Volumetric Accuracy Check After Crash: Laser Interferometer and Ballbar

Volumetric Accuracy Check After Crash: Laser Interferometer and Ballbar

Post-crash volumetric check with laser interferometer positioning tests and ballbar circularity and backlash runs, including point layout and pass limits.

Introduction

After the collision, the reliable evidence for determining whether the spatial accuracy is damaged or not, and how much it is damaged, is the data measured by instruments other than the machine. The reason is straightforward: the position feedback of Equator comes from the grating installed on the driving strut, and the driving point and feedback point are of the same origin. The machine will believe as much as the grating reports. The hinge point gap and the slight deformation of the strut, which truly affect spatial accuracy, are all outside the closed loop of the grating.

So we need to ask in two steps: whether the laser interferometer can accurately measure the single axis movement, and whether the direction change will result in loss of measurement; Is the circle produced by the three-axis linkage of the golf club still round or not. The testing sequence includes geometric rough inspection, single axis positioning accuracy, three-axis linkage circle, probe and calibration review. Each step has a passing criterion and a non passing direction. The qualified limit value is based on the standard and contract acceptance indicators, and no publicly available indicators of any model are used. The core of interpretation is to find correlation: errors are related to position, direction, or speed, and point to completely different faults.

1. Original text: Why spatial accuracy must be determined by external instruments

1.1 What the collision really moved was the coordinates of a point in space

To choose the right instrument, first think carefully about what the collision has changed. What it changes is not a directly readable number, but a set of geometric relationships: whether the effective length of the three driving struts corresponds to the nominal length, whether there are gaps in the Hooke hinges at both ends, whether the platform posture deviates from the design relationship during motion, and whether there is displacement of the probe assembly relative to the platform. None of these quantities can be read from the screen, only manifested as a deviation in where the same instruction position actually falls in space.

The essence of spatial accuracy is this mapping: the controller calculates the length of each of the three struts according to kinematics, and after the struts have traveled, the platform reaches a certain spatial point with the center of the stylus. If a collision causes a deviation in this mapping, the deviation form is not one-dimensional addition or subtraction, but a field that varies with position, direction, and velocity. So the investigation cannot start with ‘see which axis is alarming’ – an alarm only indicates that the system has detected an abnormality, not alarming does not mean that the geometric relationship is still present.

1.2 Machines cannot prove themselves: feedback loop closes errors outside the door

One sentence to clarify this limitation: The error of the measurement system can only be discovered by something more independent of the benchmark, and cannot be discovered by oneself.

The position feedback of Equator comes from the grating on the driving strut, and the length of the strut read by the controller is the input for kinematic solution. This closed loop has a characteristic: any error that occurs after the grating reading and before the real space point is not in the feedback path. Those belonging to this category include: hinge point clearance (the grating shows that the strut has traveled 10.000 mm, but the actual effective length has only changed by 9.996 mm), trace elastic deformation under the load of the strut, offset between the platform and the probe component, and return loss that cannot be read by the grating.

The most common misjudgment after a collision comes from here: if it returns to zero successfully, walks a few points, and the screen is normal, it is assumed that the geometry is not damaged. A normal screen is inevitable because it will always be normal. External instruments can break through because the feedback chain between the reference and the machine is completely independent – interferometers use laser wavelength as the reference, and ball club instruments use calibrated rod length as the reference, without consulting with gratings.

1.3 Division of labor between two instruments: one for measuring straight lines and one for measuring linkage

Space accuracy contains two types of information with different properties, requiring two types of instruments.

The first type is the linear motion property of single axis: the positioning error at each position after completing the entire journey in a straight line, how many times it travels back and forth, where it stops each time, and how much is lost at the moment of turning from forward to reverse. These are measured by a laser interferometer, and its measurement axis is a straight line.

The second type is the trajectory property of multi axis linkage: walking a circle, whether the actual path is a circle or a bit flat or square, whether the two axes move at the same time match, and whether there are steps on the trajectory when reversing. These are measured by a golf club instrument, which holds a fixed length rod between two points, and any cause that changes the distance between the two points is recorded as a change in rod length.

One sentence for division of labor: Interferometer measures the accuracy of “one axis walking straight”, and ball bar measures the coordination of “two axes walking together”.The collision simultaneously destroys both types, and neither type of normal can be deduced from the other type of normal.

1.4 Laser Interferometer: One sentence principle explanation and three measured items

Principle statement: Split a laser beam into reference light and measurement light. The measurement light hits the mirror on the moving part and returns. The number of interference fringes generated by the superposition of the two beams corresponds to the optical path difference, which is divided by the laser wavelength to obtain the displacement; Environmental sensors synchronously measure temperature, air pressure, and humidity, convert the laser wavelength in the air to a standard state, and obtain the true geometric displacement.Optics and refractive index correction are internal matters of the instrument, and only two prerequisites need to be remembered on site: environmental compensation and a stable optical path. No compensation, readings drift over time; The light path is unstable and the reading jumps directly.

It measures three items.

Positioning accuracy: The instruction reaches a certain position, and the difference between the actual arrival position and the instruction is measured point by point along the entire journey. The answer is’ Is the destination correct? ‘.

Repetitive positioning accuracy: the ability to walk multiple times in the same position and stop at the same place each time, with the answer being ‘walking the same time is different’. It is two different things from positioning accuracy – it can deviate by the same amount every time (good repetition, poor positioning), or it can average the position but scatter every time, the latter being more harmful and making compensation ineffective.

Reverse clearance: The difference in readings between forward and reverse at the same point, answer ‘how much was lost at the moment of reversing’. The repeatable parts can be compensated with parameters, while the random parts cannot. This difference determines the direction of subsequent disposal.

All three items must be tested after collision, because the most common result of collision is that the reverse clearance increases while the positioning accuracy does not change significantly – the clearance is hidden in the hinge point and only exposed during reversing. Only measuring one-way positioning, the curve may look good.

1.5 Ball Rod Instrument: One sentence principle explanation and four tested items

Principle statement: A rod with precision ball seats at both ends, one end fixed to the machine tool and the other end sucked into the magnetic ball cup of the probe seat. The machine follows the program to form a nominal circle, and any changes in the distance between the two ball centers are recorded by the displacement sensor inside the rod, resulting in radial deviation distributed with angle. **When walking a circle, two or three axes are moving simultaneously, so it measures the linkage property.

It reports four types of information.

Roundness: The degree to which a circle deviates from an ideal circle, manifested as the shape and amplitude enclosed by the radial deviation curve.

Radial deviation: The difference between the actual radius and the nominal radius at each angle, dividing roundness into directional information.

Servo mismatch: Shape distortion caused by inconsistent dynamic response of two axes, typically an ellipse in the diagonal direction, pointing towards the matching of drive and control, not geometry.

Reverse jump: The radial deviation peak at the reversal point (quadrant boundary) corresponds to the release of clearance, insufficient preload, and friction at the moment of reversal.

There are two easily overlooked prerequisites for a golf club instrument: the length of the club must be calibrated with a calibrator, and failure to calibrate is equivalent to no benchmark; The circular speed should be selected according to the detection purpose, with low-speed deviation towards geometry and clearance, and high-speed deviation towards dynamic following. Walking the same circle at two different speeds once each can separate static and dynamic errors.

2. Criterion layer: Three types of error fingerprints, two definitions of ISO 230-2, and one criterion summary table

The core of the criterion layer is not to provide a “qualified line number”, but two things: how to classify errors, and what evidence is used to determine whether each step passes. The testing method shall be carried out in accordance with ISO 230-2 (Determination of Positioning Accuracy and Repeat Positioning Accuracy), and the qualified limit shall be executed according to the contract acceptance indicators. The two shall not be mixed.

2.1 ISO 230-2 clearly distinguishes several similar words

Long term mixing of these words in spoken language can directly lead to misinterpretation.

Unidirectional positioning accuracy refers to the maximum range taken throughout the entire journey, which is the difference between the actual position and the target position when approaching the target position point by point in one direction. Bidirectional positioning accuracy introduces directional information on the basis of unidirectionality, taking into account the deviation of both directions reaching the same position, thus inevitably including directional differences. The reverse difference refers to the difference between two readings approaching the same position from both the forward and reverse directions, which directly exposes the gap and pre tension state. Repetitive positioning accuracy refers to the degree of dispersion between actual positions when approaching the same position from the same direction multiple times – it measures’ walking the same but different each time ‘, regardless of accuracy.

This set of definitions has direct use in interpretation: poor positioning but good repetition indicates the existence of stable and compensable deviations; **If the positioning is acceptable but the repeatability is poor, it indicates the presence of random sources (gaps, looseness, dirt) that cannot be solved by compensation. ‘The more you make up, the more chaotic it becomes’ is mostly due to not looking at the repetition accuracy first.

The standard has clear regulations on the sampling method: the allocation of positions, the number of cycles, and the order of approaching in both directions are written in the clauses. If the methods are not consistent, the data cannot be compared with each other; There are also requirements for average bidirectional positioning accuracy (comprehensive indicator of the entire journey), temperature measurement location, and data recording method. One rule: This article does not provide specific qualification limits, nor does it reference publicly available indicators of other models. The publicly available data tables for Equator 300 and 500 provide a comparison uncertainty of ± 0.002 mm (provided that the measurement is taken immediately after re mastering, the clamping position difference is ≤ 1 mm, and certified standard parts are used), which is the measurement capability of the comparison mode, not the geometric accuracy of the machine tool. The public accuracy term of X-500 is expressed in ISO 10360-2 (± 2.1 μ m+L/300, 18-22 ° C), and in comparison mode, it is expressed as a deviation from the standard part certificate value, which does not belong to the ISO 230-2 system. Mixing is the beginning of identifying accidents.

2.2 Three types of error fingerprints: using ‘what is related to’ as a criterion

The essence of interpretation is to find correlation: errors of the same amplitude are related to position, direction, or velocity, and point to completely different directions.

Related to location.The error increases monotonically with the stroke, suddenly jumps at a certain position, or fluctuates periodically. Monotonic growth is usually a proportional error; The reading problem (dirt, scratches) or jamming of the local jump pointing grating at that position; Periodic fluctuations indicate periodic errors in the transmission chain. This type of error is highly reproducible and most easily compensated and reduced, but the prerequisite is to first confirm that it was not hit out – deviations caused by hinge clearances are often disguised as position related.

Related to direction.The forward and reverse readings are different at the same position, or when moving in different directions, the error direction rotates accordingly. The typical sources are clearance and pre tension: the reverse difference stabilizes beyond the limit, pointing to the hinge point clearance, contact between the ball head and the seat ring, and insufficient pre tension.

Related to speed.Using different feed rates for the same stroke, the error varies with the speed; Or sudden amplification of errors in the deceleration zone and reversing zone. This type of direction control side: servo gain and filter, acceleration and deceleration planning, friction compensation, drive and load matching. The common reason for the sudden increase in such errors after a collision is not geometry, but the change in the friction torque of a certain strut, while the control still follows the original parameters.

The usage rule is: First confirm what is irrelevant, then confirm what is relevant.The error after collision is often mixed, and comparing the results of different speeds can cut off the speed factor, and then see whether the remaining changes with position or direction.

2.3 List of post collision detection criteria

Serial number Test item Instrument/tool used Passing criterion (how to pass) Point when not passing
1. Appearance and structure rough inspection Visual inspection, flashlight, lint free cloth No bending of stylus, no breakage of ball head; There is no debris, oil leakage, or visible cracks in the strut and hinge point area; Platform without jamming Damaged stylus or probe → Rework item 5 after replacing parts; Oil stains on hinge points → suspicious frame, transfer to professional evaluation
2 Geometric rough inspection (straightness, flatness, perpendicularity) Ruler, dial gauge+magnetic gauge holder, level The dial gauge reading is within the contracted geometric indicators and there is no trend change before and after collision; Level without drift local changes → mechanical interference or loose fastening; Overall change → leveling and foundation or structural deformation
3 Single axis positioning accuracy (including reverse difference) Laser interferometer+environmental compensation Sampling and cyclic collection according to ISO 230-2, positioning accuracy, bidirectional positioning accuracy, and reverse difference are all within the contract indicators Reverse difference exceeds the limit → Joint clearance or pre tightening; Positioning exceeding the limit with good repeatability accuracy → proportional or scale deviation; Poor repeatability accuracy → random source
4 Three axis linkage circle (one wheel at low speed and one wheel at high speed) Ball bar instrument+calibrator The peak and peak values of the radial deviation of both wheels are within the contract specifications; No sudden changes in shape, quadrant peaks within allowable range ellipse → servo mismatch; Four petals → geometry; Quadrant peak → reverse jump; High speed is significantly inferior to low speed → dynamic following
5 Probe and Calibration Review Standard Ball, Deviation Reading, Needle Replacement Rack Process Deviation ≤ 4 μ m; The multi-point repeatability of the standard ball returns to the level before the collision; Needle consistency meets standard Deviation exceeds limit → Needle and probe module; Needle position difference → Remake needle holder positioning (Part 1/Part 2)
6 Comparison benchmark review Certified standard parts, original measurement program Measure immediately after re mastering, clamping position difference ≤ 1 mm, standard part retest matches certificate value Benchmark does not match → Mastering or calibration process issues; Geometry has passed but still does not match → Software and process side

The deviation of in the table is ≤ 4 μ m (path UCCServer Equator → Advanced → Equator Server → Coordinate Measuring Machine Controller → General → Measuring Head Deviation) and clamping difference ≤ 1 mm are from official documents, and the rest are in accordance with the contract and standards.

3. Disposal layer: Detection sequence and step-by-step operation after collision

The sequence cannot be skipped, because each step is creating conditions for the next step: rough inspection eliminates “physically broken”, single axis inspection fixes the linear properties of each axis, linkage inspection looks at coordination under this premise, and the probe and calibration review are sent to the end of the measurement chain. Conversely, the common outcome is holding a beautiful circular chart but not knowing where the peaks come from.

3.1 Step 0: On site security, prior to any instrument entering the site

The most difficult thing to supplement after a collision is information, not accuracy. Before anyone touches the machine, fix the scene: alarm code and time, which program and coordinate the collision occurred in, whether the stylus is bent, whether there is any abnormal noise, contact marks between the workpiece and the fixture, environmental temperature and humidity; Take photos of the probe and ball head, the probe and strut surroundings, and the contact point and screen alarm.

Then perform the first visual inspection in the power-off state. Once the stylus has bent, remove it first and replace it with a known good stylus for subsequent testing to avoid mixing stylus errors with geometric readings.

Safety reminder: After turning off the power, make sure there are no residual movements before reaching out; Don’t try to run the program back to zero by powering it on first, as bearing a cyclic load on the damaged hinge point will only increase the gap; Do not allow anyone other than the testing personnel to operate the machine.

3.2 Step 1: Geometric rough inspection, use the cheapest tool to eliminate major problems first

The tools are a ruler, a micrometer with a magnetic gauge holder, and a spirit level, aimed at quickly eliminating “macroscopic structural changes”.

Action 1 | Flatness of fixture board and platform reference surface.The micrometer is attached to a stable reference, and the probe lightly touches several evenly distributed positions on the fixture plate. The moving platform reads one by one and compares it with before the collision.

Action 2 | Conduct a rough check of verticality and parallelism.Use a ruler or inspection rod with a dial gauge, and mark the gauges in two vertical directions to check the total change. The accuracy is not high, but it is enough to detect ‘obvious deviation’.

Action 3 | Horizontal state.The level meter reads in both directions and compares it with the leveling record. The overall situation has changed, and we first suspect the feet, legs, and tabletop – collisions often occur with the movement of the entire machine’s position.

Action 4 | Confirm the tactile sensation of gap and clearance.According to the official operation boundary: the installation guide only states that moving/lifting the entire machine requires two people to use the base casting handle, and does not provide any manual operation instructions for the user to push the motion platform. After the power is turned off, caution can be taken to use tactile assistance to judge (two people cooperate, only walk a small section, and do not stop halfway). Do not manually push the 500 and X-500 (they are heavy and difficult to control), and use a controller to jog and compare with three axes instead;It is strictly prohibited to push or pull the platform while it is powered on. Only answer whether there is obvious empty space left, not how much empty space left.

Direction that does not pass: sudden change in local reading → mechanical interference, loose fastening or local damage at that location, stop the machine for troubleshooting first; Overall change → leveling and installing foundation.

Step 2: Single axis positioning accuracy, measured axis by axis using a laser interferometer

This step is the easiest to fail, and the problem is usually not that the instrument is inaccurate, but that the optical path is incorrect.

Action 5 | Determine the measurement axis and optical path.The measuring axis must be parallel to the motion axis. Equator is a three pole parallel structure, where the direction of the pole is the driving direction, and the measurement axis is arranged along the axis of the pole. The cosine error introduced by non parallel optical paths will amplify with the stroke and disguise as “position dependent”, which is most likely to mislead the interpretation.

Action 6 | Environmental compensation in place.Place temperature, air pressure, and humidity sensors on the optical path near the middle of the measuring axis, avoiding the air outlet and door, and collect data after stable startup.

Action 7 | Collect according to ISO 230-2.Sampling points, cycle times, and positive and negative directions shall be executed according to standards, and the suspected section shall be encrypted. Do not change the air conditioning, open the door, or walk around during collection.

Action 8 | Read three things and compare them with the table below.The three criteria are the contract acceptance indicators, and the combination relationship is the key to diagnosis.

Reading combination Meaning Pointing Next action
The positioning accuracy exceeds the limit, the repeatability accuracy is normal, the curve is smooth and monotonic stable and repeatable deviation proportional, scale or kinematic solution class; It could also be a stable geometric offset preserving the original data; Confirm the source first and avoid using compensation to mask geometric changes
The positioning accuracy exceeds the limit, the repeatability accuracy is normal, and there is a local jump in the curve There is a reproducible abnormality at this position There is a local reading problem with the grating (dirty, scratched), local jamming Cleaning and appearance review, operate according to the official kit, and disable cleaning agents; There is still a transition from professional evaluation to professional evaluation
Repetitive accuracy exceeds the limit, multiple curves scatter Random source Gap between hinge points, loose fasteners, connectors, grating signals Check fasteners and cables first, then check gratings; Don’t change the parameters first
Reverse difference exceeds the limit, other items are still acceptable Reversing loss Joint clearance, contact between ball head and seat ring, insufficient pre tightening Record the original readings in both directions; Professional evaluation without on-site adjustment and tightening
The overall translation of the entire line has excellent repeatability reference offset return to zero or reference point, clamping and installation status reset to zero and retest; Confirm that the installation foundation remains unchanged
Error varies with speed, low-speed normal and high-speed degradation Dynamic tracking Servo parameters, friction compensation, drive and load matching First make low-speed reference, gradually increase speed and retest

Step 3: Three axis linkage circle, use a cue stick to check the coordination

After all the single axes are qualified, the linkage test answers’ still not accurate when moving together ‘.

Action 9 | Clamping and rod length calibration.One end is fixed in a rigid position on the fixture board, and the other end is sucked into a magnetic ball cup on the side of the probe seat. The rod length is first calibrated with a calibrator. The clamping position should cover the typical working area.

Action 10 | Walk in a circle at low speed.The purpose is to suppress dynamic factors and make the reading lean towards geometry and gaps. The criterion is to compare with the contract indicators and determine if the shape is distorted.

Action 11 | Walk in a circle at high speed.At the same position, use a higher input speed and walk again. The differential novel indicates good dynamic following, while obvious deterioration indicates control side issues. The increase in peak size indicates that the reverse jump has been dynamically amplified.

Action 12 | Repeat by changing positions and planes.At least two different heights or radii should be grouped together to form a comparison matrix of “same position but different speed” and “same speed but different position”.

Comparison of shape and source.

Circle shape Morphological features Main direction Verification method
Near circle, small amplitude Radial deviation curve close to flat Normal Comparison with contract indicators and baseline before collision
Ellipse, with the long and short axes at a 45 ° angle Deviation in opposite diagonal directions Mismatch between servo gain and dynamic response on both axes High/low speed comparison, with high speed being more pronounced, confirmed
Ellipse, with the long and short axes along the coordinate axis maximum deviation along both axes a certain axis ratio or positioning error is converted into radial deviation in linkage locate the specific axis based on single axis data
Four petal shape Converging along the diagonal and concave along the coordinate axis Geometric categories: perpendicularity, straightness, platform posture Geometric rough inspection and cross validation of uniaxial data
Peak at quadrant boundary Narrow peak at reversing point Reverse jump: instantaneous release of clearance, preload, and friction Compared with the single axis reverse difference, if both exceed the limit at the same time, they will be locked
One side drum and one side flat Asymmetric, monotonically changing with angle Unidirectional error or uneven friction on this side Bidirectional uniaxial detection to check load distribution
The starting and ending points do not coincide One circle does not close Collection and programming, repeated positioning, temperature drift Repeat once to see if it is reproduced

Read the order of the linked circles.First, look at the general ledger (whether the peak to peak value is within the contract indicators), then look at the shape (whether it is symmetrical and in what shape), then look at the change with speed (the ratio of low/high speed circles at the same position, the control class deteriorates significantly with speed, while the mechanical class remains basically unchanged, which is the most effective single criterion for distinguishing between mechanical and control), and finally look at the change with position (changing radius and height, geometric class is sensitive to position).

Safety Reminder: The golf club is not designed to withstand collisions, and the lateral force that obstructs the movement of the ball can damage the club. The radius, center, and safe height must be confirmed by running empty first.

Step 4: Probe and Calibration Review, Close the Measurement Chain

Action 13 | Check the stylus and probe.Check the straightness of the shaft and the appearance of the ball head; The criterion for probe deflection is ≤ 4 μ m. If it exceeds the limit, check the probe and probe module first.

Action 14 | Standard ball multi-point repeatability.The same standard ball triggers in multiple directions, observe the dispersion and compare it with before the collision. Collisions often occur in a single direction, and damage is often directional.

Action 15 | Needle changing rack positioning.The collision involves the needle changing rack area or the needle changing rack has been moved and disassembled, and Part 1 and Part 2 of EQR-6 positioning must be redone.

Action 16 | Calibration and Mastering Review.After passing all the previous steps, redo the calibration and mastering, and then immediately measure without inserting any other tasks – this is the primary prerequisite for comparing uncertainty within ± 0.002 mm.

Action 17 | Record and archive.Each item should include the instrument number and calibration validity period, environmental temperature and humidity, optical path or clamping diagram, original data, and conclusion. The original data is more important than the conclusion table, and only it can be re interpreted after a year.

4. Boundary layer: External detection cannot solve much, what situations must be stopped

The value of external detection has a clear upper limit: it measures and categorizes errors, but does not repair anything and cannot replace the evaluation of the structure. Writing clear boundaries is to avoid two types of misuse: treating the results as a repair plan or taking action in areas that should not be touched.

4.1 Three things that external testing cannot answer

External testing can be qualitative and quantitative, but there are a few things it cannot answer, and confusion can overestimate the weight of this evidence.

Where is the error source located inside the component.The interferometer tells you that ‘the reverse difference in the direction of this strut is X’, without saying whether this X falls on the upper hinge point, lower hinge point, or drive end. The component needs to be disassembled and inspected, which is not within the scope of inspection.

Whether there are invisible cracks in the structure or whether it can return to its factory state.The hidden damage of internal microcracks and hinge seat rings in castings may only show small deviations or even not at all in geometric data; Testing can only speak to the contract indicators and baselines, and if the indicators are wide, it cannot indicate a return to the factory state. So ‘data is okay’ cannot be used as proof of structural integrity, especially when there are obvious impact sounds or high-energy collisions.

The economic consequences of errors.Whether the deviation affects the qualification judgment of the parts depends on the part tolerance and tolerance allocation, not on the reading itself, and needs to be judged jointly by the process and quality departments.

4.2 Situations where it is necessary to stop and switch majors for evaluation

  • There are obvious signs of the structure: cracks, deformation, and oil leakage can be seen in the strut or hinge area, and there are abnormal traces around the fasteners.
    –Stuck, abnormal noise, overload: Do not “walk once to see if it can go smoothly”, as these phenomena will expand the damage with exercise.
  • Grating related: There is a strong position related jump in the interferometer data, which persists after cleaning and visual inspection. Scratches on the grating body and malfunctions in the reading system are not within the scope of on-site disposal; Officially, only the cleaning kit A-5504-9055 and controller fan filter A-5696-0120 are self maintainable, and cleaning agents are prohibited.
    –Geometric relationships need to be reconstructed: Deviation in verticality and platform posture, there are no conditions on site that can be reconstructed without dismantling the machine.
    –500 and X-500 manual pushing platform: The official has not provided this operation guide. Do not manually push (due to its high quality and difficulty in stopping). Instead, use a controller to jog and compare with three axes.

5. Verification layer: How to use the same set of instruments to prove ‘okay’ after repair

The inspection after collision needs to be done twice: qualitative and quantitative inspection before repair, and closing inspection after repair. The same set of instruments, optical path layout, sampling points, and environmental conditions must be used twice, otherwise the comparison is not valid.

5.1 Four “same” principles for retesting

Same instrument.The same interferometer, ballbar, and environmental sensor set. Changing the instrument means changing the entire uncertainty budget.

Same arrangement.The height direction of the optical path, the reflector bracket, the fixing point of the cue stick and the ball cup, and the radius and plane of the circle are all consistent. When the layout changes, the cosine error changes and the readings are incomparable.

Same sampling.The number and allocation of sampling points, the number of cycles, the arrangement of positive and negative directions, and the speed and direction of rounding are all consistent.

Same environmental caliber.The temperature and humidity should be consistent with the recording time, preferably done at the same time period. When the four requirements are not met, truthfully state ‘cannot be directly compared’ and do not force to provide a percentage improvement.

5.2 Post repair acceptance form

Acceptance criteria Methods Criteria Record fields Disposal in case of failure
Geometric rough inspection retest micrometer, level, the same measuring points as before repair trend improvement or restoration to the pre repair state, and within the contract indicators readings, position indication, temperature local abnormalities of each measuring point
Single axis positioning accuracy retest Laser interferometer, same optical path, same sampling, including environmental compensation Positioning accuracy, bidirectional positioning accuracy, and reverse difference are all within the contract indicators and comparable to the baseline Complete error curve, forward and reverse original readings, discrete Reverse difference still exceeds the limit → Gap not eliminated; Poor repeatability → random sources still present
Three axis linkage circle retest Ball bar instrument, one round at the same position, low speed and high speed Radial deviation peak to peak value within the contract specifications, no characteristic distortion in shape Polar coordinate raw data, peak to peak value, shape Ellipse still in → Servo matching; Four petals still in → geometry; Peak still in → reverse jump
Probe deflection Official viewing path reading ≤ 4 μ m Reading, time Exceeding limit → Replace probe or probe module and retest
Standard ball repeatability Standard ball triggered in multiple directions, under the same conditions Discretize back to the level before collision Readings and range in each direction Large dispersion in a specific direction → Mechanical or probe abnormality in that direction
Consistency of needle replacement Two needle positions are measured with the same standard ball Needle position consistency meets the standard Reading and difference of each needle position Remake the needle replacement frame Part 1/Part 2
Comparison benchmark review Testing standard parts immediately after re mastering Consistent with certificate values; Clamping difference ≤ 1 mm Actual measurement value, certificate value, mastering time Inconsistent → Re check calibration and mastering process

5.3 How to read ‘There is still a slight deviation after revision’

Minor deviations after repair are common, depending on which category it belongs to.

Acceptable: Deviation is stable, repeatable, and of the same magnitude as the baseline before the collision, indicating that the structure has returned to its original state, and the residual error is the inherent error of the machine.

What needs to be observed: The deviation is stable but slightly larger than before the collision, and the direction is single. Perhaps there is a slight difference in the assembly status of a certain area compared to the original. The approach is to observe it closely and see if it converges or expands over time; Continuously expanding will be treated as unacceptable.

Unacceptable: Deviation varies with position, direction, or speed. This indicates that the geometric relationships have not been completely reconstructed and cannot be suppressed by adjusting parameters.

5.4 Test report skeleton

The report should be written in the following order so that readers can follow the interpretation chain: a section of background (collision time, location, program segment, alarm code, on-site key points); A set of conditions (instrument model and calibration validity period, environmental temperature and humidity, optical path and clamping diagram, sampling basis standards); A pre repair data sheet (for geometric rough inspection, single axis three items, linkage two groups, probe and calibration, providing readings and judgments for each item); A paragraph of interpretation conclusion (classified according to three types of fingerprints, providing the basis for pointing to machinery or control); A section of disposal and retesting (what was done, retest data, comparison before and after repair); A statement of limitations (which conclusions are subject to conditional limitations, which ones lack baselines, and which ones need to be tracked). Finally, attach the original data list and the schematic diagram of the measuring point optical path.

6. Supplementary layer: One complete interpretation (desensitization) and two quick reference tables

6.1 Quick Reference Table 1: Testing Items, Instruments, and Non directional Inspection

Testing items Instruments What to collect (criteria are based on contract indicators and baseline before collision) Not directed
Rough check of flatness and verticality Micrometer+gauge seat, ruler Reading and total change of each measuring point Local interference or fastening; Overall installation foundation
Horizontal state Level Two directional reading Foot, leg, table movement
Positioning accuracy Laser interferometer Difference between actual position and instruction of each sampling point Proportional, scale, kinematic solution class; Local grating inspection and jamming
Repetitive accuracy Laser interferometer Discrete approach to the same point multiple times Joint gap, fastening looseness, grating signal, connector
Reverse difference Laser interferometer Reading difference between forward and reverse directions reaching the same point Joint clearance, preload, friction
Roundness and radial deviation Ballbar apparatus Distribution of polar coordinate radial deviation with angle Four lobes → geometry; One side drum → unidirectional error
Servo Matching Ball Rod Instrument (Low Speed vs. High Speed Comparison) Elliptical Shape and Its Variation with Speed Servo Gain, Filter, Acceleration and Deceleration Planning
Reverse jump Ballbar apparatus Peak amplitude and width at quadrant boundary Gap, preload, friction released at the moment of reversal
Probe status Probe deflection reading, standard ball Deflection value and multidirectional repeatability Probe and probe module
Needle consistency Standard ball, needle replacement rack process Measuring the difference of the same ball at different needle positions Remake EQR-6 positioning Part 1/Part 2

6.2 Quick Reference Table 2: What to Test and When Not to Test

Situation Suggestion Reason
Low energy probe collision, normal appearance probe deflection+standard ball repeatability, if necessary, single axis retesting low energy, low probability of geometric impact
High to medium energy collision with abnormal noise Four steps in the entire process Load has been transmitted to the hinge point and platform
After moving and changing positions Geometric rough inspection+single axis+linkage Installation foundation and geometry may change
After major repairs or replacement of hinge components Full process+pre repair and post repair comparison Need to prove that the reconstruction is in place
Environmental conditions are not met Only perform relative method testing and declare limitations Absolute readings are unreliable and cannot be used for acceptance

Frequently Asked Questions (FAQ)

Q1: The machine comes with a built-in probe. After running the program once to see if it is accurate, can’t it prove that the spatial accuracy is not compromised?

No. The position read by the machine comes from the grating on the driving strut, which provides feedback from the same source as the driving: the grating says how much the strut has gone, and the controller considers the platform to be where it is. The hinge point gap, slight deformation under the load of the strut, and displacement of the probe component relative to the platform all occur after the grating reading and before the real spatial point, and the controller has no way of knowing. The screen displays not the actual position, but the position believed by the machine.

Q2: Which of the three parameters measured by the laser interferometer is the most sensitive to the collider?

Reverse gap. The collision load is mostly transmitted along the hinge point, and the most direct result is a change in the contact state between the ball head and the seat ring, insufficient preload, and an increase in clearance, which are only exposed at the moment of reversing. The accuracy of one-way positioning may vary very little because the preload force is always pressed in the same direction when walking forward, and the gap is not reflected when it is pressed. Only measuring one-way positioning is the most prone method to miss judgment, and it is necessary to arrange the collection in a cyclic manner according to ISO 230-2 in both directions, and calculate the reverse difference separately. Repetitive positioning accuracy can also indicate mechanical status, but it is sensitive to gaps, looseness, and dirt, and its directionality is not as good as the reverse difference.

Q3: Can the golf club instrument measure circles? Does it include all errors and eliminate the need for an interferometer?

Cannot be replaced. The radial deviation read by the golf club instrument is the sum of the projections of all error sources in the circular direction: if two axes move simultaneously, the errors are superimposed, and it is impossible to separate which axis is different in which position. It provides a general ledger and shape, but cannot provide a single axis curve. Space accuracy has two types of properties – single axis straightness and multi axis coordination, both of which are destroyed by collision. In terms of sequence, interferometers are used first and then ballbar instruments, because if the reverse clearance of one axis is already large, geometric distortion and reverse jump will occur simultaneously on the circle, making it difficult to distinguish the source.

Q4: A nearly perfect ellipse appears on the circular chart. Is it due to geometric problems or control issues?

In most cases, it is a mismatch between the two axis servos. The criterion is that the long and short axes are along the diagonal or coordinate axis, and the high-speed circular ellipse is more obvious, while the low-speed one is close to a circle – the shape changes with speed, indicating that it is caused by dynamic following, not structure. On the other hand, if the four petal shape is clear at low speeds and still clear at high speeds, it is more like a geometric problem. The two treatments are completely different. The former adjusts the servo, while the latter moves the geometry, so it is necessary to look at the low-speed circle and the high-speed circle together.

Q5: Can the positioning accuracy of ISO 230-2 be judged as qualified with ± 0.002 mm?

No. ± 0.002 mm is the comparison uncertainty between Equator 300 and 500, which describes the ability to measure parts in comparison mode, and has prerequisites: measure immediately after re mastering, clamp position difference ≤ 1 mm, and use certified standard parts. It does not belong to the positioning accuracy system of ISO 230-2. The precision term disclosed by X-500 is based on ISO 10360-2 (± 2.1 μ m+L/300, 18-22 ° C), not ISO 230-2. The qualified limit can only be implemented according to the standard execution method plus the contract acceptance indicators.

Q6: If all the test data is qualified, can production be directly resumed?

We still have one step left. The qualified data indicates that the testing items are within the contract indicators, but to resume production, three things need to be confirmed: whether they are comparable to the baseline before the collision, whether the probe and calibration chain have been reviewed, and whether the comparison benchmark has been re established. Especially for the third item – geometric qualification does not mean that the benchmark is valid, the latter requires immediate measurement after re mastering, clamping position difference ≤ 1 mm, and the use of certified standard parts. Suggest encrypting and retesting the first few batches after repair to observe whether there is a trend in repeatability and retesting of standard parts. The length and frequency of the observation period are common practices in the industry and should be agreed upon by the user and the process department.

Q7: If you see that the reverse difference exceeds the standard, can you first add gap compensation in the software?

It is not recommended to do so before confirming the source. The reverse gap can be divided into two parts: repetitive and random. The former can indeed be compensated and reduced, so compensation is effective in some cases; However, the gap caused by the collision is often accompanied by an increase in random parts, and compensation will reduce the repeatable deviation and leave the random dispersion in place, which may appear to improve the data, but the actual repeat positioning accuracy may be worse. The correct order is to first determine: whether the reverse difference is stable or scattered, whether there is a degradation in repeated positioning accuracy, and whether it is amplified at high speeds.

Q8: Can environmental compensation be simplified by using fixed values without connecting sensors?

No way. The significance of environmental compensation lies in converting the changes in laser wavelength with temperature, air pressure, and humidity; Using a fixed value is equivalent to assuming that the environment remains unchanged, which never holds true in the workshop. The typical situation is to adjust the machine in the morning and measure it in the afternoon: a change of a few degrees in room temperature is enough to make the positioning error differ by an order of magnitude within a few tens of meters of travel. The feasible approach is twofold: first, allow the sensor to be fully isothermal with the workpiece and guide rail before starting; Secondly, do not write the compensation value as a constant and use it for a long time. Collect it again before each detection. If the on-site conditions are indeed unable to connect the sensor, it should be stated in the report that “no environmental compensation has been made” and the error should be included in the uncertainty, rather than quietly using a fixed value to suppress the data.

Data source explanation

The technical parameters related to the Equator ontology in this article are taken from the following official documents: Equator 300 data sheet H-5504-8251-06-A (2026-02), Equator 300 specification manual H-5504-8252-04-A, Equator 500 data sheet H-6078-8310-01, Equator-X 500 data sheet H-6620-8560-01-A (2025-06), Equator comparator usage guide H-5504-8500 (ID 145539), Equator installation and controller guide (ID 132816/107657).

  • The three preconditions for probe deflection criterion ≤ 4 μ m and viewing path, cleaning kit A-5504-9055 (cleaning agent prohibited), controller fan filter A-5696-0120, EQR-6 needle changing frame positioning (Part 1/Part 2), and ± 0.002 mm comparison uncertainty are all cited from the official documents listed above.
  • The working temperature of 300 is+5~+50 ° C in English 06-A and the user manual, and+10~+40 ° C in Chinese 04-A. The text has been listed and the following single version is indicated as the standard.
  • The working principle description of laser interferometer and ball bar instrument belongs to the general measurement principle and does not include any brand’s model, accuracy or range parameters; When these data are needed, please refer to the official data sheet of the corresponding manufacturer.
  • The definitions of positioning accuracy, bidirectional positioning accuracy, repetitive positioning accuracy, and reverse difference in ISO 230-2 are expressed in accordance with their standard clauses; This article does not provide specific qualification limits, and qualification judgments will be based on standard execution methods plus contract acceptance indicators.
  • The classification method of three types of error fingerprints (related to position, direction, and speed), the correspondence between the shape of the cue stick and the error source, and the practice of detection records and observation periods are common practices in the detection industry and are not official specifications. Please distinguish them from official parameters when quoting.
  • The case in the article is reconstructed based on real working conditions, and the data has been desensitized and does not refer to specific customers or parts.

Need a specific repair plan?

Every Equator fault eventually comes down to a specific machine type, duty cycle and batch. The criteria and procedures in this article can be used as a shop-floor checklist; when something does not add up, talk to us directly:

Item Detail
Company NINGBO JCE INDUSTRIAL TECHNOLOGY CO., LTD. (JCETech)
Address Room 12B17, Building 2, Zone A, Ningbo R&D Park, Juxian Subdistrict, Hi-Tech Zone, Ningbo, Zhejiang 315048, P.R. China
Contact Mr. Li / Mr. Zhang
Phone +86 157-5780-7400 (Mr. Li), +86 132-2194-1413 (Mr. Zhang)
Online booking jcetech.cn/contact

Repair intake and on-site service are handled at the same address. Please call before shipping a unit so transport and scheduling do not have to be repeated.