Post on 10-Jul-2020
Procedure for performing linear measurement
Linear measurement is the most common form of measurement performed with a laser. The laser system measures linear positioning accuracy and repeatability by comparing the position displayed on the axis read-out with the true position measured by the laser system.
This section leads you through a practical exercise where you calibrate the positioning accuracy of a linear axis of a machine
The steps required to perform a linear measurement are as follows:
Linear measurement
User guide
This manual is available in English only
© 2000-2001 Renishaw - 14 August 2001 Issue 5.1
IMPORTANT - please read the SAFETY section before proceeding.
Set up the laser system to perform linear measurements.
Align the laser beam with the machine s axis of travel. Activate the automatic environmental compensation and ensure that the correct material expansion coefficient has been entered into the software.
WARNING
IN LINEAR MEASUREMENTS, THE USE OF ENVIRONMENTAL COMPENSATION IS STRONGLY
Page 1 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
Specifications
- Gives a full specification for linear measurement accuracy.
Procedure for performing linear measurement - Issue 5.1
© 2000-2001 Renishaw
Linear measurement set-up
A typical system set-up for measuring linear position is shown in Figure 1. Select the text labels on the set-up picture to obtain more information on the components in the system.
Figure 1 - Typical system set-up for measuring a position
RECOMMENDED. IF AUTOMATIC ENVIRONMENTAL COMPENSATION IS NOT USED, VARIATIONS IN THE ATMOSPHERIC CONDITIONS CAN LEAD TO SIGNIFICANT MEASUREMENT ERRORS.
WHEN ENVIRONMENTAL COMPENSATION IS USED, LINEAR DISPLACEMENT MEASUREMENTS MAY BE MADE TO AN ACCURACY OF BETTER THAN ±1.1 ppm.
Measure and record the machine's linear errors.
Analyse the captured data. Before analysing to your preferred National or International standard, examine the captured data with the All data plot' graph type to inspect for measurement errors. This is discussed further in Factors affecting accuracy of linear measurement.
The Principles of linear measurement section describes how linear optics work and how they should be set up to perform linear measurement.
Page 2 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
Perform the following steps to set up the laser system for a linear measurement:
Connect together the ML10, EC10 and interface card. Plug one end of the datalink cable into the 5-pin socket on the PC10/PCM20 interface card and the other socket on the rear of the ML10 laser. Then connect the EC10 to the interface card in the same way. The two 5-pin sockets on the PC10/PCM20 interface card are common and therefore it does not matter which socket the ML10 or EC10 is connected to.
For safety, the shutter of the ML10 laser should initially be rotated to its closed position shown in Figure 2 below.
Figure 2 - ML10 shutter position - no beam emitted
Apply mains power to the ML10 laser and the EC10 and PC. The order in which power is applied is not important. Allow the ML10 to stabilise. This will take approximately 10 to 15 minutes.
Linear measurement set-up - Issue 5.1
© 2000-2001 Renishaw
If you have not already done so, install the calibration software. In addition, you must ensure that one of the following Renishaw interfaces has been installed and configured on your computer:
PCM10 or PCM20 (PCMCIA) card for notebook computers
PC10 card for desktop computers
Attach the linear optics to the machine to be calibrated. Typical linear optical set-ups for different machine configurations are shown in the Linear measurements set-up section.
Mount the ML10 laser head on the tripod.
Connect the environmental sensors to the EC10. Place the EC10's air sensors in suitable positions on or around the machine. Place the material temperature sensors in suitable positions on the machine.
WARNINGS
1. TO AVOID RISK OF EYE DAMAGE, DO NOT STARE INTO THE OUTPUT BEAM.
2. DO NOT LET THE BEAM PASS INTO YOUR EYES OR ANYONE ELSE'S, EITHER DIRECTLY OR BY REFLECTION FROM AN OPTICAL ELEMENT OR ANY OTHER REFLECTIVE SURFACE.
Run the linear data capture software Align the laser beam with the machine's axis of travel
Page 3 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
Linear measurement optics
Figure 1 - Linear measurement optics
The linear measurement optics are used for measuring linear positional accuracy. A linear measurement optics kit as shown in Figure 1 consists of:
beam-splitter two linear reflectors two targets to help with optical alignment
Note: When you combine a beam-splitter and linear reflector it becomes a linear interferometer.
To perform linear measurements, you may also need an optics mounting kit and suitable clamps for fastening the optics to the machine to be calibrated.
Renishaw linear optics use lightweight aluminium alloy to reduce machine droop and minimise thermal lag so optics can stabilise more quickly.
See the specifications section for the full specification of linear measurement.
The dimensions of the linear optics are shown in the dimensions and weights section.
See correct orientation of reflectors for a description of how to use the red dots to correctly orientate the reflectors.
The linear and angular optics combination kit is also available for users who also want to make angular measurements.
Linear measurement optics - Issue 5.1
© 2000-2001 Renishaw
Principles of linear measurement
For details of how to perform linear measurements using these optics, refer to the linear measurement section.
Page 4 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
Figure 1 - Optical set-up for linear measurement
To set up for a linear measurement, attach one of the linear reflectors to the beam-splitter with the two captive screws provided. This combination element is known as the 'linear interferometer' and it forms the reference path for the laser beam. The linear interferometer is positioned in the beam path between the ML10 laser and the linear reflector, as shown in Figure 1. The beam-splitter housing is marked with two arrows showing its orientation. The heads of the arrows should point towards the two reflectors as shown above.
Figure 2 - Principle of measurement
The beam from the ML10 laser enters the linear interferometer, where it is split into two beams. One beam (known as the reference beam) is directed to the reflector attached to the beam-splitter, while the second beam (the measurement beam) passes through the beam-splitter to the second reflector. Both beams are then reflected back to the beam-splitter where they are re-combined and directed back to the laser head where a detector within the head monitors the interference between the two beams.
During linear measurements, one of the optical components remains stationary, while the other moves along the linear axis. A positional measurement is produced by monitoring the change in optical path difference between the measurement and reference beams (note, it is a differential measurement between the two optical components and is independent of the position of the ML10 laser). This measurement can be compared to the read-out from the scale of the machine under test to establish any errors in the machine's accuracy.
Page 5 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
Usually, the reflector is set up as the moving optical component, with the interferometer as the stationary component as shown in Figure 2. These roles can be reversed, but this reduces the maximum range over which measurements can be made from 40 m (133 ft) to 15 m (49 ft). Therefore on long axes, the linear interferometer is normally kept stationary and the other reflector is moved to take the measurements. On shorter axes, these roles can be reversed if it's more convenient.
Principles of linear measurement - Issue 5.1
© 2000-2001 Renishaw
Mounting of linear optics
Attach the linear optics to the machine to be calibrated.
To mount the linear optics you will need an optics mounting kit and suitable clamps for fastening the optics to the machine under test. Figure 1 shows a typical set-up for mounting the linear optics to the clamp blocks and mounting pillars.
Figure 1 - Mounting linear optics to clamp blocks and mounting pillars
The mounting pillars are mounted to the machine using magnetic clamp blocks or/and the steel base plates.
The linear interferometer and linear reflector should be mounted at the tool or probe holder and workpiece positions.
Measurements will then accurately reflect the errors that will occur between the workpiece and tool or probe, and are not contaminated by other errors. Even if machine guards and covers make access difficult, always try to fix both the interferometer and the retro-reflector rigidly to the machine. Also, try to arrange the laser and optics so that the interferometer is the stationary optic. This avoids errors that can occur if there is any beam deflection introduced by the moving interferometer.
An example of positioning the interferometer and reflector on a machining centre and CMM are shown in Figures 2 and 3 below.
Typical linear optical set-ups for different machine configurations are shown in the linear measurements configuration section
Page 6 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
Figure 2 - X axis linear positioning measurement on a vertical machining centre
Figure 3 - Linear optics mounted on a CMM
Note: Figure 3 shows the laser and interferometer mounted directly on the table, with the reflector mounted on the spindle with its clearance face downward (see Correct orientation of reflectors).
Correct positioning of the optics to minimise measurement errors
The following factors should be taken into account when mounting the optics to simplify set-up and to
Page 7 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
ensure any measurement errors are minimised.
Mounting of linear optics - Issue 5.1
© 2000-2001 Renishaw
Linear measurement configurations
The linear measurement optics can be set up in many different configurations. Which configuration is used will depend on the machine type and which axis is being calibrated.
Most linear measurement configurations fall into one of the following categories:
Horizontal axes -
moving interferometer or moving reflector
Horizontal axes -
moving interferometer or moving reflector with vertical reference arm
Horizontal axes -
at right angles to laser head -
moving reflector
Horizontal axes -
at right angles to laser head -
stationary turning mirror and interferometer -
moving reflector
Vertical axes -
moving reflector
Inclined axes -
stationary swivel mirror and interferometer -
moving reflector
Note: To keep the following pictures as simple as possible, the clamp blocks and mounting pillars are not shown. The best position to mount the clamp blocks to the linear optics will depend on the machine being calibrated and the availability of flat accessible surfaces on the machine.
Horizontal axes - moving interferometer or moving reflector
Minimise Abbé offset error
Fix optics directly to the points of interest
Fix the optics rigidly Bring optics together at one end of axis travel Avoid localised heat sources
Minimise material deadpath error Use turning mirrors Cleanliness of optics
Page 8 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
Figure 1
Horizontal axes - moving interferometer or moving reflector with vertical reference arm
Figure 2
Horizontal axes - at right angles to laser head - moving reflector
Page 9 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
Figure 3
Horizontal axes - at right angles to laser head - stationary turning mirror and interferometer - moving reflector
Note: Use this optical set up if you are also going to measure angular pitch and yaw or straightness on the same axis.
Figure 4
Page 10 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
Vertical axes - moving reflector
Figure 5
Inclined axes - stationary swivel mirror and interferometer - moving reflector
Figure 6
Linear measurement configurations - Issue 5.1
© 2000-2001 Renishaw
Page 11 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
Correct orientation of linear reflectors
The optical element inside the reflector housing is not orthogonally symmetrical. If you look into the optic, you'll see what appears to be a six-faceted element, as shown in Figure 1.
Figure 1 - Correct orientation of reflectors
For best results, each reflector should be mounted so that the laser beam hits the middle of one of these facets and not one of the 'lines' which join them.
The linear reflectors and beam splitter are marked with a red spot next to the optical apertures. These red spots have two purposes:
1. They help in defining the correct orientation of the reflectors - as explained above, the red spot is next to a facet and not an edge (or 'line').
2. They define a clearance face (see Dimensions and weights section). The distance from the optical centre of the optic to the side of the housing is 1 mm shorter on a side with a red dot than a side without a red dot. Therefore, for ease of set-up on CMMs, the interferometer can be mounted directly on a machine table and the reflector can be mounted on the spindle. The reflector can be aligned with the interferometer while leaving a 1 mm clearance between itself and the table, provided its red spot is facing downwards and those on the interferometer are not.
Correct orientation of linear reflectors - Issue 5.1
© 2000-2001 Renishaw
Linear beam alignment
To ensure that enough signal strength is obtained over a machine's full axis of travel and to minimise cosine measurement errors, the ML10's laser beam must be aligned so that it is parallel to the axis of
WARNINGS
1. TO AVOID EYE DAMAGE, DO NOT STARE INTO THE OUTPUT BEAM.
2. DO NOT LET THE BEAM PASS INTO YOUR EYES OR ANYONE ELSE'S, EITHER DIRECTLY OR BY REFLECTION FROM AN OPTICAL ELEMENT OR OTHER REFLECTIVE SURFACE.
Page 12 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
travel.
The alignment procedures discussed in this section assume that the optics are set up as shown in Figure 1, where the linear interferometer is the stationary optic and the retro-reflector is the moving optic.
Figure 1
These procedures can also be adapted to other optical configurations shown in Linear measurement set-ups.
Positioning of the linear interferometer and reflector
1. Position the tripod and laser so that it is pointing normally at the measurement optics. Roughly align the laser to the axis of travel using the housing as a sight line.
2. Rotate the laser shutter so that the laser is outputting a reduced diameter beam as shown in Figure 1. The smaller diameter beam makes it easier to see any misalignment.
Figure 1
3. Move the machine so that the linear reflector is close to the laser and fit a target on the front with the white spot at the top. Translate the laser or the machine until the beam hits the white spot on the target.
Note: The linear interferometer should not be positioned between the laser and the linear reflector at this stage.
The laser steerer can be used to simplify beam alignment. It reduces the amount of adjustment that needs to be made to the laser head and tripod.
Set up the ML10 on top of the tripod and stage.
Page 13 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
Figure 2
4. Remove the target and check that the beam returned from the reflector hits the centre of the target on the ML10 shutter. If it does not, then translate the laser or the machine until the beam hits the centre of the target.
Figure 3
5. Position the linear interferometer as close as possible to the reflector as shown in Figure 4. If they are positioned close together, the remainder of the alignment can then be achieved by only adjusting the laser head.
Figure 4
6. Ensure that the outside faces of the interferometer and reflector are square with the machine and aligned with one another. If the interferometer is skewed, then degradation in accuracy and possible failure to detect when the beam is obstructed may occur. As a general rule, it is advisable to align the interferometer to better than ±2° in roll, pitch and yaw, which can often be done by eye.
7. Fit a target to the input aperture of the interferometer with the white spot at the top and translate the interferometer vertically and horizontally so that the beam hits the target.
Page 14 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
Figure 5
8. Remove the target from the linear interferometer and check that the returned beam from the interferometer hits the centre of the ML10 shutter on top of the beam returned from the retro-reflector. If it does not, then translate the interferometer until the beam hits the centre of the white target. Note: You may find it useful to block the beam being returned from the reflector by positioning a card between the interferometer and the reflector.
Figure 6
Simple linear alignment procedure
If you are an experienced user of the system, you may wish to use the faster quick alignment procedure.
1. Drive the machine a short distance along the axis being calibrated so that you can fit the target on to the retro-reflector with the white target at the top.
Align the laser so that the laser beam strikes the centre of the target over the length of axis travel.
2. Remove the target from the linear reflector and if necessary horizontally translate the laser head so that the beams returned from the linear interferometer and the retro-reflector hit the horizontal centre line of the shutter's target.
Figure 1
3. Then vertically translate the laser head to bring the beam back onto the centre of the target. Note: At this point it may be necessary to make another small horizontal rotational adjustment to get the laser beam back on the centre of the target.
Page 15 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
Figure 2
4. Recheck the alignment of the returned beams at the laser head as described in the quick alignment procedure. At the laser head, the effect of any beam misalignment error is doubled, making misalignment easier to detect.
Quick alignment procedure
1. Drive the machine along the axis of travel so that the reflector moves away from the interferometer.
Figure 1
2. Move the machine until you see that the beam starts to move off the target on the ML10 shutter. As soon as you see that one of the beams has reached the edge of the target, stop moving the machine.
Figure 2
Vertical beam adjustment
Page 16 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
3. Use the thumbwheel at the rear of the laser head to adjust its rotational pitch to bring the two beams back to the same height.
Figure 3
4. Now use the tripod's height adjustment wheel on the tripod's central spindle to wind the laser head up or down to bring both beams into the centre of the target.
Figure 4
Horizontal beam adjustment
5. Using the small knob at the rear of the left hand side of the tripod stage, adjust the angular yaw of the laser head to bring the two beams on top of one another.
Figure 5
6. Now, using the larger knob midway along the left-hand side of the tripod stage, horizontally translate the laser to bring both beams into the centre of the target.
Figure 6
Page 17 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
7. Now resume driving the machine along the axis of travel. One again, stop the machine when you
can see that the laser beam has moved off the target. Repeat the laser alignment in steps 3 to 6 until you have reached the end of the axis.
8. When the end of the axis is reached, move the machine back so that the retro-reflector and the linear reflector are close together.
Note: If one of the beams has moved off the shutter's target, this is caused by the retro-reflector being laterally offset. Move the retro-reflector up/down and left/right until the beam returned from the retro-reflector overlaps with the beam from the interferometer on the target of the shutter.
9. Repeat steps 1 to 10 until the two beams remain in the centre of the target over the entire length of axis travel.
10. The laser beam is now aligned with the axis of travel. Rotate the shutter into its measurement position as shown in Figure 7 and check the signal strength displayed in the linear data capture software as the retro-reflector is moved over the machine's full length of travel.
Figure 7
11. Clear the BEAM OBSTRUCT error by datuming the laser. To datum the laser use the [Ctrl] + [D]
keys or use the button on the toolbar.
12. Block the laser beam between interferometer and reflector and confirm that the BEAM OBSTRUCT lamp is displayed in the calibration software. If a beam block does not occur, check that outside faces of the interferometer and reflector are square with the machine and aligned with one another as discussed in Positioning of the linear interferometer and reflector.
Note: If the machine s axis of travel is less than a metre long, further adjustments may be required to minimise the effect of cosine error. Refer to Cosine error for more information.
13. Next activate environmental compensation and ensure that the correct material expansion coefficient has been entered in the software. Then capture linear data.
Linear beam alignment - Issue 5.1
© 2000-2001 Renishaw
EC10 environmental compensation unit
Page 18 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
Figure 1 - EC10 environmental compensation unit
The EC10 environmental compensation unit compensates the wavelength of the laser beam for variations in air temperature, air pressure and relative humidity.
The EC10 can also accept inputs from up to three material sensors, which measure the temperature of the machine. Provided the appropriate material thermal expansion coefficient has been entered into the Laser10 software, this will allow measurements to be normalised to a machine (material) temperature of 20 °C (68 °F).
The EC10 is only required for linear measurements. In linear measurements, if an EC10 is not used, variations in the refractive index of air can lead to significant measurement errors. Using the EC10 environmental compensation unit, linear displacement measurements may be made to within the system accuracy specified for linear measurement.
Environmental compensation can be performed in three ways:
The pictures shown in this section are of the EC10 Gold Standard. The EC10 Gold Standard provides a better system accuracy than earlier EC10s.
A full EC10 specification is given in the specifications section.
Environmental sensors
Wavelength compensation
Material thermal expansion compensation
Automatic environmental compensation with EC10
Manual environmental compensation with EC10
Compensation using manually entered data
Page 19 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
The air temperature and material temperature sensors shown in Figure 2 below are separate items and are supplied complete with moulded leads terminated in connectors.
Figure 2 - Air and material temperature sensors
These connectors plug into sockets at the rear of the EC10 housing. Take care to connect the air temperature sensor to the correct socket; the sockets are clearly marked and are as shown in Figure 3 below:
Figure 3 - Rear view of EC10
Important note: The EC10's air and material temperature sensors should be connected before the EC10 is switched on. Once the EC10 has identified one of its sensor channels is not connected, it will not be interrogated again. When using fewer than three material temperature sensors, the sockets must be used in ascending order.
The air and material temperature sensors are all supplied with 5 m (16.5 ft) leads. Extension leads of lengths 5 m (16.5 ft), 10 m (33 ft) and 15 m (49 ft) are obtainable from your local Renishaw distributor. The leads are for use with either type of sensor and can be combined as required to give a total length of up to 20 m (65.6 ft) on each sensor.
The air pressure and relative humidity sensors are situated behind two ports in the front of the EC10 housing as shown in Figure 4 below:
Page 20 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
Figure 4 - Humidity and air pressure sensors
EC10 calibration
To maintain the Renishaw calibration system within its specified accuracy, we advise that the EC10 and its sensors are recalibrated annually. More frequent calibration is advised for units used in extreme environmental conditions, or where damage is suspected. The requirements of your quality assurance programme or national/local regulations may also dictate more frequent recalibration. Also, during storage, transportation and use, they should not be subjected to excessive shock, vibration or extremes of temperature, pressure or moisture, since any of these factors could invalidate their calibration. If in doubt, refer to your local Renishaw distributor.
The EC10 compensation unit is calibrated as follows:
The readings of the pressure sensor are compared to those from a UKAS-calibrated reference pressure indicator at each of five nominal pressures (in the range 750-1150 mbar) applied on both the rising and falling halves of a pressure cycle.
The readings from the relative humidity sensor are compared to those from a UKAS-calibrated reference relative humidity probe and meter at three applied humidities.
The error contribution of the two internal precision resistors used as references for temperature measurements are derived by connecting two precision wire-wound resistors (whose values are defined by comparison to a UKAS-calibrated standard resistor) to each of the temperature sensor input connectors in turn.
The air and material sensors can be calibrated separately from the EC10 environmental compensation unit. Therefore, sensors can be interchanged between EC10 units without invalidating their calibration. Separate certificates are issued for the EC10 and for the air and material sensors.
The air temperature sensor is calibrated by direct comparison with a UKAS-calibrated reference sensor in a lagged enclosure. The material temperature sensor is calibrated by direct comparison to two UKAS-calibrated semi-standard platinum resistance thermometers in an ice bath.
Page 21 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
The measurement instruments and reference standards used in this calibration have themselves been calibrated either directly by the National Physical Laboratory (NPL) or traceable to NPL through a UKAS-accredited calibration house.
The system accuracy makes an allowance for reasonably foreseeable drift over the recommended recalibration period but:
the EC10 unit and sensors cannot be guaranteed to remain within specification between calibrations
the value quoted makes no allowance for uncertainties associated with the set -up and alignment of the system
the value quoted makes no allowance for uncertainties associated with the inappropriate positioning of the sensor
The errors and uncertainties associated with normalisation of readings to a material temperature of 20 °C are not included; these will depend not only on the material temperature sensor being within specification (as evidenced by a recent Renishaw calibration certificate), but also on the value of expansion coefficient
entered into the calibration software and the temperature differential from 20 °C.
Renishaw offers a recalibration service for EC10 environmental compensation units and their sensors. For more information, refer to your local Renishaw distributor.
EC10 data link
At the rear of the EC10 housing are the mains power socket, mains (on/off) switch, a status lamp and a socket for the data link cable which connects the EC10 unit to the PC10 or PCM20 interface cards.
Power supply
The EC10 has been factory set for the supply voltage and frequency of your country. However, before using the EC10, ensure that the power supply is set to the appropriate voltage, especially if using the EC10 outside your country or on a site with a non-standard supply. To do this, check the setting of the switch on the underside of the unit, as shown in Figure 5 below:
EC10 environmental compensation unit specification
WARNINGS
1. THE MAINS LEAD (POWER CORD) IS A THREE CORE CABLE. THE EARTH (GROUND) CORE MUST BE EARTHED/GROUNDED EFFECTIVELY AT THE SUPPLY.
2. THE NEUTRAL LINE IS NOT FUSED. WHEN CONNECTING THE UNIT TO THE MAINS SUPPLY, TAKE CARE THAT THE CORRECT CABLE CORE IS CONNECTED TO THE LIVE. PARTICULAR CARE SHOULD BE TAKEN WHEN TERMINATING AN UNTERMINATED MAINS LEAD, OR WHEN ADDING AN EXTENSION TO AN EXISTING MAINS LEAD (POWER CORD).
Page 22 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
Figure 5 - Voltage selection switch
The voltage selection switch should be set as follows:
Change the voltage setting as follows:
1. Ensure the power is turned OFF.
2. Using a coin or a large flathead screwdriver or similar implement, turn the switch to the appropriate setting.
3. Switch the unit on.
The unit is now safe to use.
The live (line) input is protected by a Class T cartridge fuse located in a retractable tray immediately next to the mains socket at the rear of the EC10 housing. The tray also has space for a spare fuse. The active fuse is the one in the inner position.
EC10 environmental compensation unit - Issue 5.1
© 2000-2001 Renishaw
Supply voltage EC10 voltage setting
100 volts 120
120 volts
220 volts 240
240 volts
Page 23 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
Wavelength compensation
The accuracy of linear positional measurements depends on the accuracy to which the wavelength of the laser beam is known. This is determined not only by the quality of the laser stabilisation, but also by ambient environmental parameters. In particular, the values of air temperature, air pressure and relative humidity will affect the wavelength (in air) of the laser beam.
If the variation in wavelength is not compensated for, then linear laser measurement errors can reach 50 ppm. Even in a temperature-controlled room the variation in day-to-day atmospheric pressure can cause wavelength changes of over 20 ppm. As a guide, an error of approximately 1 ppm will be incurred for each of the following changes in the environmental conditions:
Air temperature 1 °C (1.8 °F) Air pressure 3.3 mbar (0.098 in Hg) Relative humidity 30%
Note: These values are worst case, and they are not entirely independent of the values of the other parameters.
These errors can be reduced by using an EC10 environmental compensation unit.
The EC10 measures the air temperature, pressure and humidity, then calculates the air's refractive index (and hence the laser wavelength) using the Edlen equation. The laser read-out is then automatically adjusted to compensate for any variations in the laser's wavelength. The advantage of an automatic system is that no user intervention is required and that compensation is updated frequently.
Wavelength compensation only applies to linear measurements. For other measurements (angle, flatness, straightness etc), environmental influences are far less significant, as environmental changes affect both the measurement and the reference beams to a similar degree.
Note: accuracy can be further improved (down to a theoretical limit of the ML10 long term frequency accuracy) by manual entry of environmental data read from very high precision instrumentation, provided that the environment is maintained within the entered parameters throughout the period of measurement. Refer to Compensation using manually entered data (no EC10).
Wavelength compensation - Issue 5.1
© 2000-2001 Renishaw
Positioning of air sensors
Positioning of air temperature sensor
The air temperature sensor supplied with the EC10 environmental compensation unit is mounted on a small cylindrical pedestal with a strongly magnetic base so that it can be 'clamped' to a machine or to a Renishaw mounting kit base plate or mounting pillar.
CAUTION
TO ENSURE THERMAL STABILISATION, THE AIR TEMPERATURE SENSOR SHOULD BE IN THE MEASUREMENT ENVIRONMENT FOR UP TO 15 MINUTES BEFORE STARTING MEASUREMENT.
Page 24 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
The air temperature sensor should be placed as close as possible to the laser beam's measurement path and about halfway along the axis of travel. Avoid placing the sensors close to localised heat sources, for example motors, or in cold draughts.
When measuring long axes, check for the presence of air temperature gradients. If the air temperature changes by more than 1 °C along the axis, use a fan to circulate the air. (This is particularly relevant on long vertical axes where air temperature gradients are more likely.) Avoid routing sensor signal leads close to sources of major electrical interference such as high power or linear motors.
Air pressure/relative humidity sensors
The pressure and humidity sensors are mounted within the EC10 environmental compensation unit. In general, it is not necessary to measure air pressure or relative humidity in the immediate vicinity of the beam path. This is because large variations in pressure and humidity are required to give a significant error in measurement and there should be no significant variation in either, across the work area. However, the relative humidity sensor should be positioned away from sources of heat or draught.
When calibrating vertical axes over 10 metres long, it is also recommended to place the pressure sensor halfway up the axis of travel.
Positioning of air sensors - Issue 5.1
© 2000-2001 Renishaw
Material thermal expansion compensation
The international reference temperature used by the calibration community is 20 °C (68 °F) and CMMs and machine tools are normally calibrated with reference to this temperature. In a normal factory environment where precise temperature control is often not available, the machine will not be at this temperature. Because most machines expand or contract with temperature, this could cause an error in the calibration.
To avoid this calibration error, the linear measurement software incorporates a mathematical correction called thermal expansion compensation or 'normalisation' which is applied to the linear laser readings. The software normalises measurements using the coefficient of expansion, which must be entered manually, and a mean machine temperature measured using the EC10. The objective of this correction is to estimate the laser calibration results that would have been obtained if the machine calibration had been performed at 20 °C (68 °F).
Material thermal expansion coefficients
The amount that most materials expand or contract with changing temperature is very small. For this reason, the thermal expansion coefficient is specified in parts per million per degree C or F (ppm/°C or ppm/°F). These coefficients specify the amount that the material will expand or contract for every degree rise or fall in material temperature. For example, suppose the coefficient of thermal expansion is +11 ppm/°C. This means that for every 1 °C rise in material temperature, there will be a material expansion of 11 ppm, which is equivalent to 11 micrometres per metre of material or 11 microinches (.000011") per inch of material.
CAUTION
DO NOT PUT THE RENISHAW AIR TEMPERATURE SENSOR NEAR COMPUTER DISKS OR TAPES OR ANYTHING WHICH WOULD BE DAMAGED BY ITS MAGNETIC FIELD.
Page 25 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
Incorrect compensation for material thermal expansion is one of the primary sources of error in laser linear distance measurements in non-temperature controlled environments. This is because the expansion coefficients of common engineering materials are relatively large compared to the coefficients associated with wavelength compensation errors and laser beam alignment errors.
The normalised measurement will have an error relating to the measurement accuracy of the material temperature sensor. The size of this error depends on the thermal expansion coefficient of the machine under test. The material temperature sensor has an accuracy of ±0.1 °C and therefore if the machine under test has a thermal expansion coefficient of 10 ppm/°C, then the error in the normalisation of the measurement would be ±1 ppm. This is in addition to the system measurement accuracy (0.7 ppm) when using the EC10 environmental compensation unit.
However, since the two errors are uncorrelated, their combined effect is the square root of the sum of their squares and not their arithmetic sum. Thus, for the above example, the normalised measurement accuracy will be 1.2 ppm for the ML10 and EC10 Gold Standard systems (1.5 ppm for earlier ML10 and EC10 units).
Additional measurement errors will occur if an incorrect thermal expansion coefficient is entered into the software. Since the values of the thermal expansion coefficients of different machines can vary by 10 ppm/°C (5.5 ppm/°F) or more, care should be taken to ensure that the correct values are entered. If necessary, seek the advice of the machine's manufacturer.
The expansion coefficient of the machine's feedback system is normally entered into the software, unless you are estimating the accuracy of machined parts when returned to 20 °C (68 °F). Table 1 below gives typical expansion coefficients for different materials used in construction of machines and their position feedback systems.
Note: Since material expansion coefficients can vary with material composition and treatment, these values are for guidance only and should only be used in the absence of manufacturer's data.
Table 1
When trying to identify the expansion coefficient, be particularly careful where there are two materials with different coefficients fixed together. For example, in the case of a rack and pinion feedback system, the expansion coefficient may be closer to the cast iron rail to which the rack is fixed. In the case of large gantry machines with floor mounted rails, the expansion coefficient of the rail may be reduced by the restraining action of the concrete foundations. Also, many modern scales are composed of a number of different materials, e.g. a glass scale may be bonded to an aluminium spar which is mounted, in turn, on a cast-iron machine member. In such cases, selection of the appropriate coefficient can be difficult. You should seek the advice of the manufacturer of the scale and/or the machine on which it is used.
Material Application Expansion coefficient
ppm/°F ppm/°C
Iron/steel Machine structural elements, rack and pinion drives, ballscrews
6.5 11.7
Aluminium alloy Lightweight CMM machine structures 12 22
Glass Glass scale linear encoders 4.5 8
Granite Machine structures and tables 4.5 8
Concrete Machine foundations 6 11
Invar Low expansion encoders/structures <1 <2
Zerodur glass "Zero" expansion encoders/structures <0.1 <0.2
Page 26 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
Material thermal expansion compensation - Issue 5.1
© 2000-2001 Renishaw
Positioning of material sensor
When positioning the material temperature sensors, the first step is to decide on your primary objective for performing material expansion compensation. This is usually one of four possible objectives.
The differences between these objectives are often significant, particularly if the machine position feedback system gets hot during machine operation (for example a ballscrew), or if the workpiece expansion coefficient is significantly different from that of the position feedback system, for example, an aluminium workpiece with glass scale linear encoders.
The material temperature sensor supplied with the EC10 has a strong magnetic base for 'clamping' to the machine under test. Ensure there is good thermal contact between the material temperature sensor and the material being measured.
Calibration in accordance with National and International Standards
To calibrate the accuracy of the machine in accordance with a National or International Standard, the procedure defined in the standard should be followed. This should cover where to place the material sensor, what expansion coefficient to use, and what machine warm up cycle to perform. If a thermal drift test is also defined in the standard, this must also be included.
Estimate accuracy of the machine if it was operated in 20 °C environment
CAUTION
TO ENSURE THERMAL STABILISATION, THE MATERIAL TEMPERATURE SENSOR SHOULD BE FIXED TO THE MATERIAL FOR 25 MINUTES BEFORE STARTING MEASUREMENT.
To estimate the linear positioning accuracy that would be obtained if the machine was operated in an ambient environment of 20 °C (68 °F). This is often the objective during machine build, sign off, commissioning or recalibration, and in most cases is the same as defined in a National or International Machine Acceptance Standard. To perform a calibration in accordance with a National or International Machine Acceptance Standard. To estimate the linear accuracy that the machine feedback system could achieve if the feedback system was at a temperature of 20 °C (68 °F). This is useful for diagnosing faults in the feedback system. To estimate the accuracy of parts that the machine will produce when those parts are returned to 20 °C (68 °F) for inspection. This objective is particularly important in the production of accurate non-ferrous parts in non-temperature controlled shops, where machine feedback and workpiece expansion coefficients differ significantly.
CAUTION
DO NOT PUT THE RENISHAW MATERIAL TEMPERATURE SENSOR NEAR COMPUTER DISKS OR TAPES OR ANYTHING WHICH COULD BE DAMAGED BY ITS MAGNETIC FIELD.
Page 27 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
To estimate the accuracy of the machine if it was operated in an environment of 20 °C (68 °F), the material temperature sensor(s) should be placed on the table of the machine or on some other massive part of the machine structure that is NOT close to any sources of heat such as motors, gearboxes, bearing housings, exhausts etc. The material expansion coefficient should be set to that of the feedback system.
If the air and machine temperatures are significantly different, then it is also likely that there are significant temperature differences between material surface and core temperatures. Under these circumstances, care should be taken to locate the material temperature sensors where they will measure the core temperature. The temperature can be measured at a number of points using up to three material sensors and the compensation factor applied will be based on an average value.
It is a common misconception that material sensors should always be placed on the ballscrew, or feedback system. This is not always the case, as the following example illustrates.
Example
Suppose a machine is being calibrated in a shop at 25 °C, and because of heat generated by machine operation, the ballscrew is 5 °C warmer, at 30 °C. If the material sensors are placed on (or very close to) the ballscrew, the laser readings will be compensated to estimate the readings that would have been obtained if the ballscrew was operating at 20 °C. However, if the machine were being operated in an environment at 20 °C, the ballscrew would NOT be at 20 °C. The heat generated by operation of the screw and the motor would still be there, so the ballscrew temperature would still be about 5 °C warmer than ambient (25 °C). Putting the material sensor(s) on the ballscrew will therefore result in overcompensation. It is better to place the sensor(s) on a massive part of the machine to give a temperature reading related to the average ambient temperature around the machine over the last few hours.
Estimate accuracy of machine feedback system if it was at 20 °C
This procedure is often used for diagnostic purposes. Perhaps the machine has failed calibration against Objective 1 or 2, and the accuracy of the feedback system at 20 °C now needs verifying. To meet this objective, the laser beam should be aligned as close to the axis of the feedback system as possible (to minimise Abbé offset error). The material temperature sensor(s) should be placed on (or very near to) the feedback system and the expansion coefficient should be set to that of the feedback system. The temperature can be measured at a number of points using up to three material sensors.
Manufacture of parts which must be accurate at 20 °C
If a machine tool is always used to machine workpiece materials with a significantly different expansion coefficient to those of the feedback system, for example, aluminum alloys, carbon composites, ceramics, etc., it may be beneficial to use the expansion coefficient of the workpiece and not the one of the machine feedback system. Although this will not give a calibration that represents the performance of the machine at 20 °C, it can improve the accuracy of the workpieces when they are returned to 20 °C for measurement.
The material temperature sensor(s) should be located to measure a temperature similar to that expected of the workpiece. This is often on the table of the machine, but other factors like the type of coolant system employed and the metal removal rates may need to be considered. Care should also be taken to perform this type of calibration under typical conditions, and it can only be truly effective if the temperature and expansion coefficients of the various workpieces are relatively consistent.
Positioning of material sensors - Issue 5.1
© 2000-2001 Renishaw
Page 28 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
Automatic environmental compensation with EC10
Automatic environmental compensation uses the EC10 environmental compensation unit to perform laser wavelength compensation and material thermal expansion compensation. If calibration is being performed in an environment where the atmospheric conditions are likely to vary during the test, then automatic compensation is strongly recommended.
To perform automatic compensation connect the EC10 environmental compensation unit to the PC10/PCM20 interface card using the data link cable provided, and connect the air and material temperature sensors to the appropriate sockets in the rear of the EC10. Refer to environmental sensors for more information.
Switch on the EC10 unit after all sensors are plugged in. From the data capture window, select the Capture/Environmental Set-up menu option to display the Environmental Set-up dialog box and ensure that the Automatic compensation checkbox is ticked. After a short time the EC10 AUTOMATIC lamp in the data capture status window will indicate that the EC10 is available. Environmental compensation is now performed automatically. EC10 readings are updated every two minutes, and are used to compensate the laser readings accordingly. Refer to EC10 update cycle for more information.
Alternatively, automatic compensation can be enabled in the environmental window that is normally displayed on the right hand side of the screen. Again ensure that the Automatic compensation checkbox is ticked. If this area is being used for another function, i.e. showing the optical configuration for linear measurement or displaying the current session properties, select the Window/Environment menu option to display the environmental window. The environmental window is useful for monitoring the current environment as measured by the sensors.
Where only one or two material temperature sensors are used, the lines corresponding to those not in use will read '--NC--' (not connected).
The units of any environmental parameter (except relative humidity) can be changed by clicking on the arrow to the right of the units box.
To define the default units which are used when the calibration software is started refer to configuration.
CAUTION
Page 29 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
EC10 update cycle
The PC10 and PCM20 PC interface cards interrogate the EC10 for environmental sensor readings on an interleaved two-minute cycle. After the first minute, air pressure, air humidity, air temperature and the environmental factor are updated. After the second minute, the three material sensors and the environmental factor are updated. This cycle is repeated every two minutes. In addition to this interleaved interrogation cycle, the software updates the display reading every minute. Therefore, in the worse case it may take three minutes for the EC10 readings to be correctly displayed on the computer screen after switching the system on.
Important note: The EC10's air and material temperature sensors should be connected before the EC10 is switched on. Once the EC10 has identified one of its sensor channels as being not connected or faulty, it will not be interrogated again. Therefore swapping sensors between EC10 sensor ports once the system has been switched on can lead to false N/C and FAULTY messages being displayed for the air and material temperature sensors.
Automatic environmental compensation with EC10 - Issue 5.1
© 2000-2001 Renishaw
Manual environmental compensation with EC10
With the EC10 still connected to the laser system, the data capture software can be set to manual mode. In manual mode, the EC10 will continue to interrogate the sensors and update the values displayed on the screen. However, the 'environmental factor' is updated only on manual command, by pressing [Ctrl]+[L]. This allows a measurement run (or series of runs) to be made using constant compensation parameters.
Manual mode can be used for the calibration of an axis where the effect of continuing thermal expansion of the machine during the calibration is not to be normalised. It can be argued that any thermal effect which occurs during calibration will also occur during normal operation and should not be excluded from the statistical analysis of positional accuracy and repeatability.
If calibration is being performed in an environment where the atmospheric conditions are likely to vary during the test, then automatic compensation is strongly recommended. If calibration can be performed quickly, or is being performed in a temperature-controlled room, then manual compensation may be acceptable.
On entering the linear measurement software, the currently active environmental set-up is normally displayed on the right hand side of the screen. To set the software to manual mode, click on the Automatic compensation checkbox and make sure the tick is cleared. The environmental compensation lamp in the data capture status window will show EC10 MANUAL.
BEFORE STARTING ANY CALIBRATION RUN:
MAKE SURE THAT THE MACHINE TO BE CALIBRATED HAS BEEN EXERCISED SUFFICIENTLY TO WARM UP THE DRIVE AND SCALE OF THE AXIS TO BE CALIBRATED.
MAKE SURE THAT THE CORRECT VALUE HAS BEEN ENTERED FOR THE COEFFICIENT OF THERMAL EXPANSION BY ADJUSTING Exp.Coeff PARAMETER.
Page 30 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
Manual environmental compensation with EC10 - Issue 5.1
© 2000-2001 Renishaw
Compensation using manually entered data (no EC10)
In manual compensation, the user reads the air temperature, pressure and humidity from separate instruments and then manually enters the values into the laser calibration software. The linear data capture software then calculates the environmental factor and this value is applied to the linear distance reading.
Because the system is manual, it is usually impractical to update the compensation frequently. If calibration can be performed quickly, or is being performed in a temperature-controlled room, then manual compensation may be acceptable.
Entering environmental parameters manually
Do not connect the EC10 to the PC10 or PCM20. Run the linear data capture software and ensure that the EC10 status lamp in the data capture status window is empty, i.e. it does not show the EC10 MANUAL or EC10 AUTOMATIC status messages.
From the data capture window, select the Capture/Environmental menu option to display the Environmental Set-up dialog box.
Ensure that the Exp. Coeff parameter is set to the expansion coefficient of the machine or object under test.
Using the appropriate measuring instruments:
1. Take readings from the air temperature, air pressure and relative humidity sensors and enter the values in the appropriate boxes.
2. Take reading(s) from the material temperature sensor(s) and enter the (mean) value in the appropriate boxes.
3. The Environment factor, which is used to compensate and normalise the raw measurement, is displayed at the bottom of the screen. This value will be updated each time the value of any of the parameters within the sub-menu is changed.
4. When you have finished, click OK.
To change an environmental value during data capture, stop the calibration at a suitable point. Click on the Finish button at the bottom of the data capture window. Modify the environmental values as described above and then select Continue from the Capture menu to continue capturing data.
Of the ambient environmental parameters, the air temperature is the one most likely to change during a calibration run. You should check this parameter regularly during the course of a calibration and update it as required. It is also advisable to carry out the calibration of each axis as quickly as possible to reduce the risk of a significant change in the air temperature.
In general, variations in air pressure and relative humidity are unlikely to be significant during the course of
Page 31 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
a calibration. However, this will not always be true. For example, where measurements are made in an environmentally controlled room, where the relative humidity inside the room is kept low but the relative humidity outside is very high, the opening of a door or window could lead to a significant localised change in relative humidity. In this case, it would be advisable to delay the start or continuation of a calibration until the relative humidity has re-stabilised.
The temperature of the axis of a machine tool is likely to rise during the course of a calibration run, owing to the heat generated in driving the machine. The extent of this problem will depend on the type, design and size of the machine and how much it was exercised before starting the calibration measurements. It is recommended that the material temperature readings are checked and recorded during the course of the calibration, so that any analysed data can be interpreted in the light of any change which occurs.
Selection of manual sensors
If compensation is performed manually, it is important to select environmental sensors with appropriate measuring accuracies. To achieve the system's measurement accuracy, the instrumentation used must comply with the minimum specification given in the table below:
Note 1: These figures are based on a collective effect of all parameters and also incorporate other system errors. The accuracy value when material compensation is enabled assumes that the appropriate coefficient of thermal expansion is entered, correct to sufficient decimal places in ppm.
Note 2: The air and material sensors will normally be different. An air temperature sensor requires a low thermal mass so that it can respond to rapid changes in air temperature. Material temperature sensors require an efficient thermal bonding so that they can reach the true scale temperature as quickly as possible.
Note 3:The atmospheric pressure value needed for compensation is not the sea level pressure quoted by meteorologists, but is the actual pressure at the current altitude. If pressures are taken from a normal weather barometer or local weather reports, they must be corrected to take into account the height above sea level. (Atmospheric pressure falls by approximately 0.115 mBar/metre, from 0-1000 metres).
Note 4: It takes a significant change in relative humidity (30%) to give an additional 1 ppm error in measurement. If measurement errors of 2-3 ppm are acceptable, it may be sufficient to enter an estimated value of relative humidity.
Compensation using manually entered data (no EC10) - Issue 5.1
Minimum specification for environmental sensors
Sensor Accuracy Resolution Range
Metric units
Air temperature Material temperature Air pressure Relative humidity
±0.20 °C
±0.1 °C ±1.0 mbar
±15%
0.03 °C 0.03 °C 0.3 mbar 5%
0-40 °C 0-40 °C 750-1150 mbar 0-95%
Imperial units
Air temperature Material temperature Air pressure Relative humidity
±0.36 °F
±0.18 °F ±0.03 in Hg
±15%
0.05 °F 0.05 °F 0.01 in (Hg) 5%
32-104 °F 32-104 °F 22-34 in (Hg) 0-95%
Page 32 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
© 2000-2001 Renishaw
Data capture
Data capture is carried out by moving the machine to a number of different positions (or 'targets') along the axis under test and measuring the machine s error. You can write a part program to drive the machine from one target position to the next, pausing for a few seconds at each target position. Measurements are taken during each pause. The steps required to perform data capture are as follows:
1. The machine's temperature will often rise during operation. So that this effect is included in the calibration, it is recommended you perform a warm-up sequence of moves.
2. Define the units (English or Metric) and resolution of the laser measurement display and error value readings.
3. Set the direction sense of the laser system to be the same as the machine under test using button on the toolbar or the [Ctrl] + [-] keys. For linear measurements, the sign of the laser display should correspond to the sign of the machine's axis. For angular and straightness measurements, the laser display should match the sign convention defined for the test.
4. For linear measurements, move the linear interferometer and linear reflector close together, then
datum the measurement display using the button on the toolbar or the [Ctrl] + [D] keys. Datuming with the optics close together minimises deadpath error.
5. Move the machine to the calibration start position. To ensure that any backlash in the machine's axis is removed, approach the start position in the same direction as the first run.
6. For linear measurements, if the machine read-out does not agree with the laser read-out, the
preset function may now be used to adjust the laser reading accordingly using the button on the toolbar or the [Ctrl] + [P] keys. For angular and straightness measurements, datum the
measurement display to zero using the button on the toolbar or the [Ctrl] + [D] keys.
7. Set up the target positions and the data capture sequence. If a calibration is being carried out in accordance with an approved national or international standard then set up the number of target points and run sequence as defined in the standard.
The software has an automatic set-up option which guides the user through the normal data capture set-up process by displaying the automatic target generation, capture initialisation and automatic data capture dialog boxes in sequence before starting data capture. To use this option, select
File/New/Automatic Set-up from the menu bar or click on the button on the toolbar.
CAUTION
THE RENISHAW SOFTWARE ASSUMES THAT ODD-NUMBERED RUNS ARE POSITIVE APPROACHES. TO COMPLY WITH THE STANDARD CONVENTION THAT POSITIVE
APPROACHES ARE FOR POSITIVELY INCREASING TARGET POSITIONS, PARTICULARLY ON AN AXIS WHERE THE SCALE IS NEGATIVE, ENSURE THAT THE FIRST TARGET POSITION IS MORE
NEGATIVE (OR LESS POSITIVE) THAN THE LAST TARGET POSITION (E.G. FIRST TARGET POSITION = -560; LAST TARGET POSITION = 0).
Select the measurement target positions
Select the measurement target sequence and the number of runs
Set up automatic data capture settings if required
Page 33 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
Data capture - Issue 5.1
© 2000-2001 Renishaw
Analysis
Running analysis
To run the Renishaw Laser10 analysis software, select Renishaw Laser10 from the Windows Start menu, then select the Analysis option. The Renishaw Laser10 analysis window will appear.
If new data has been captured then, from the Renishaw Laser10 capture software, select Data/Analyse
from the menu bar or click on the button on the toolbar.
Note: Before analysing newly captured data, it should first be saved.
The Renishaw Laser10 analysis window will open with the analysis plot of the new data displayed.
Selecting the data to be analysed
To analyse a data file, click the button on the toolbar then use the Open dialog box to load the required data file.
Note: There are specific sections describing how to analyse data captured using the following measurement modes.
Selecting the analysis plot type
Clicking Analysis in the menu bar displays a list of the available analysis plot types, the current selection being indicated by a tick alongside the option. Select a plot type according to the national or international standard you need to use (e.g. ISO 230-2, NMTBA, VDI 2617 etc.) or select one of the Renishaw options as appropriate. The list of plot types will change depending upon the type of data file loaded, i.e. linear, angular, straightness, dynamic or flatness.
If the selected plot type does not support the data you are analysing, you will see a warning, for example:
Measure and record machine errors Save the captured data to disk
Analyse the captured data
Flatness analysis
Straightness analysis
Squareness analysis
Parallelism analysis
Page 34 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
To set the default plot type, press the [Ctrl] key at the same time as you select the plot type i.e. press [Ctrl] when you click on a plot type with the mouse, or highlight the plot type with the cursor keys and press [Ctrl] [Return]. The default plot type will be in bold.
Displaying an analysis plot
To display an analysis plot of the type selected in the Analysis menu, click the button on the toolbar. You can also display an analysis plot by selecting Plot/Plot from the menu bar or by pressing the [Ctrl]+[P] keys.
Note: Selecting an option in the Analysis menu will automatically display the associated analysis plot.
The analysis window
A typical analysis display is shown below.
The analysis window consists of:
Title bar Toolbar
Menu
Analysis plot
Exiting analysis
Page 35 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
To exit analysis, press [Ctrl]+[X] or select File/Exit from the menu bar or click the button on the toolbar.
Working with analysis plot types
When the button is depressed on the toolbar, you can display more than one plot type in the window, provided that the data being analysed is supported by the selected option.
If you have several analysis plots displayed in the window, you can use the Windows menu to arrange them as 'tiled' (side by side) or 'cascaded' (overlapping), the default being 'cascade'. If you minimise an analysis plot window by clicking the ' - ' box in the top right hand corner, you can use the Windows/Arrange Icons menu option to organise the icons in a neat row along the bottom of the main analysis window.
Trend analysis plot - When this option is selected, a dialog box is displayed which allows you to define the file names for up to four separate plots on the same graph.
Selected runs plot - When using this option, a dialog box pops up enabling you to choose which data capture run(s) you want displayed on the plot. Simply highlight the runs you want to display by selecting the runs in the list box.
Page 36 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
VDI 2617 template - When using this plot type, the VDI 2617 Template Set-up dialog box pops up enabling you to edit the template information.
Using cursors
When you select the Cursors option from the Tools menu, a dialog box pops up and the mouse pointer becomes an active cursor. If the mouse pointer is now positioned in the currently selected plot, a line 'snaps' to the nearest data point and the X and Y coordinates, target number (T) and run number (R) for that data point are displayed in the dialog box.
Setting the measurement units for a plot
Select Plot/Units from the menu bar or press the [Ctrl] + [U] keys, and the Units Selection dialog box will be displayed. Select the required units for the X and Y axes then click the OK button. Select the Set as default button to make the selected units the default.
Setting the scale values for a plot
Printing
Using a different printer or plotter
Viewing analysis plots
Customising the toolbar
Analysis - Issue 5.1
© 2000-2001 Renishaw
Factors affecting accuracy of linear measurements
A common error seen on linear calibration graphs is a steady gradient or slope. The steady gradient can be caused by an error in the machine's positional feedback system. However, this can also be caused by the following measurement errors.
See plot scaling for more information
See printers and plotters for more information
See printers and plotters for more information
You can change the way in which the plot appears on your screen by using the View menu options.
See customising toolbar section for more information
Incorrect wavelength compensation
Page 37 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
Other factors that can affect linear measurement accuracy are:
Factors affecting accuracy of linear measurements - Issue 5.1
© 2000-2001 Renishaw
Deadpath error
Deadpath error is an error associated with changes in the environmental factor during a linear measurement where EC10 automatic compensation is being used. Under normal conditions, deadpath error is insignificant and only occurs if the environment changes after a datum and during measurement.
The deadpath error of the laser measurement of path L2 is related to the distance between the two optical elements when the system is datumed L1, see Figure 1. If there is no motion between the interferometer and the reflector, and the environmental conditions surrounding the laser beam change, the wavelength (in air) will change over the entire path (LI + L2) but the laser measurement system will be compensated over the distance L2 only. Therefore a deadpath measurement error will be introduced due to beam path L1 not being compensated.
Incorrect material thermal expansion compensation
Incorrect positioning of air temperature sensors
Material deadpath error
Incorrect positioning of material sensors
Deadpath errors
Cosine error
Abbé offset error
Localised heat sources
Optics not positioned correctly
Optics not fixed rigidly
Page 38 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
Figure 1 - Deadpath error
However, the deadpath error will be negligible if the stationary and moving optic are abutted when the datum is set, as shown in Figure 2 below:
Figure 2 - Correct set-up for negligible deadpath error
When possible, datum the laser with the optics close together. If the optics are within 10 mm of one another when the laser is datumed, then deadpath error will be insignificant under normal conditions. Where the geometry of a machine means that the two optics are furthest apart when the moving optic is at the zero position of the axis, the preset facility can be used to avoid the potential deadpath error associated with datuming the laser interferometer system at this point.
Deadpath error - Issue 5.1
© 2000-2001 Renishaw
Cosine error
Any misalignment of the laser beam path relative to the axis of motion will result in an discrepancy between the measured distance and the actual distance travelled as shown in Figure 1.
Figure 1
- Cosine error
Page 39 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
This misalignment error is usually referred to as cosine error. The size of this error is related to the angle of misalignment between the laser beam and the axis of travel, shown as in Figure 1.
When the laser measurement system is misaligned to the axis of travel, cosine error will cause the measured distance to be shorter than the actual distance. The error increases significantly as the angular misalignment increases, as seen from the following table:
To minimize cosine error, the measurement laser beam must be aligned so that it is parallel to the axis of travel. On axes longer than a metre, this is relatively easy to achieve using the alignment procedures
provided. With shorter axes, it becomes increasingly difficult and the following techniques can be used to optimise alignment and minimise cosine error:
Maximise the laser reading
The auto reflection technique
Slope removal during the set-up for a straightness measurement
Note: Do not assume, because the signal strength is constant all along the axis of travel, that alignment is perfect. The signal strength meter in the calibration software has insufficient sensitivity and resolution to ensure accurate alignment on short axes.
Maximise the laser measurement reading
If there is a cosine error in the laser measurement, the laser reading will be smaller than it should be. Therefore, on short axes it is possible to eliminate cosine error by carefully adjusting the pitch and yaw of the laser head until the largest laser reading is obtained. The procedure is as follows:
1. Align the beam along the axis of travel.
2. Move the axis so that the optics are close together and datum the laser reading.
3. Move the axis so that the optics are at their greatest separation.
4. Carefully adjust the laser head's pitch and yaw controls to give the largest laser measurement.
Note: This is a delicate but highly effective procedure. It may be necessary to make a series of small adjustments and to let go of the adjustment controls after each one, before observing the effect on the laser read-out. It may also be necessary to translate the laser head to maintain alignment. It may also be necessary to select the maximum resolution setting on the measurement display and to set short term averaging ON. Once completed, it is a good idea to repeat the above steps to confirm alignment.
The auto-reflection technique
If the machine axis is very short and there are flat surfaces known to be suitably perpendicular or parallel
Angle
(mm/metre)
Angle
(arc minutes)
Error (ppm)
0.45 1.00 1.40 3.20 4.50
10.00
1.53 3.43 4.87
10.87 15.39 35.39
0.1 0.5 1.0 5.0
10.0 50.0
Page 40 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
(within 0.05°) to the axis of travel, then the auto-reflection technique can be useful. The procedure is as follows:
1. Align the beam along the axis of travel.
2. Place a steel gauge block in the path of the laser beam (after the interferometer) and against one or more of the flat surfaces.
3. Adjust the yaw and pitch controls so that the reflected beam from the gauge block surface is returned into the output beam aperture on the laser head. When this occurs, the laser beam is aligned with the axis of travel.
This technique works particularly well if the laser head is set some distance away from the interferometer.
Straightness measurement slope removal
If you intend to carry out both linear and straightness measurements on an axis, it is advisable to perform the straightness measurement first, as manual removal of slope can be used to optimise beam alignment. Provided that the straightness optics can be replaced by the linear optics without disturbing the mounting arrangement or the laser, the optimum alignment will be immediately available for the linear measurement.
Cosine error - Issue 5.1
© 2000-2001 Renishaw
Abbé offset error
Where a measurement is made with the beam aligned parallel to, but offset from, the defined axis of calibration, machine angular errors (e.g. pitch or yaw) can introduce an Abbé offset measurement error (see Figure 1).
Page 41 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm
Figure 1 - Abbé offset error
To minimise the effect of Abbé offset error, the laser measurement beam should be coincident (or as close as possible) to the line along which calibration is required. For example, to calibrate the linear positioning accuracy of the Z axis of a lathe, the laser measurement beam should aligned close to the spindle centre line. This will minimise the contamination of the linear measurement from machine pitch or yaw angular errors.
For each arc second of angular motion, the error introduced is approximately 0.005 µm/mm of offset. The following table gives examples of the errors in microns due to increasing pitch or yaw deviations:
* Example shown in Figure 1
Where the angle is constant, this does not lead to an Abbé error, since the moving optic will lead (or lag) the 'pivot point' by a constant distance, and the system operation is based on differential measurement (i.e. from the datum point).
The above relates to the Abbé offset error associated with the linear interferometer optics. The machine under test may itself be subject to an Abbé error. In this case, the position of the measurement axis with respect to the axis of the machine scale would affect the measurement.
Abbé offset error - Issue 5.1
© 2000-2001 Renishaw
Material deadpath error
For linear measurements, material expansion compensation is normally only applied to a material path length equal to the measured laser distance. If the measurement loop includes additional structures, then any thermal expansion or contraction of this 'material deadpath', or deflection under load, will cause measurement errors. Mounting the optics directly to the points between which measurement is required will minimise these errors.
Material deadpath error - Issue 5.1
© 2000-2001 Renishaw
Angle (arc-secs)
Offset distances (mm)
5 10 50 100 500
1
2
5
10
20
60
120
0.024 0.048 0.120 0.240 0.480 1.450 2.900
0.048
0.097
0.240
0.480
0.970
2.900
5.800
0.24
0.48
1.20
2.40
4.80
14.50
29.00
0.48
0.97
* 2.40
4.80
9.70
29.00
58.00
2.40
4.80
12.00
24.00
48.00
145.00
290.00
Page 42 of 42Linear measurement title page
20/05/2009file://C:\Documents and Settings\NospheratuDaemon\Local Settings\Temp\~hh893F.htm