Sorting through the somewhat confusing array of gaging systems and measurement technology on the market can be time-consuming and frustrating. With this in mind, Marposs Corp. surveyed some 250 metalworking companies on-site during 18 months to determine what shop floor gaging technologies they use and how they rate the gages' capabilities against various performance criteria.
Based on the survey results, the following overview clarifies some basic gaging terminology, outlines the primary applications of these technologies and systems, and addresses their relative strengths and weaknesses. Armed with this information, future buyers can select gaging systems that meet their needs cost-effectively and offer optimal returns on their investment dollars.
Fundamental gage components
The Automotive Industry Action Group's reference manual of gaging standards defines a measurement system as "the collection of operations, procedures, gages and personnel used to obtain measurements of workpiece characteristics." A gage is defined as "a device used to obtain measurements." This means that just about any measurement product on the shop floor can be called a gage, from hard go/no-go gages to measuring machines.
All measurement technologies include the same basic components -- a standard, a comparator and an interpreter. The standard is a master, or qualified, workpiece that serves as a reference for determining whether the part compared to it falls within specification limits. The comparator compares a workpiece characteristic to a reference. The interpreter -- which can be a human being or a complex gage amplifier -- processes information from the comparator and determines whether the workpiece characteristic is good or bad, depending on specification limits.
Beyond these fundamental components, however, measurement technologies vary widely. Before acquiring a particular gaging system, parts manufacturers should carefully assess what value it will bring to their process. This includes expected performance reliability and the level of gage effectiveness needed to justify the investment. Like a new machine tool, a gage should improve productivity, add value to the product and increase the payback from manufacturing operations.
When selecting a particular gaging system, prospective buyers must first consider whether the system meets three general application requirements: specified part tolerance limits, the type of characteristics to be gaged -- whether dimensional or geometrical -- and the manufacturing or end-product priorities vs. critical characteristics to be gaged.
Other application considerations take into account requirements for an attribute- or variable-type gage, machine feedback or closed-loop machining, the number of part characteristics to be checked simultaneously, the number of hard characteristics to be gaged or, conversely, the flexibility required to accommodate predicted process variability. Still other application criteria include system suitability for the application environment, any cost constraints, production rate and the required inspection rate.
In the Marposs survey, parts manufacturers were asked to rate currently used measurement technologies and gaging systems with respect to performance criteria that affect a system's suitability for particular applications. The criteria included gage repeatability and reproducibility; design robustness and reliability; system accuracy, flexibility and user-friendliness; measurement resolution; inspection speed; maintainability; feedback capabilities; and sensitivity to environmental factors. A rating system of 1 to 5 points was used, with 1 equaling poor and 5 equaling excellent.
Attribute-type gaging systems
Attribute type gages compare part characteristics to specification limits and either accept or reject the part based on whether the limits are satisfied. These gages often are referred to as go/no-go. They tell users only whether a part is good or bad, not how good or bad it is.
Attribute gaging is suitable for applications where the life of the tool generating the gaged characteristic is predictable and where anticipated process variation is relatively high -- a Cpk or Cp of 1.33 or more. Other criteria include applications where machine tool compensation is not feasible or needed, or where production rates are low enough to allow checking every part.
The companies surveyed gave attribute-type gaging systems an average score of 2.5, which is between fair and acceptable. Attribute gages were rated poorly with respect to gage repeatability and reproducibility because they don't give direct dimensions, only indicate whether a part is good or bad. Moreover, extrapolating a dimension from the gage finding requires the gage operator's analysis, which will be necessarily subjective.
For accuracy, attribute gages were rated a 3 because measurement resolution isn't a factor with gages that provide only a good/bad verdict. The gages were rated as poor for flexibility because they are generally dedicated to the inspection of a specific part characteristic.
Because attribute gages typically feature simple designs, they were rated as excellent for user-friendliness. The same design simplicity also brought high ratings for inspection speed and reliability -- which translates to cost-effectiveness. The rating for maintainability was poor, however, because attribute gages must be reworked mechanically once they are chipped, damaged or dropped. Typically, they must be recertified in a gage lab.
Machine feedback isn't applicable with attribute gages because they don't provide an absolute number with which to offset the machine. Finally, the gages were rated acceptable in terms of sensitivity to environmental factors such as chips, coolant and temperature.
Variable-type gaging systems
Distinct from attribute gages, variable-type gages provide a quantitative value for the part characteristic being checked. For example, if the nominal dimension for a machined part diameter is 1.125", variable-type gages will give a numerical measurement that indicates exactly how close a measured piece is to nominal. The measured value can be compared to the specification limits, which helps in qualitative decision making about the machined characteristic.
Variable gages include hand micrometers, bench fixture gages, automatic machines and coordinate measuring machines. Such gages often are used for maintaining part size in a grinding or machining process by feeding back measurement data used for tool-size compensations.
Variable gages are suitable for three other applications: optimizing tool life and uptime by gaging the part and making corresponding machine adjustments; reducing scrap rates and increasing productivity; and processes where there is a concern that a Cpk or Cp of 1.33 or less will occur without gaging and proactive intervention. Surveyed gage users rated variable gages between 4.3 and 4.6, which is good to excellent.
Dial indicators received an acceptable-to-good overall rating of 3.6. This included acceptable ratings for gage R&R and excellent ratings for user-friendliness. Inspection speed was rated only fair, reflecting the difficulty operators often have in reading the indicators on the plant floor. Because dial indicators are manual systems that require operator decisions relative to machine adjustments, machine feedback capabilities were rated very low. Environmental sensitivity, on the other hand, was judged good or excellent.
Linear variable differential transformer gages were rated on average 4.5, one of the highest scores overall. Ratings for individual characteristics of this technology, including gage R&R, accuracy and measurement resolution, proved uniformly high. With respect to measurement resolution, some LVDT transducers can resolve measurements to as fine as 0.25 micron, and certainly to the range of 4 millionths of an inch. The lowest, though still acceptable, rating given this technology was for maintainability, reflecting these gages' susceptibility to coolant impairment and wear and tear.
There are several types of LVDT gaging systems, including standard line pencil probes and measuring cells. Some standard pencil probes and Marposs resistive technology gages (a form of strain gage technology) offer repeatability performance in the range of 4 millionths of an inch. The MRTs, in addition, limit thermal drift to 4 millionths of an inch per degree Fahrenheit while providing outstanding linearity, settle time and inspection speed. Only 1/2" and 3/16", they are also more compact than the pencil probes, which are generally either 3/8" or 8 mm in diameter and between 0.5" and 5" in length, depending on the model. The measuring cell, however, is the most reliable and durable of the LVDT technologies.
Air-to-electronic variable gages
Electronic/pneumatic gages based on this technology operate as comparative inspection systems, using pneumatic line pressure to make dimensional, geometrical and positional checks on production parts. Part size is determined based on the amount of air that escapes between the comparator, or the air gage itself, and the part. The sizes obtained, however, are susceptible to measurement error induced by the part's surface finish.
An air-to-electronic gaging system's components include the comparator and a part-measuring detail that contains one or more pneumatic orifices, such as a nosepiece, for inside diameter inspection. Another component is the pneumatic sensor, or air-to-electronic converter, which translates the line pressure to electronic size data. Air gaging also requires up to two standard, or master, parts as well as an interpreter, which is the electronic amplifier that processes the electronic signal to establish meaningful measurement data.
Air-to-electronic measurement systems are used for applications requiring noncontact with the workpiece and only simple dimensional or geometrical checks. Because typical air gage measurement ranges are well under ±0.004", the systems generally are used where very small tolerance ranges of less than ±0.003" are expected. These applications require a smooth and ample measuring surface.
These gaging systems also work for applications where a low initial investment is required and flexibility isn't. Air-to-electronic gages must be specifically designed for particular applications. They work best where inside diameters are very small or deep, where the ratio of the bore size to depth is small, or where surface averaging of inside or outside diameters is desired.
In the Marposs survey, gage users rated air gage technology at between 3.4 and 3.5. The highest ratings were for gage R&R capability and accuracy. The lowest rating was for flexibility, reflecting air gages' very limited measuring range and tailor-made design specifications.
Comparing this noncontact technology with electronic contact gages can be useful. In terms of measuring range and accuracy, electronic contact gages are clear winners. Their linear range is about 0.200", compared with only 0.003" for air-to- electronic systems, and electronic contact gages can be accurate over this range to within about 2 millionths of an inch. These gages also have a faster response because it takes time for the pressure to build up and change in air-to-electronic systems and for the sensor to detect it. The electronic contact gages also are unaffected by surface finish roughness because they use a radius contact that bridges the RMS value of the surface finish.
On the other hand, air-to-electronic gages have better thermal stability. As noncontact devices, they also avoid leaving a witness mark on the part, unlike some electronic contact gages. Most importantly, air-to-electronic systems are simply designed, which makes them easier to maintain and results in fewer service problems.
This measurement technology features an electronic contact probe mounted on a machine tool. Because touch probing relies on the coordinate feedback system of the machine to produce a measurement reading, it is rejected by many quality control people who can't accept the concept of measuring a part on the same machine that is making it.
Actually, these skeptics have a valid point -- in terms of taking measurements, plotting the data on a chart and doing an SPC analysis. The touch probing system's accuracy depends on the machine tool's positional accuracy. However, the touch probe, offering a system locational repeatability of better than 0.00015", often is a viable solution for controlling the machining process. Applications of these systems include pre-process setup of machine workshifts, in-process size control, post-process statistical size control and on-machine tool qualification, all of which are discussed below.
In the survey, gage user ratings of touch probing systems fell between 4.1 and 4.4. The performance criterion rated most highly was flexibility. When programmed correctly, the touch probe system offers as much flexibility as the computer numerical control machining center or lathe on which it is used. Generally, the system will provide process control for any parts cut within a machine tool's work envelope.
The lowest-rated performance criterion for touch probes was gage R&R capability, but this requires some interpretation. For in-process probing on machine tools, a 20-percent gage R&R capability -- only half as good as that required for a post-process gage -- may be perfectly adequate. Process control is possible with touch probes on CNC lathes, for example, whenever the dimensional tolerance to be maintained is 0.001" or greater. If the tolerance is tighter than 0.001", post-process gage fixtures usually will provide a better solution.
Optical-electric variable gages
Optical gaging systems use light to make dimensional, geometrical and positional checks on production parts. These systems consist of four principal components: a light emitter, a receiver that converts the light to an electrical signal, a series of optical lenses and an electronic amplifier that processes the signals and establishes meaningful measurement data.
Optical measurement technology is suitable for parts inspection in operations where noncontact with the workpiece is critical, where a large measuring range must be covered without retooling, where machining is performed at high operating speeds and where workpieces can be supplied to the gage clean and dry.
Optical-electric measurement technology comes in a number of gaging formats. One of them is light-intensity comparative gaging, now almost obsolete. Other types include laser scanning gages; shadow-cast, or CCD array, gaging; and laser diffraction gaging. Laser diffraction gaging is limited to a very small measurement range and generally is used only for wire inspection. In the survey, optical gaging received a good overall rating of 3.75. High marks were given for gage R&R capability, flexibility, user-friendliness and inspection speed. The optical gages were rated very low for environmental sensitivity, however, due primarily to their susceptibility to coolant.
Using variable-type gages to control the metalworking process effectively is especially important today. Along with tighter-than-ever tolerance specifications, we now have CNC capabilities, spindle speeds, superabrasive cubic boron nitride grinding wheels, and CBN and whisker-type inserts capable of very high productivity rates. Also, the latest CNC machines have improved capabilities for parts production.
But ability, or potential, remain the key words here. Both large and small manufacturers are learning that, even though the CNC machine can produce parts faster and to better tolerance, it often can't produce the parts consistently without closed-loop process control. Many process variables still adversely affect the metal-cutting process. Along with material variances such as part hardness and excessive or insufficient stock, there are also variances due to part drift, machine deflection and machine thermal instability.
Another major variable is the machine operator. Operators can either over-control or under-control a process, making subjective tool-compensation decisions that differ from one operator to another. There's also the variability caused by different operators running the same machine and part at different times.
Insert quality itself is still another variable. Although pressing technology has greatly improved in recent years, inserts still display batch characteristics, providing variable performance.
Closed-loop machining and process control provided by variable-type gaging systems can eliminate these and other process variables. Closed-loop machining also offers the advantage of predictability -- because it makes the same compensation decisions the same way every time, based on all process variables. The variables are all summed up in automatic decisions aimed at what's really important: the quality of the part being manufactured. In addition, individual machine processes can be targeted to very narrow tolerance bands. Generally, for parts that go through many processes, the tighter the tolerances maintained in early stages, the easier it will be to maintain quality later.
Closed-loop machining automatically feeds gage information directly back to the machine producing the part for establishing size control. This helps minimize errors introduced into the machining operation through the up-and-down size oscillations caused by too much tool control. The small, but firm, tool-wear offsets offered by closed-loop machinery provide the best size control.
Process control for metal-cutting
There are four kinds of closed-loop process control:
Pre-process machine control -- This measures a part before it's machined and transmits the data to the machine CNC. The measurement can be made by something as simple as a handheld gage, a bench fixture or a machine-mounted touch probe. Generally, this type of tool compensation is not statistically based. Its purpose is to confirm whether a part is loaded or is the correct part.
Pre-process control -- This allows you to set a machine work shift. For example, with multiple parts on a gang fixture, individual work shifts can be set for each part, using a touch probe or a handheld gage. Controlling thermal drift to determine the amount of material to remove is especially important for a cast part. It's also possible to control the process to fit a mating part by measuring a shaft and turning the bearing surface to fit it.
In-process machine control -- This makes preliminary cuts on the part, leaving a certain amount of material behind. The part is then measured, and the depth of cut for the last pass or several passes is calculated automatically. The measurement can be made by a handheld gage, a machine-mounted fixture or a machine-mounted touch probe. The procedure is effective for all machines and parts.
Post-process control -- This measures the part after it's machined and transfers the data to the CNC. Again, the gage data can be produced by simple gages, but in post-process applications, the data will almost always represent a statistically based averaging. These applications are most effective on CNC lathes and commonly employ touch probes and/or post-process gages.
In all post-process applications, the gaging system or fixture can be manual, semiautomatic or automatic. Manual systems often will use multiple fixtures that employ standard gages to measure the part in stages. Many different strategies and devices can assist the operator using these manual systems, including tactical feedback from lamps located by the machine the part came from, go/no go indicators and logic locks that prohibit the operator from manufacturing another part until he or she measures the one just made. Gage amplifier displays give the operator the operation's complete status.
For machine compensations in post-process gaging, the parts are measured and the running average of each dimension calculated. The process running average is accumulated and compared with upper and lower tool compensation limits. When the process running average exceeds the tool compensation limit, the software automatically initiates an absolute tool offset to the nominal part size or, when programmed, to the target part size. If an individual measurement violates the upper or lower reject limits, the process is stopped.
There are several reasons for programming a target value beyond nominal. If an operator continuously compensates down to nominal only, the parts distribution aggregates on one side of nominal, which is unacceptable statistical practice. When an operator compensates down to the target value, the parts distribution aggregates around nominal, which is acceptable and also allows the machine to run longer between tool comps.
The system also can track tool wear. When the wear limit is reached, a call-up for the redundant tool can be triggered automatically. The tool-replacement function also can be integrated with a tool-life management package based on a running average of tool wear.
Although any type of measurement can be made, the CNC machine generally will accept offsets only for diameter, length, distance and, in some cases, taper. In some applications, it's possible to split an offset mathematically between the x and z axes to compensate for contour. In all cases, the measurement system should be flexible enough to handle any part within the working envelope of the CNC machine.
Closed-loop machining offers the major advantage of predictability. Because this form of process control makes compensation decisions the same way every time, it allows production managers to target the machine process to narrower tolerance spans. This yields the benefits of an important principle: the less tolerance used up in a part early in the process, the easier it is to maintain high quality in subsequent manufacturing operations.
About the author
Tom Stewart is sales and marketing manager in the standard products division of Marposs Corp., located in Auburn Hills, Michigan. He can be reached by fax at (248) 370-0621.