Fig. 2. – Examples of block tooling.
Datum And Location Checks
Workholding devices are designed to hold a workpiece, restricting movement in all six degrees of freedom. It is not desired that the device damages the workpiece and it must achieve these two aspects with speed and simplicity. In an automated system it would be necessary for the clamping procedure to occur without human intervention. This can be achieved through the use of hydraulics or pneumatics, which can be controlled electronically. It is also necessary that the workholding device provide repeatability for recurrent products of the same type. This is particularly important in an automated process where any misalignment in the workpiece can result in the datum for that workpiece moving and scrap, potentially, being produced. Workpieces that are to be produced from castings are generally mounted on fixtures or jigs, which have their own clamping methods. If it is not possible for the datum to be maintained between recurrent components then a system of three-dimensional co-ordinate sensing can be used to identify the location, datum and orientation of the workpiece. This operation, obviously, is performed on the CNC machine prior to machining but does extend the production time for each component.
Grid plates are a means of rapid set-up for components with straight or round edges and that do not require machining around the lower sides. The plates consist of a cast and ground slab of metal (usually steel or cast iron) and precisely drilled holes that are alternatively reamed or tapped. Each hole is referenced by and alphanumeric code that can be referenced as a datum. The tapped holes are used for clamping purposes. See. Fig. 3.
Fig. 3. – An example of a Grid plate
The grid plate is clamped to the bed of the machine and then the workpieces are clamped to the plate. With a large enough grid plate several jobs can be set up at once and whilst one job is being machined another one can be clamped to the plate out of the way of the machining process. The differences in datum’s are able to be programmed into the CNC control with a datum shift by stating the alphanumeric code for the new positions. In the above figure the alphanumeric code for the datum would be A1.
Pallets can be used for milling and drilling operations instead of grid plates and are often used in modern machining workshops. Whilst one workpiece, mounted on one pallet, is being machined another is being mounted on a second pallet outside of the machining area. When the piece being machined has finished then the pallets are swapped and the whole process repeated. Using this process means that machine downtime is kept to a minimum whilst ensuring maximum safety to any personnel passing by. To further increase time efficiency, in larger cells the pallets may pass from machine to machine for extra operations to be performed before the component is removed from the pallet.
In Process Measurement
In process gauging is a system whereby the workpiece quality can be continuously monitored whilst the piece is being machined. Data is returned continuously to a control, usually the CNC’s control, where appropriate action can be taken if the workpiece is exceeding specified tolerances. This is known as adaptive control. Detection of items such tool breakage can be achieved through the use of force monitors attached to the tool. If the tip should break then the force monitor will detect a greater load being applied to the tool and the control can determine that it needs to remove that tool from the process to prevent damage to either the machine or the component.
If the workpiece should start moving out of tolerance then corrective measures can be taken to bring the process back into control. In process measurement can use instruments such as probes or acoustical force sensors.
Post Process Inspection
Post process inspection can take the form of either “near machine” and “off machine” post process gauging.
The “near machine” post process is a widely used method of measuring the quality of a component and sending any corrections to the CNC control to compensate for situations such as -:
- Tool wear
- Machine variability as environment changes temperature
- Change in material for same profile
The process can be performed by the use of tools such as verniers or micrometers, some of which have a special port specifically for this purpose. The port is known as an SPC output. SPC stands for Statistical Process Control and the same readings that would allow the CNC control to modify its settings can be used to generate SPC charts for monitoring quality (see Qualitative Data and Attributes). This method is not usually used when short cycle times are in effect but generally when high volumes of parts exist.
“Off machine” post process gauging is generally used when there is a wide part mix or if short runs are used. This is because this process offers greater flexibility and does not take up much processing time. Using this method, however, does present longer measurement cycle times and does not allow for swifter response to dimensional changes provided by the in process system. CNC co-ordinate machines are often used and form a closed loop for the assessment of part quality.
Qualitative Data and Attributes
A means of measuring the quality of a component is through the use of statistical process control (SPC). Statistical analysis of a process can provide a company with valuable information about what is happening within a specific process. Qualitative data is defined by using variables, which are very specific. See Statistical Analysis.
Variability is a process may occur for many reasons, including -:
- Operator error – setting, accurate positioning
- Machine errors – bearing wear, clearance
- Material inconsistency – hardness, composition, size
- Tooling deviation– design, maintenance, wear
- Methodology problems – feeds, speeds
- Environmental changes – temperature, power supply
Any one, or combination of, these effects can cause a process to go out of control and the component to be made incorrectly. When this occurs the cause must be identified and rectified.
Statistical Analysis
In SPC data is obtained from attributes and variables, which are defined as -:
- Attributes – Only has two states. It is either acceptable or not acceptable. E.g. the wall will be painted red – it either is or it is not.
- Variables – Has a numeric value as an output. E.g. Length, mass, temperature, voltage, etc.
Attributes are very subjective since what one person calls red may not be same as the next person. Due to this inherent problem attributes are not often used but if they are then their data is determined by making comparisons with an agreed standard. Variables, on the other hand, are objective and the measuring equipment provides empirical evidence. Variable, or qualitative, data is generally more expensive because skilled personnel are required to use the measuring equipment.
The data that is obtained, be it variable or attribute, needs to be translated into a form that can be easily read. This can be achieved through the use of control charts. There are different types of control charts, which are listed as -:
- Mean and range charts
- Mean and standard deviation charts
- Median and range charts.
The most common charts used are mean and range charts. There are two types of chart associated with each kind of data. They are -:
- Variables – n and p charts, where n is the mean chart and p is the range chart.
- Attributes – c and u charts, where c is the mean chart and u is the range chart.
These charts are developed from the data obtained and limits are calculated and applied to the charts. If a piece of data exceeds those limits then actions can be employed to remedy the cause of the spurious data.
Data Capture
Methods of Data Capture
Data capture is the method of obtaining data with or without human intervention. In an automated system, such as being discussed in this assignment, the retrieval of data must occur with out the aid of humans in the machining cells. There are two forms of sensing that can be used. These are -:
Tactile sensing refers to sensors that have to make physical contact with the data transmitter to obtain the information such as annular rings. Non-tactile sensors are the inverse or tactile sensors and do not require direct contact but can scan from a short distance from the transmitter. These can take the form of microchips, barcodes or acoustic emission sensors. Three examples of tactile and non-tactile sensors are given below.
Examples of Data Capture
Annular Rings
This system uses a series of rings mounted on the toolholder. These rings are of equal thickness but are of two different diameters. The tool is passed over a reading head that consists of a series of levers mounted on microswitches. The levers that contact the large rings slide out the way activating their appropriate microswitches whilst the small rings do not make contact with a lever. This allows a binary signal to be generated that can be translated into a tool identification (see Fig. 4.).
Fig. 4. – Example of how annular rings work
This system is limited in that there is only so much information that can be presented to the reading head. The tool identification number and the tool diameter are about the maximum amount of data that can be realistically transmitted. No information regarding tool offsets could be included as the decimal value of the offset can vary so much; hence the binary conversion could be very long.
Microchip
Tool holders with silicon chips inserted in them can be used to transmit data about the tools that they are carrying. A comprehensive set of information about each tool can be held within the microchip. This can include the -:
- Tool number
- T.L.O. (Tool Length Offset)
- Tool type
- Wear on tips
- Speeds and feeds to compensate for wear
- Time in service
- Location of placement in CNC tool turret (for error checking)
This data can be written to the chip at the tool room and is then able to be read
at any desired point along the manufacturing line. The chips are re-writeable to allow different tools and, therefore, different information to be placed into the toolholder. The data can be used by the CNC machines to -:
- alter the T.L.O.’s to the new values of an introduced tool
- to ensure the tool is placed in the right turret position on the CNC machine
- vary the speeds and feeds to compensate for tool wear.
The data can also be passed back to a CPU (central processing unit) for tool management which, among other aspects, ensures that the tool is not in service too long without being inspected by an operator in the tool room. Data could also be written to the chip at the CNC machine to inform the overall system of a tool breakage. The tool can then be removed from service and not inserted into another machine until it has been inspected and repaired. Microchips can currently hold 512 bits (1/8th of a byte) of information which is more than adequate for most applications.
Barcode
This is a “one shot” method of transmitting data about the tool in a toolholder. A label is printed and then stuck on the tool with certain information about the tool. Like the annular rings the quantity of information is limited by the physical size of the label. However, unlike the microchip and annular rings, the data can be transmitted to a reader from a distance of up to several metres through the use of a laser to scan the labels (similar to the method of reading barcodes at the supermarket checkout). In the environment that toolholders operate in though there is a distinct risk of labels falling off whilst in transit on the magazine or when the tool is in use in the CNC machine. This means that a form of error detection would be required with this system. This could take the form of – if a tool is detected without a label on it then it is returned to the tool room for a new label immediately. This could, potentially, mean machine downtime if there is no alternative tool that can be used already in the magazine.
Fig. 5. – Matrix comparing sensor methods
Data Transmission Features
Data transmission occurs in many forms throughout a company. When related to CAM systems the most common form of transmission occurs through the RS-232 ports. The RS-232 interface was formulated in 1969 by the EIA (Electronics Industries Association), Bell laboratories and other manufacturers of communications equipment as a form of serial data transmission. After it induction it almost immediately went through revisions to become RS-232-C and then recently RS-232-D. The RS-232 interface was developed for one single purpose that was clearly defined as -:
Interface Between Data Terminal Equipment and Data Communications Equipment Employing Serial Binary Data Interchange
Every word in this definition is significant. The Data Terminal Equipment (DTE) refers to the computer whilst the Data Communications Equipment (DCE) refers to the equipment receiving the transmitted signal, in this case the CNC machine.
In its simplest form, the RS-232 interface consists of only two wires; one to carry the data plus a circuit common often ambiguously referred to as the “signal ground”. The circuit common provides an absolute voltage reference for all the interface circuitry and must be connected on every RS-232 connection no matter how simple or complex. The RS-232 interface is not generally, however, a simple form. There are, in a fully connected cable, 21 separate wires linking a DTE to a DCE. This allows bi-directional data flow, handshaking and error detection to take place.
Transmitted signals can take the form of -:
- Simplex
- Half duplex
- Full duplex
In a simplex system data can only flow from the DTE to the DCE whilst in a half duplex system the data can flow in both directions but not at the same time i.e. the DCE has to wait for the DTE to stop transmitting before the DCE can transmit. In a full duplex system the data flow is in both directions and can occur at the same time.
Handshaking is the term associated with the DTE and DCE being ready to transmit and receive data, at what speed the information the data will be transmitted and what form of error checking will take place. Error checking is used to ensure that the data has not been corrupted between the point where it leaves the DTE and arrives at the DCE. For many CAM systems it is possible to eliminate many of the wires reducing their number down to three, one line to transmit, one to receive and the signal common. By reducing the number of wires, thereby reducing the cost of installation, it is necessary that the DTE and DCE are set up to the same configurations to prevent transmission and reception problems.
RS-232 interfaces can be run over long distances but the greater the run of cable the slower the signal will have to travel to prevent signal degradation. In a very non-scientific experiment performed in 1983 the distance a signal could travel at a given speed (baud rate) was determined and the results are listed on the next page.
RS-232 stands for Recommended Standard number 232 from the Electronic Industry Association.
DTE’s and DCE’s that are connected together are considered to be in a topology or network. There are several forms of topology that networks can take. These are -:
Star topology is where the DTE is located at the centre of a group of DCE’s and the data cables spread radially from the DTE to the DCE’s. One cable is required for each of the DCE’s which means that the DTE must have an equal or more number of RS-232 ports to the number of DCE’s in order to accommodate each of them. This process of connecting DCE’s to a DTE is more expensive as a special card is required for the DTE to give it the required number of RS-232 ports and also the extra cable would increase the initial outlay. However, it its favour, fault finding is extremely easy to achieve, as only one line of communication needs to be investigated at a time.
Bus topology takes the form of one single cable which runs the length of the DCE’s with each DCE (along with the DTE) joining that cable at its own individual junction. This reduces the amount of cable required and also the number of ports at the DTE. However, bus topology typically operates in half duplex mode with a synchronous transmission control scheme.
In ring topology a network cable passes from the DTE to a DCE and then round to any other DCE’s in a loop or ring before returning to the DTE. The data transmission rates for ring and bus topologies are between 1 to 10 megabits per second and is ideal for use by interconnecting local communities of computer based equipment.
The final topology is the hub. This is a variation of the ring or bus topology and although it may have the appearance of a star topology it is actually where the bus or ring wiring has been brought into a central unit (the hub).
-o-o-o-o-o-
Appendix A
Acknowledgements
Cambell, Joe: The RS-232 Solution – How To Use Your Serial Port.