In our continuing effort to provide the most accurate information to the research, defense and industrial communities, we have expanded many of our product specifications to reflect the true performance users can expect. We refer to these as "guaranteed" and "typical" specifications. This Motion Control Metrology Primer provides a thorough discussion of what these guaranteed and typical specifications mean as well as how they are derived and verified. Knowing how specifications are derived can give you a better understanding of the true performance of a product than simple numerical values. This eliminates the mysteries in "specmanship" and allows you to select the best solution for your application.
This is our promise!
For Motion, Think Newport!
Guaranteed and Typical Performance Specifications
Newport's Guaranteed Specification
Although not explicitly stated in the past, our motion products have always been held for release until their production specifications could be guaranteed. The guaranteed specifications are the worst-case values that users can expect for this particular product, tested per Newport Metrology Procedure following ASME B5.57 and ISO 230-2 standards. Any motion product that exceeds these guaranteed specifications is either re-worked until it meets them or scrapped if there is no possibility of meeting the specifications.
Newport's Typical Specification
Typical specifications have been provided for our most popular motorized products and are based upon initial or ongoing statistical production results. Typical values shown can be considered to be the mean performance values for that particular specification, and for some stages, the statistical test data is presented. This specification provides the users with a more realistic description of the performance of a particular stage family. For instance, the VP25XA series stages have a guaranteed Bi-directional Repeatability specification of 0.2 µm but in reality historical testing has shown this stage typically provides 0.15 µm Bi-directional Repeatability performance.Figure 1. Bi-directional Repeatability data
Newport's Motion Control Metrology
For almost 50 years the Newport Resource has been the reference standard used by researchers, scientists and engineers to advance their technologies, discoveries and products. In today's world of shrinking feature sizes, nanometer-scale devices, as well as higher accuracy and resolution requirements, it is critical to use the highest quality components available. This is why Newport provides testing and test data for on-axis accuracy, repeatability and trajectory errors (e.g. Pitch and Yaw for linear stages and Wobble and Eccentricity for rotation stages) on every XM, GTS, IMS, ILS, MTM, UTS, MFA, VP, RGV, RV, URS, PR/SR, LTA and TRA products. A sample test data sheet can be found in Figure 2 below.Figure 2: Sample Test Report
As mentioned above, Newport provides test data for on-axis accuracy, repeatability and linear/rotational errors on every XM, GTS, IMS, ILS, MTM, UTS, MFA, VP, RGV100BL, RV, URS, PR/SR, LTA and TRA product sold. The final performance of a positioner depends on many factors in the manufacturing process, such as the precision of machined parts or tolerances in the assembly. A rigorous quality control and metrology procedure is the only way to guarantee that a stage or actuator meets its published specifications.
Newport's internal metrology procedure is a very comprehensive combination of other widely known procedures including ASME B5.57 and ISO 230-2. The following are the major components of Newport's test procedure, which assures our customers receive only the highest quality products:
- Controlled test bench data taken for 21 measurement points distributed over the full travel of the stage. This high number of measurement points ensures careful identification of the stages characteristics along its entire travel range. A typical positioner may perform very well along specific points along its travel, but it might have significantly different characteristics at other points due to imperfections in the bearings or in the drive screw. Procedures that collect only a few data points (4 or 5 for example) do not accurately represent the true performance of the positioner and may not truly meet published specifications along its full travel range. Newport's dedicated metrology room is temperature stabilized to 19.5°C +/-1.5°C. All stages undergoing testing are stored in the room for at least 12 hours for stabilization before the actual measurements are taken. A certified third party metrology institute provides regular calibration of the test tools to guarantee the highest precision possible.
- Controlled test bench data taken for 4 complete cycles of motion, in both directions.This element is not only essential to verifying the repeatability and reversal errors of the positioner, but also provides a more precise qualification of positional accuracy. This type of test also confirms that the characteristics of the positioner are consistent and eliminates any "lucky" test results.
These two tests combined provide 168 points of data that are used to determine the positioner's true performance. The statistical methods used to evaluate this data are presented in Figure 3.Figure 3. Statistical Equations
- These statistical methods equate to stated specifications, which are the MAXIMUM Uni-directional Repeatability, the MEAN Reversal Value and the MEAN Bi-directional On-Axis Accuracy. These provide the best representation of true positioner performance that users can expect from repeated use of the product.
- Uni-directional repeatability is specified at a factor of 3 standard deviations (3-sigma). The standard deviation is calculated from the positioning data taken at all 21 measurement points in the forward and reverse directions. With 4 cycles of motion, this makes a total of 168 data points. The high number of data points and the 3-sigma approach provides a very accurate indication of the worst-case repeatability with 96% certainty. This means when moving to the same position hundreds of times, 96% of the moves will be repeatable within the specified span.
- NOTE: It is important to clarify that Newport specifies positioning products as the "guaranteed worst-case repeatability" as opposed to some companies that specify an RMS (Root Mean Square) or mean repeatability to make their data look more impressive. RMS or mean repeatability is the expected repeatability error when moving to the same position many times. It converts to only a 1-sigma specification certainty. This significant difference means that when comparing a product with a mean RMS repeatability of 0.5 µm this same product would have a 1.5 µm specification using Newport's A167 procedure, which is more representative of the actual performance of the product.
- 100 data points are taken for pitch and yaw of linear stages, as well as, wobble and eccentricity of rotation stages using a Newport LDS1000 autocollimator. These data points are taken dynamically, on the fly, along the travel. The performance specification presented in the test report is the true magnitude of the error (the difference between the max and min error).
- Wobble and eccentricity values of rotation stages are measured during 2 full rotations. Taking data during one rotation only will not detect all the runout errors as the bearings rotate at only half the speed of the stage.
- All test tools are mounted on the center of the carriage at a height of approx. 50 mm, which closely mimics typical applications. Repeatability, reversal value, and accuracy for linear stages are measured in comparison to a calibrated laser interferometer. On rotary stages, repeatability, reversal value, and accuracy are measured in comparison to a high accuracy rotary encoder that is coupled to the rotating platen of the stage.
- Test reports verifying performance of every XM, GTS, IMS, ILS, MTM, UTS, MFA, VP, RGV100BL, RV, URS, PR/SR, LTA and TRA product, (as shown in Figure 2) are supplied AT NO ADDITIONAL COST to our customers. These test reports not only verify the performance of our products, but they also provide sufficient detail to allow users to improve the performance of their positioning products through backlash compensation, error mapping or linear compensation.
The old practice of evaluating products based solely upon product catalog specifications such as repeatability, accuracy, pitch, roll, and yaw as supplied by the manufacturer, can be a very risky way to choose a product especially in the submicron and nanometer performance range. In today's ultra-high precision environments, it is imperative to have a clear understanding of the origin of the motion control specifications. Knowing how they are derived can be ultimately more telling than the numerical value itself.
Upon request, Newport will supply historical test data for most of its motorized positioners. For more information on Motion Control specifications and technology, please review our Motion Control Tutorial and our Application Notes at www.newport.com, or contact your local Newport Applications Engineer