium, the stiffness of the system formed of the pump, restrictor

154 A. Olarra, I. Ruiz de Argandoña and L. Uriarte

and cell can be determined. In case of a cell supplied through a capillary by means

of a pump which operates at a constant pressure, the resulting stiffness is:

⋅⋅=

(4.11)

As can be seen in the previous expression, the stiffness depends on the load

supported by the cell. For this reason, it is very interesting to use hydrostatic

guides with a “retaining plate” in order that a high preload can be applied. In addi-

tion to gaining stiffness, the guide can absorb loads in both directions.

4.6.3.1 The Damping

The hydrostatic guides provide much higher damping values than the roller-based

guides against movements perpendicular to the cell; this is due to the friction force

that the oil lamina presents on sliding. In case of movements parallel to the cell,

the damping is reduced because there is no displacement of fluid.

4.6.3.2 The Energy Consumption

The energy consumed by hydrostatic guide depends, on the one hand, on the

work carried out due to friction in the land and, on the other, the work carried out

by the pump. This energy is integrally transformed into heat which increases the

oil temperature.

The work carried out to overcome the hydrodynamic friction can be determined

as:

AvFW

rrr

⋅⋅=⋅=

(4.12)

while the work carried out by the pump is:

⋅

(4.13)

The total work results in:

()

WWW

rpp

ηε

⋅

⋅⋅+

⋅

⎟

⎠

⎞

⎜

⎝

⎛

⋅

=+=

(4.14)

The dimensioning of a hydrostatic guide is relatively complex due to the strong

dependency shown by the variables which define the status of this. A very important

variable is the temperature rise of the oil throughout the hydraulic circuit and the

temperature in the restrictors and throughout the hydrostatic guide cells. The oil

4 New Developments in Drives and Tables 155

viscosity is heavily dependent on the temperature such that the calculation of the

state of equilibrium requires repeating the calculation several times. The prepara-

tion of a computer program which assists the designer in the calculation of this state

of equilibrium for different speeds or loads applied to the guidance is of interest.

4.6.3.3 The Hydraulic Circuit

The functions of the hydraulic circuit are to ensure an adequate supply of oil to the

hydrostatic guidance, in regard to pressure and flow, to suck from the oil the heat

generated by the losses and prevent contact between the sliding elements in case

of pump failure. It generally consists of a pressurised oil tank, a pressure control

system, a safety valve, a suction pump, dirt and a pump filter. Sometimes, an oil

cooling system located after the suction pump is also necessary for the collection

of the oil exiting from the guidance. The function of the pressurised tank is to

provide oil to the bearing in case of pump failure, as well as to equal the oil flow

at the pump discharge. The most common type of pump used is the gear pump.

In general, the cost of the hydraulic system tends to be high compared to other

types of guides.

4.6.3.4 The Design Criteria

In general terms, it is of interest to keep the gap as small as possible. However, the

minimum value is limited by the accuracy which can be obtained during the manu-

facture and by the elastic deformation of the components, in order to avoid contact

between sliding surfaces at all costs.

In order to determine the stiffness of the guidance, it is appropriate to consider

the flexibility of the slides themselves. It is probable that this is comparable with

that of the guidance.

On the other hand, special attention should be given to the system for collecting

oil released through the hydrostatic guidance and its return without contamination.

For this, appropriate seals should be provided in each case. As an example, the

characteristics of a rotary table are shown in Table 4.5.

Table 4.5 Design parameters of an aerostatic rotary table by Precitech

Table outer diameter 380

Axial error motion

Radial error motion

0.035

μm

0.079

μm

Load capacity (safety factor 3) 6,670

Axial stiffness (17

bar)

Radial stiffness

Tilt stiffness

1,315

N/μm

700

N/μm

Nm/μrad

Effective stiffness 300

mm above the centre of bearing 162

N/μm

156 A. Olarra, I. Ruiz de Argandoña and L. Uriarte

4.6.4 Aerostatic Guides

In the case of aerostatic guides, the medium which separates the sliding surfaces is

the air. There are no fundamental differences in respect to the hydrostatic guides

in that referring to the operating mode.

The following points are highlighted from among the characteristics and advan-

tages of this technology [4]. The aerostatic guides present exceptionally low losses

due to friction, including at very high speeds, due to the low viscosity of air. At

the ambient temperature, the air viscosity is approximately three orders of magni-

tude less than those of the oils used in hydrostatic guides. In addition, it is not

necessary to consider the means for returning the fluid and seals are not required.

An interesting characteristic of the aerostatic guides is their aptitude for applica-

tions subject to large differences of temperature, due to that the viscosity of the air

presents a good consistency in a wide range of temperatures.

On the other hand, the dimensioning of an aerostatic guide is more complicated

than in the hydrostatic case. Air compressibility, for example, contributes to this.

This compressibility, combined with the fact that the air barely provides damping,

generates different dynamic phenomena which, in general, are problematic. There

may be, for example, self-excited excitation, known as “air hammering” which

may become audible and on occasions seriously affect the correct operation of the

guidance. This vibration can be mitigated through the use of a high number of

cells, each with its restrictor. For this reason, nowadays, the guides are manufac-

tured in porous material. This tends to consist of a sintered part in which the po-

rosity emulates the operation of multiple restrictors.

This type of guidance is seldom used in machine tool applications, mainly due

to the difficulty in attaining sufficient load capacities at a reasonable cost of the

air-supply equipment. For this reason, it is necessary to machine the sliding sur-

faces to a very high degree of accuracy, such that the gap can be extremely small.

As an example, certain characteristics of interest of a workholding spindle of an

ultraprecision lathe are presented in Table 4.6.

Table 4.6 Design parameters of a workholding aerostatic spindle by Precitech

HSS SP75 HSS SP150

Material Steel shaft/bronze journal

Swing capacity 220

mm diameter

Load capacity 180

N 680

Axial stiffness 70

N/μm 228

N/μm

Radial stiffness 22

N/μm 88

N/μm

Motion accuracy Axial/radial

nm Axial/radial

nm through dynamic range

Max. speed 15,000

rpm 7,000

rpm

4 New Developments in Drives and Tables 157

4.7 The Present and the Future

Some new ideas are now developing, with prototypes and even released products

with improved features. Some of them are explained below.

4.7.1 Rolling Guides with Integrated Functions

The manufacturers of rolling guides have made an effort to integrate functional-

ities in their systems (Fig. 4.24). For example, presently, there are rolling guides

with position pickup systems. The pickup system offers a similar performance to

that of the average level linear encoder and is compatible with the electronics that

this uses. Another product are the racks prepared to support a rolling guidance

system. These products permits us to reduce the time required for design and as-

sembly and allow the machine tool manufacturer to offer more compact solutions.

Fig. 4.24 Rolling guides with integrated functions, by Schneeberger

4.7.2 The Hydrostatic Shoe on Guide Rails

Another interesting product which has recently appeared on the market is the shoe

designed to operate on guides similar to those used in rolling systems, shown in

Fig. 1.16b in Chap. 1.

This product allows the machine tool manufacturer to obtain the performance

of the hydrostatic guides without having to design and manufacture the guides

themselves.

158 A. Olarra, I. Ruiz de Argandoña and L. Uriarte

4.7.3 Guiding and Actuation through Magnetic Levitation

State-of-the-art technology has allowed the development of machine tool slides

guided by active magnetic fields [1]. The characteristics of a levitated slide, driven

by means of magnetic fields, are shown in Fig. 4.25.

Stroke (X and Y) 100

Max velocity 60

m/min

Max aceleration 20

m/s

Jerk 1,000

m/s

Kv 10

m/mm.min

Linear resolution 60

Angular resolution 0.046

arcsec

Linear accuracy 0.2

μm

Workpiece weight 120

Linear force 2,000

Fig. 4.25 Planar motor magnetically levitated, by Tekniker

Among the advantages of this type of slide is the fact that there are no forces of

friction, the dynamics are very high due to the fact that there are no transmission

force elements and the accuracy is only limited by the measurement system used.

Today, this type of guidance is still in the research phase.

Acknowledgements Our thanks to the companies which have kindly collaborated, by giving

pictures and technical information. Special thanks to Mr. R. Gonzalez from Shuton and

P. Rebolledo from Redex-Andantex for their time dedicated in discussing several aspects of

this chapter.

References

[1] Etxaniz I et al. (2006) Magnetic levitated 2D fast drive. IEEJ Trans Industr Appl

126(12):1678–1681

[2] Shamoto E et al. (2001) Analysis and Improvement of Motion Accuracy of Hydrostatic

Feed Table. Annals of the CIRP, 50/1:285–288

[3] Slocum A (1992) Precision Machine Design. Prentice-Hall

[4] Takeuchi Y et al. (2000) Development of 5-axis control ultraprecision milling machine for

micromachining based on non-friction servomechanism. Annals of the CIRP, 49/1:295–298

[5] Uriarte, L et al. (2004) Rectificado de precisión de piezas helicoidales (in Spanish, Grinding

of helical parts). IMHE, 305:52–58

[6] Zaeh MF et al. (2004) Finite Element Modelling of Ball Screw Feed Drive Systems. Annals

of the CIRP, 53/1:289–292

[7] Zatarain M et al. (1998) Development of a High Speed Milling Machine with Linear

Drives. CIRP Int Seminar on Improving Machine Tool Performance 1:85–93

L. N. López de Lacalle, A. Lamikiz, Machine Tools for High Performance Machining,

159

Chapter 5

Advanced Controls

for New Machining Processes

J. Ramón Alique and R. Haber

Abstract This chapter describes the basic concepts involved in advanced CNC

systems for new machining processes. It begins with a description of some of the

classic ideas about numerical control. Particular attention is paid to problems in

state-of-the-art numerical control at the machine level, such as trajectory genera-

tion and servo control systems. There is a description of new concepts in ad-

vanced CNC systems involving multi-level hierarchical control architectures,

which include not only the machine level, but also include a process level and

a supervisory level. This is followed by a description of the sensory system for

machining processes, which is essential for implementing the concept of the

“ideal machining unit”. The chapter then goes on to offer an introduction to open-

architecture CNC systems. It describes communications in industrial environ-

ments and an architecture for networked control and supervision via the Internet.

Finally, there is a brief summary of the systems available to assist in program-

ming and the architectures of current CNC systems. There is also a description of

the most recent developments in manual programming for current CNC systems

and possible architectures for these systems, with the different uses of PCs and

their various operating systems.

5.1 Introduction and History

A numerical control is any device (usually electronic) that can direct how one or

more moveable mechanical elements are positioned, in such a way that the move-

_________________________________

J. Ramón Alique, R. Haber

Industrial Computer Science Department, Industrial Automation Institute

Spanish National Research Council (CSIC) Arganda del Rey, Madrid, Spain

ralique@iai.csic.es

160 J. Ramón Alique and R. Haber

ment orders are put together entirely automatically through the use of numerical

and symbolic information defined in a program.

The first attempt to provide a mechanical device with some form of control is

attributed to Joseph Marie Jacquard, who in 1801 designed a loom that could be

made to produce different kinds of fabrics purely by modifying a program that was

entered into the machine by means of punch cards. Subsequent examples included

an automatic piano, which used perforated rolls of paper as a way of entering

a musical program. While these devices were actually automatic controls, they

cannot be regarded as true numerical control systems.

The great leap forward in numerical control evolution came when numerical

control was applied to the machining of complex parts. The introduction of auto-

mation in general, and numerical control in particular, came about as the result of

several different circumstances: 1) the need to manufacture products that could not

be obtained in large quantities at high enough levels of quality without resorting to

automating the manufacturing process; 2) the need to produce items that were

difficult or even impossible to manufacture because they required processes that

were too complex to be controlled by human operators; and 3) the need to produce

items at sufficiently low prices.

In order to solve these problems, inventors came up with a number of automatic

devices using mechanical, electromechanical, pneumatic, hydraulic, electronic and

various other kinds of systems. In the beginning, the main factor that drove the

whole automation process was the need to increase productivity. Other factors

have subsequently emerged that, both individually and as a whole, have been

enormously important in the industrial sector, such as the need for precision, speed

and flexibility. Viability is not included here, as it has been of little importance

from a quantitative point of view, but thanks to these devices it has been possible

to manufacture parts with highly complex properties that could otherwise never

have been made.

One early attempt to apply numerical control techniques as an aid to part ma-

chining was made in 1942, in response to demands by the aeronautics industry.

However, the key advance in the automation of machining processes occurred in

the 1950s, when Parsons, a company under contract with the US Air Force, asked

the Servomechanisms Laboratory at the Massachusetts Institute of Technology

(MIT) to develop a three-axis milling machine with numerical control. This con-

troller was the state of the art in the 1960s. Since then the use of computers and

particularly the first microprocessors that emerged in the early 1970s has led to

spectacular advances, as more powerful, reliable, economical computers have

become available.

5.1.1 Computer Numerical Control and Direct Numerical Control

Until the appearance of microprocessors in the early 1970s, numerical control sys-

tems were divided into two main groups: 1) systems designed to control costly, so-

5 Advanced Controls for New Machining Processes 161

phisticated machine tools in which a minicomputer could be included as a basic

control system without excessively increasing the cost of both control and ma-

chine; and 2) small and medium-sized control systems designed for simpler ma-

chines and invariably hardware implemented. After computers began to be includ-

ed as a basic element, numerical control became known as CNC (computerized

numerical control).

Almost immediately, however, in the mid-1970s, the microprocessor began to

be used as a basic unit. This placed very strict conditions on system organisation,

the choice of microprocessor and the design of the system with different func-

tional units. In the case of system organisation, depending on the type of numeri-

cal control used, this meant the use of parallel processing techniques in the broad-

est sense of the term. In high-range equipment, parallel processing was achieved

using multi-microprocessor architectures, with some specific functions supported

by microcomputer LSI peripherals. These peripheral microcomputers considerably

enhanced numerical control performance, as they required minimum attention,

since they operated in parallel with the central microprocessors.

In the creation of multi-microprocessor systems and particularly mono-

processor systems, there were certain basic conditions that the systems in question

had to meet. The most important of these was determined by the size of the nu-

meric values to be handled, always in integer arithmetic. The use of 32-bit micro-

processors has offered an essential advantage in performing operations in integer

arithmetic.

Furthermore, as shall be observed below, these CNC systems require particu-

lar characteristics in the interpolation unit and the position control servomecha-

nisms. The algorithms to be used in the interpolation process must be reference

word algorithms, while the servo control systems must be sampling systems,

with a sampling period of T

. The value assigned to the T

sampling period is

of vital importance as regards errors in the contour shape generated by the

machine tool.

While the first microcomputer-based CNCs were being used, the first steps

were being taken on the road towards optimizing the machining process to some

degree. The optimisation of machining processes has traditionally relied on part

programs based on unreliable pre-processed data. It was soon found that it would

be impossible to optimise such processes, and particularly to implement the con-

cept of the objective function (which shall be discussed at a later point) based

solely on the use of existing part programs.

The first serious attempt at optimisation was made in the 1960s during a US Air

Force contract with Bendix Corporation. The system that was designed, which was

given the name of Adaptive Control Optimization (ACO), included on-line opti-

misation and adaptive controls that were very advanced for their time. However,

this system never went into practical use in the industrial environment, as its op-

eration was based on the existence of sensors that were not yet available, such as

tool condition sensors.

This showed that sub-optimum control systems could be more suitable and eas-

ier to use in industrial environments, leading to the appearance of adaptive control

162 J. Ramón Alique and R. Haber

systems with constraints (ACCs) and geometric adaptive controls (GACs). These

systems, which were developed in laboratories back in the 1980s, also failed to

win full industrial acceptance due to their highly limited properties.

Another advance in CNC systems came with the creation of distributed nu-

merical control architectures (DNCs), new manufacturing architectures in which

several machine tools with CNC controls were run by a central computer via

a direct real-time connection. DNC systems connect a group of machines to

a central computer that is used to program and record part programs, transmit

these programs on demand and, in general, manage the activities of the machines

in question.

The general aims sought by DNC systems are: 1) increasing the efficiency of

the programmer, the operator and the machine itself; 2) providing a flexible struc-

Fig. 5.1 Historical evolution of the DNC concept