Introduction

After the collision, the machine was powered on again and all three drive rods were able to move, which only indicates that the motion link is connected, not that the geometric relationship is still present. Equator is a parallel motion structure, with three drive struts connected at both ends by Hooke hinges and only capable of withstanding tension and compression. Any impact will inject the load into the hinge point, floating platform, and top casting. The order of recovery is fixed: mechanical reset, leveling and anchoring, repositioning of needle changing frame, calibration mastering, Re testing verification must be completed in six steps with acceptance criteria to return to a comparison uncertainty of ± 0.002 mm (this value is subject to immediate measurement after re mastering, clamping deviation ≤ 1 mm, and the use of certified standard parts).

1. Original text level: Why is “active” still one whole geometric reconstruction away from “accurate measurement”

1.1 The error propagation path of parallel structure is completely different from that of series structure

To understand the handling after a collision, first think about the structure. Equator is not a series structure of “column hanging cross arms”, but a parallel kinematic mechanism (PKM): three linear drive struts correspond to the P/Q/R axes, with one end of the strut connected to the fixed side and the other end connected to the floating/moving platform, both ends of which are Hooke hinges. The SP25 probe assembly is mounted on a floating platform, which is constrained by both the driving strut and the supporting strut.

The official component diagram shows the following parts: Base casting, SP25 probe assembly, floating platform, support strut, top casting, drive strut, counterbalance mechanism, upper pivot/Hooke’s joint, support arm, drive housing, parallel constraint mechanism, fixture plate, and stop button. This structure has no screw, no belt drive, and no Z-axis column – displacement is directly generated by the linear drive of the strut, and the grating is mounted on the strut. The driving point and feedback point are the same source.

The first conclusion is that position error is not one-dimensional. One shaft in the series structure collided, and the worst case scenario is that the bearing or guide rail of that shaft is damaged, while the other two shafts are basically innocent. In a parallel structure, each spatial coordinate of the moving platform is determined by the length of three struts simultaneously. A slight attitude deviation on the platform will be mapped to the readings of the three struts at the same time, manifested as “all three axes are present, but the spatial point is incorrect”.

The second conclusion is that the strut only bears tension and compression, and has no bending moment bearing capacity. This is not a disadvantage but a design premise – rigidity comes from closed geometric constraints, without relying on sturdy components to resist bending and hard top. The cost is that the collision energy is not “bent” inside the component, but transmitted along the hinge point: entering the connection end between the Hooke hinge pin and the bearing, the strut and the platform, and the driving end (which may trigger overload protection). So the most suspicious thing after a collision is not the thickest looking casting, but the invisible hinge clearance.

What are the layers between “active” and “accurately measured”

Breaking down the recovery into four layers reveals why the six steps cannot be saved.

The first layer is the electrical and limit layer. After collision, the controller may enter an error state, the drive may enter protection, and the limit may have been activated. After powering on again and clearing the errors, the machine can move, indicating that this layer has passed. The most common misjudgment is here: seeing “successful zeroing and normal positioning” means it has been fixed.

The second layer is the mechanical reset layer. The collision may push the platform to an abnormal posture before stopping, and the hinge point, counterweight mechanism, and support arm may stop off center. The criterion is not ‘able to walk’, but ‘able to return to the design position and stop stably’. The function of the counterweight balance mechanism is to counteract the gravity of the platform and the probe assembly. If it is displaced or stuck after collision, the phenomenon is often “particularly difficult in one direction, slowly sinking after stopping”, and it is also kinematic.

The third layer is the geometric reference layer, where the accuracy truly lies: the geometric relationship between the machine coordinate system and the fixture plate, needle changing frame, and standard parts. When the collision reaches the hinge point or platform, this relationship becomes invalid. The typical manifestation is that the repeatability of the single point is good, but the systematic deviation in different spatial positions, or the consistency of each needle position deteriorates after needle replacement.

The fourth layer is the benchmark layer for comparison. ± 0.002 mm is the uncertainty of comparison, not absolute accuracy, with three prerequisites: immediate measurement after re mastering, clamping position difference of parts relative to standard parts not exceeding 1 mm, and use of certified standard parts. Continue testing after only completing the first two layers, using a reference benchmark that has already failed. The direction of deviation depends on the geometric relationship of the failure rather than the part itself.

These four layers explain the arrangement of six steps: the first three steps align the machine mechanically and geometrically, the fourth and fifth steps reconstruct the reference, and the sixth step proves the validity of the reference. Jumping will only push the error down to the next level, ultimately resulting in ‘varying degrees of accuracy’.

1.3 What is a ‘collision’ and where are the boundaries

In engineering, collision is not just about hitting a plane. From low to high energy: the stylus or probe collides with the workpiece and fixture (most commonly, with the lowest energy but highest frequency); Driving or program errors causing the platform to collide with workpieces, fixtures, or needle changers (strong impact of moderate energy); Contact between tools and measuring machines during transportation or clamping (external impact); Falling or lateral impact caused by whole machine handling, forklift collision, and unstable lifting (high-energy, usually accompanied by damage to the shell and castings).

Different energies correspond to different depths of disposal: low-energy collisions usually only require needle inspection, single point repeatability retesting, and one mastering; High to medium energy must complete six steps; For high-energy and visible casting cracks or deformations, directly enter the boundary judgment in the fourth section.

A commonly overlooked point: A snapshot of the collision scene is much more valuable than a retrospective.. After a collision occurs, before anyone touches the machine, record the position of the probe, program segment number, alarm code, platform posture, whether there is any abnormal noise, whether the stylus is bent, and where it stops on the screen. These pieces of information determine which geometric quantities may be altered later on. Once someone withdraws the knife first, clears the alarm first, and returns to zero first, the causal clues are broken.

2. Criterion layer: Divide the ‘damage’ into five levels, with each level providing executable inspection actions

2.1 Why is it necessary to grade before starting

The first reaction after a collision is usually ‘should we find someone to fix it?’. This question cannot be answered before classification: the depth of repair depends on the depth of damage, which can only be obtained through inspection. The purpose of grading is to replace ‘looks quite serious’ with a set of on-site executable and reproducible inspection actions.

There are three logical axes for grading: whether the appearance is distinguishable, whether the motion is free, and whether the accuracy is reproduced, progressing from “rough but certain” to “fine but requires baseline”. Grading itself is not an official regulation, but a compilation of industry practices; However, the parts of the Equator body involved in each level of criteria (such as probe deflection, needle holder positioning, mastering, and pre conditions for comparison uncertainty) are supported by official documents, as shown in the table below.

2.2 Level 5 Damage Grading Table

Level	Name	On site executable inspection actions	Passing criteria (how to pass)	Typical phenomena and their impact on the conclusion of this level
Level I	Appearance inspection	Visual inspection in power-off state: stylus ball head and shaft, probe component housing, strut and hinge point periphery, floating platform, fixture plate and stop button area; Gently push the platform with your hands (power off, gentle) to feel for any jamming or looseness	The stylus is not bent or the ball head is broken; The shell has no cracks or dents; There is no debris or oil leakage in the hinge area; Lightly pushing without any metallic friction sound	The appearance passes completely, only indicating that there is no obvious structural damage, and it cannot be concluded that the geometry is intact
Level II	Freely movable	After powering on and clearing the alarm, use a jog or low-speed program to make each of the three struts travel through the entire range, and observe whether there are any abnormal noises, overloads, or position deviation alarms	There are no alarms, abnormal noises, or jamming in the entire three-axis process; The direction of motion is consistent with the instructions; Limit and emergency stop actions are normal	Single axis jamming or abnormal noise occurs, pointing to the hinge point or drive end of the strut, requiring mechanical inspection before level III
Level III	Able to return to zero	Perform zeroing (home/reference point reset), then repeat 3 times to see where to stop and whether the deviation is consistent each time	Successfully return to zero for 3 consecutive times and stop at the same position (the difference falls within the level of machine display resolution, please refer to the official manual for details); After returning to zero, the screen coordinates remain stable and do not drift. If the return to zero fails or the stop position is different each time, it may indicate a gap between the hinge points or a jump in the grating reading, which is a geometric layer problem
Level IV	Single point repeatability	Select a position on the fixture board, trigger the same standard ball (or the same point on the standard part) with the same stylus, direction, and speed 10 times repeatedly, and record the readings	The range of the 10 readings falls within the historical level of the machine under normal conditions (with the same condition data before collision for comparison, the specific values are subject to the official data sheet/manual)	The range is significantly enlarged → The mechanical and measuring chain are suspicious; Normal range but mean offset → geometric benchmark offset, focus on level V
Level V	Spatial Accuracy	Take at least three measuring points of different heights and positions in the working area of the fixture board, measure them separately, and compare them with standard parts through retesting; Check the consistency of each needle position on the needle changing rack	The deviation direction and magnitude of each measuring point are consistent and comparable to the baseline before the collision; Consistency between different needle positions meets the standard; The retest results of the standard parts are consistent with the certificate values. The deviation directions in different spatial positions are not consistent (such as more outward and higher deviation). The geometric relationship has changed and the complete six steps must be taken

The reading method is to advance step by step from top to bottom, without jumping: passing level I does not mean passing level II, and normal level IV does not mean normal level V. The failure mechanisms of the two are completely different.

2.3 The two most easily misjudged combinations

The first combination is “good repeatability but spatial bias”. Single point repeatability only proves that ‘the same point hits the same place every time’, measuring repetition rather than correctness. When the clearance between hinge points changes and the platform posture has a fixed slight deviation, the machine consistently hits the same “wrong” position each time – the repeatability data looks good, but the system deviation is difficult to detect due to stability.

The second combination is “Normal spatial accuracy but inconsistent after needle replacement”. The positioning of the needle changing frame is part of the geometric reference. If the collision moves to the position of the needle changing frame body or platform relative to the needle changing frame, the offset of each needle position to the center of the measuring ball will change, manifested as’ no problem with a single needle, data jumps after needle changing ‘. Judgment method: Measure the same standard ball on two different needle positions and compare them. If the difference exceeds the expected consistency of the needle positions, redo the needle holder positioning.

2.4 Checking the value range of actions

The above table deliberately does not provide specific values: the three models have different working volumes (300 is Ø 300 mm/150 mm, 500 and X-500 are Ø 500 mm/250 mm), and the threshold cannot be used interchangeably; The criteria for repeatability and spatial accuracy are based on the machine’s own baseline before collision.

The official criterion for measuring heads with clear numerical values is only head deflection ≤ 4 μ m. The viewing path is UCCServer Equator → Advanced → Equator Server → Coordinate Measuring Machine Controller → General → Head deflection. If the deflection reading of the probe exceeds this value after collision, first check the probe and probe, and do not immediately suspect the structure.

The remaining values shall be based on official data sheets/manuals, or compared with archived data under the same conditions before the collision. This is the value of regularly archiving measurement data before collision: without it, the criteria for Level IV and V can only rely on guessing.

3. Disposal layer: Six steps and gradual acceptance of geometric benchmark reconstruction

3.1 Overview and Dependency Relationships of Six Steps

The six steps have strict dependencies. If the previous step is not passed, then proceed to the next step, which is equivalent to adding a new correction on the wrong benchmark.

Step	Name	Core Action	What is considered complete	Point when not passed
Step 1	Mechanical reset	Check and release the jamming after power off and standing still, manually (or slowly jog) return the platform to a movable state, confirm that the counterweight mechanism is not jamming, and there are no foreign objects in the support arm and hinge point	The platform can move unobstructed within the three strut strokes, without metal friction or abnormal noise, and the counterweight mechanism moves smoothly	If the jamming cannot be released or there is clearance → Refer to the fourth section for structural component judgment
Step 2	Leveling and foundation	Check if the foundation is loose, if the machine is level, and if there are any signs of tilting or displacement on the body (compare with ground markings)	All foundation forces are evenly distributed, and the horizontal state is consistent with before the collision. The body has no displacement on the ground markings	The foundation cannot be restored or the body has obvious displacement → It needs to be re positioned and done from scratch
Step 3	Needle exchange rack positioning	Follow the official process to perform needle exchange rack positioning (X-500 is Part 1 and Part 2 of the EQR-6 rack); 300/500 According to the corresponding chapter of the random installation guide, reconfirm that all needle position coordinates	Part 1 and Part 2 have been completed without any errors, and that the grasping and putting back actions of each needle position are normal. If there are errors or a certain needle position cannot be recognized, check and replace the relative position between the needle holder body and the platform
Step 4	Calibration	Perform calibration according to the comparison process used (no calibration/CMM calibration/calibration/tolerance calibration), confirm that the correspondence between the `.CAL` file and the nominal value is correct	The calibration process is complete, there are no alarms, and the calibration result falls within the allowable range of the process	Calibration repeatedly fails → Return to steps 1 and 2 to recheck the mechanical and positioning status
Step 5	Mastering	Use certified standard parts to perform re mastering and establish the correlation between the standard parts and the machine coordinate system	Mastering is successfully completed, and measurement is immediately carried out (cannot be measured after interval placement)	Mastering fails or results are abnormal → Check the status and temperature conditions of the standard parts
Step 6	Retest verification	Follow the method in Section 5 to perform single point range, standard part retest, and comparison with the baseline before collision	All three items pass and are consistent with each other	No one passes → Return to the corresponding upstream step and do not directly adjust parameters

This table can be used directly as an on-site inspection checklist. Please proceed step by step below.

3.2 Step 1: Mechanical Reset – First, turn the machine back to “movable”, regardless of accuracy

The first action after collision is not to turn on the power, but to visually and gently check in the power-off state: confirm that the main power is disconnected and the drive is fully discharged; Visually inspect for debris, deformation, and oil stains around the stylus, probe housing, strut, and hinge point area; Confirm that there are no workpieces or fixture residues stuck on the platform; Use extremely light force to push the platform and feel for any hard obstacles.

The goal of this step is very limited: to change the machine from “stuck” to “movable”, without making accuracy judgments, without changing parameters, and without dismantling structural components.

Completion criteria: In both power-off and low-speed jog states, the platform can move smoothly within the stroke of the strut without obstruction, without metal friction sound, without intermittent jamming, and the counterweight balance mechanism operates smoothly.

One detail: The internal structure of the counterweight mechanism has not been officially disclosed. Do not attempt to disassemble or adjust its interior within the scope of self repair. Only observe whether the external movement is smooth and whether there is any jamming or abnormal noise on site, as a basis for determining whether to send it for repair.

If the blockage cannot be resolved by removing the foreign object, or if there is still a noticeable amount of looseness after removal (the platform needs to be lightly pushed for a period of “empty walking” before being subjected to force), do not try pushing again. The tolerance means that there is already a gap at the hinge or connection point, and repeated movement will expand the damage. It should be directly entered into the judgment of the fourth section.

3.3 Step 2: Leveling and Foundation – Find the Physical Basis of the Whole Machine Coordinate System

The Equator 300 has a weight of 25 kg (27 kg for the raised type), the 500 has a weight of 74 kg (75 kg for the raised type), and the X-500 has a weight of 140 kg and a base of 50 kg. This type of mass distribution is sensitive to the in place state: when a 74 kg machine is hit off center by a few millimeters, the four corner feet will change from uniform force to two corner suspension, and a slight tilt will be observed during measurement as the load changes.

Action: Check if all four corner feet are on the ground and if the locking nuts are loose; Review the level according to the official installation guide; Check for displacement or rotation by comparing the contour markings of the aircraft body drawn on the ground; Check for new scratches, misaligned gaskets, and root thread strains.

Completion criteria: uniform force on the foundation, consistent horizontal state with before collision, and no displacement of the aircraft relative to the ground mark.

If the body has been displaced, it must be repositioned according to the installation guide before starting from step one – doing subsequent steps on the displaced body is equivalent to solidifying the erroneous benchmark into the software configuration.

3.4 Step 3: Needle Replacement Rack Positioning – Reconfirm “Can it still be aligned after needle replacement”

The 300 data sheet only states’ comes with a six position needle changing rack ‘, not listed in the 500 data sheet. The X-500 is clearly marked as EQR-6, and the model of 300/500 is subject to the order configuration; But the needle changing rack needs to be repositioned This is common: the official installation guide states that after moving or changing the rack, the needle changing rack positioning process must be carried out. The X-500 corresponds to Part 1 and Part 2300/500 of Locating the EQR-6 rack according to the paragraphs corresponding to the needle changing rack in the random installation guide.

Collisions can also move the needle changing frame body or change the position of the platform relative to the needle changing frame, both of which can cause the needle position coordinates to become invalid.

The order is strictly divided into two parts: Part 1 establishes the preliminary positions of each needle position, and then Part 2 performs precise positioning, not just halfway. After completion, the grasping and releasing responses of each needle position are normal, and there is no phenomenon of “unable to grasp” or “unable to release”.

Completion criteria: Both Part 1 and Part 2 are fully executed without any errors; Perform several consecutive needle changes across different positions, with normal movements and no abnormal prompts.

The positioning of the needle changing rack cannot be replaced by “manually changing a few coordinates”: the needle position coordinates are geometric quantities and not adjustable parameters. The smaller working volume of 300 (Ø 300 mm/150 mm) has a smaller tolerance for needle position deviation, and it is speculated that changing the coordinates will cause errors to accumulate during needle changing.

3.5 Step 4: Calibration – Re establish the relationship between the program and the nominal value

Calibration solves the problem of how the nominal value in the measurement program corresponds to the actual measured quantity of this machine. The official MODUS programming guide lists four comparison processes: no calibration, CMM calibration, calibration, and tolerance calibration, each with different applicable scenarios; In the calibrated product, the .CAL file records the nominal value and CMM measured deviation array.

What needs to be redone after a collision is the calibration corresponding to the current process used in the program, not all processes. If the program includes CMM calibration data, the geometric assumptions on which this data depends have changed and must be redone; If using a process that does not require calibration, this step can be simplified, but it still needs to be confirmed in step six.

The completion criteria are: the selected calibration process is fully executed, there are no alarms, and the results fall within the allowable range of the process; The correspondence between the .CAL file and the nominal value of the current program is correct.

Be careful not to confuse calibration with mastering. Calibration deals with the relationship between the nominal value of the program and the actual measurement of the machine, while mastering deals with the relationship between the machine coordinate system and standard components. After collision, both may fail, but the order is calibration first, then mastering – first establish the corresponding relationship on the program side, and then use standard parts to establish coordinate system anchor points on the machine side.

3.6 Step 5: Mastering – Rebuilding the benchmark using certificate standard parts

Mastering is the foundation of Equator comparison measurement: parts and standard parts are compared in the same machine and coordinate system, and the machine does not need to have absolute measurement accuracy (the 300/500 data sheet does not have an ISO 10360 chapter, only the X-500 has an absolute measurement index of ISO 10360-2:2009). Because it is a comparison, the position of the standard parts must be aligned with the coordinate system anchor points, which is the role of mastering.

The official three prerequisites for the uncertainty of ± 0.002 mm comparison are: the first one is to measure immediately after re mastering. After the collision, this is particularly crucial: if you test again after a while, the environment and structural state may have changed, and the validity of the benchmark cannot be guaranteed.

Action points: Use certified standard parts; Confirm that the standard component itself was not damaged during the collision (if the standard component was also on board during the collision, it may have been hit); Follow the official process to perform re mastering, and once completed, proceed to step six.

Completion criteria: Mastering without errors; The appearance of standard parts is free of scratches, the certificate is within its validity period, and the temperature has stabilized; Measure immediately after mastering is completed, without inserting any other assignments in between.

Temperature conditions: 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 (old version caliber). The following single version documents shall prevail; 500 is+5~+50 ° C. Regardless of the caliber, reconstruction should be done in a stable temperature state to avoid treating thermal deformation as geometric errors.

3.7 Step 6: Retest Verification – Turn ‘Fixed’ into reproducible evidence

After completing the first five steps, the machine “looks” normal. The sixth step is the only link that can prove its true normality, and the method is elaborated in the fifth section. The sequential requirement is that retesting must be conducted in the same environment, at the same temperature, and under the same clamping conditions as the first five steps, and a record must be kept – the reconstruction of missing records cannot determine whether it is “not fixed” or “changed later” when problems occur next time.

Suggest recording the completion time, executor, readings, and screenshots of the six steps into a table and archiving it with the machine. When accuracy issues arise in the future, it is the first evidence to be investigated.

4. Boundary layer: What signs indicate that self repair cannot continue

4.1 Basic division of tripartite boundaries

Regarding the maintenance authority of Equator, it needs to be viewed in three layers, and mixing them together can easily lead to incorrect conclusions.

Official statement for users: The host and controller have no parts that users can repair on their own, and component replacement is carried out by Renishaw business representatives (RBE). It constrains on-site users, not professional maintenance providers.

**Professional maintenance personnel can perform both board level and component level repairs: electrical components such as power modules, driver boards, interface boards, connectors, fans and filtration systems, storage media, and cables can be repaired at the component level by qualified professional maintenance personnel. Before performing such repairs, three boundaries must be explained to the user: non original factory repairs will void the original factory warranty for the component; After repair, accuracy, security, and license availability must be rechecked; In cases involving structural deformation, damage to the grating body, and the need for original factory calibration or encrypted upgrade packages and license migration, the original factory path should still be returned.

The issues that must be stopped from self repair, transferred to professional disposal, or returned to the factory for evaluation are at the level of the measurement framework. This is not a ‘permission’ issue, but a capability and risk * * issue – even if there is repair capability, it should not be treated as a self repair project on site.

4.2 Three types of signs indicating that self-study must be stopped

Category 1: Cracks and deformations in structural components.The base casting, top casting, floating platform, support arm, and drive housing are the solid entities of the geometric framework. Cracks in castings will propagate under alternating loads; Plastic deformation cannot be compensated by software – compensation can only correct repeatable fixed deviations and cannot correct deformation that changes with posture. If visual inspection or light tapping reveals signs of cracks (surface fine lines, oil stains seeping out along the seams, muffled tapping sound), or if the platform and support arm are at an obvious non design angle, the movement must be stopped, the site must be protected, and professional evaluation must be carried out. The risk of continuing operation is crack propagation leading to component fracture and platform drop.

Type 2: Damage to the grating body.The grating is mounted on the strut and is the source of position feedback (the resolution value varies between official documents, please refer to the random document for details). Grating scratches, contamination entering between the reading head and the ruler body, and displacement of the reading head due to collision, are manifested as reading jumps, zero drift, and position deviation alarms during movement. The installation position and gap of the reading head belong to the scope of factory calibration, and disassembling and assembling it by oneself will cause the feedback chain to lose reference. Indication: The reading of a certain strut fluctuates regularly or is not coordinated with the movement of the other two struts.

The third category: Failure of measurement framework.Refers to changes in the constraint points that determine the geometric relationship between the three struts – wear or deformation of the Hooke hinge pin, loosening of the hinge seat, damage to the positioning surface of the parallel constraint mechanism, and deformation of the platform installation surface. It has three characteristics: invisible (passed visual inspection), does not cause an alarm (completely normal in kinematics), can only be indirectly exposed by precision data (can only be detected at levels IV and V). The six step reconstruction cannot ‘fix’ it back – the six step reconstruction is based on existing geometric relationships to reconstruct the baseline, not to correct the geometric relationships themselves. Wear and tear of hinge points, loosening of constraint point seats, and damage to positioning surfaces all belong to this layer, involving structural deformation, scratching of grating bodies, or situations that require original factory calibration, encryption and upgrade packages. They should still be returned to the factory or taken through the Business Representative (RBE) channel; The scope of professional maintenance is electrical board level and component level repairs, and does not extend to the geometric framework itself.

4.3 Three General Stop Loss Rules

The three rules always apply after a collision.

Rule 1: If possible, do not move.Every ‘try it out’ is adding cyclic loads to the damaged hinge points and connections; The most expensive thing at a collision site is not downtime, but turning repairable damage into irreparable damage.

Rule 2: Do not dismantle structural components.The replacement of measuring needles and probe modules is within the scope of maintainability (needle inspection is one of the officially listed self maintainable items), but the disassembly and assembly of struts, hinge points, platforms, counterweight mechanisms, and castings are not included. Disassembly and assembly will change the pre tension and geometric relationship, and the calibration conditions required for restoration are not available on site.

Rule 3: Truthfully record and explain.When sending for repair or returning to the factory for evaluation, the program segment, location, alarm code, and every action taken after the collision must be delivered; Operations such as forcibly resetting to zero and changing parameters should also be included, as concealment can lead to subsequent judgments being based on incorrect assumptions.

4.4 Boundary Reminder for Self Maintainable Items

Avoid reading ‘boundary’ as’ nothing to touch ‘: Cleaning kit (including backup dust filter, prohibited cleaning agent), controller fan filter, and probe inspection are officially listed as self maintainable items. But it is necessary to distinguish between “cleaning and inspection” and “repair” – any actions that involve changing geometric relationships or changing the source of readings after a collision are not within the scope of self maintenance.

5. Verification layer: How to prove that it really measures accurately after repair

5.1 Composition and mutual verification of the three verifications

Verification is not about running a program to see if it looks good, but about combining three sets of independent evidence to explain the problem.

The first item is a single point with 10 extreme ranges.At the fixed position of the fixture board, use the same stylus, direction, and speed to repeatedly trigger the same standard ball 10 times, and calculate the range (maximum value minus minimum value). It answers’ Is consistency still there? ‘. The reading of this item must be compared with the archived data under the same conditions before the collision – only knowing what the current range is, not knowing what the original range was, is equivalent to having no criterion.

The second item is retesting of standard components.Using certified standard parts, retest under the same conditions as before the collision and answer ‘Is the benchmark still available?’. The certificate value is the reference origin of this link; The systematic deviation of the relative certificate value during retesting indicates that the mastering or coordinate system has not been truly reconstructed.

The third item is to compare with the baseline before the collision.Put the first two results in the same table as the pre collision archive data, check the deviation direction and magnitude, and answer ‘Is the change acceptable or structural?’: If the deviation direction is consistent, it usually indicates that the benchmark reconstruction was not done thoroughly; If different positions deviate in different directions, it indicates that the geometric relationship has changed.

The relationship between the three items: the first item is that the measurement chain has random sources (dirt, looseness, gaps); The second item is not enough, while the first item is over → benchmark offset (mastering or calibration problem); The first two items pass while the third item is inconsistent with the baseline → It is necessary to distinguish between environmental differences and geometric changes.

5.2 Acceptance Data Sheet Template

Verification item	Measurement method	Criterion	Record content	Disposal in case of failure
Single point repeatability	fixed position, same standard ball, same stylus, same direction and speed, triggered 10 times	10 range falls within the same condition archive level before collision; The specific measurement values are subject to the official data sheet/manual	10 original readings, range, average, measurement time and temperature	Range amplification → check the deviation of the stylus and probe (≤ 4 μ m), clamping and cleaning status
Standard parts retest	Certified standard parts, measured according to the original process, with the same conditions as before the collision	The difference between the retest result and the certificate value is within the allowable range of the original process	Actual measurement value, certificate value, difference, standard part number and certificate validity period	System offset → Remaking, then measuring, checking whether the clamping position difference is ≤ 1 mm
Comparison with baseline before collision	Measurement program before re run collision using the same program, fixture, and environment	Deviation direction and magnitude are comparable to before collision, with no new systematic trends	Deviation direction, magnitude, and distribution pattern of two data points	Direction inconsistency → Determine changes in geometric relationship and proceed to structural evaluation
Needle consistency	Measure and compare the same standard ball at different needle positions	Ensure consistency between each needle position to meet the needle consistency requirements of the model (based on the official manual)	Read and difference values of each needle position	Difference exceeds the limit → Rewrite the needle changing frame Part 1/Part 2
Head deflection	UCCServer Equator → Advanced → Equator Server → Coordinate Measuring Machine Controller → General → Head deflection	≤ 4 μ m	Reading and measurement time	Exceeding limit → Check the probe and probe module, eliminate the structure, and replace them

The usage of this table is to perform the first validation after collision and retain it; Subsequent routine inspections can only run the first two items with probe deflection; After each collision or relocation, execute the full table again.

5.3 Observation period after recovery

All three verifications have passed, and it is still recommended to keep the observation period. There are three reasons: the damaged hinge point may have gaps after a certain number of cycles; Mastering once does not represent long-term stability; The data after collision will be compared, retested, and even traced back to the shipped parts.

The method is to perform encrypted retesting in the first few batches of normal production (such as running a single point repeatability and standard part retest before starting work every day, with the frequency determined by the production line rhythm), and record the temperature and results; After several consecutive batches are stable and consistent with the baseline, gradually restore the regular frequency. The frequency is a common practice in the industry and is agreed upon by the user and the process department.

A clear judgment should be made during the observation period: if the standard parts are retested and stable, and the single point repeatability is also stable, but the comparison results of the parts are systematically offset from before the collision, the problem is likely not in the machine, but in the part clamping, fixture status, or program. At this point, first check the fixture and program – after collision, people often adjust the fixture or change the clamping position by hand.

6. Supplementary layer: A real disposal path after collision (desensitization)

6.1 Background

An Equator 500 (Ø 500 mm/250 mm) from a certain automotive parts supplier is used for batch comparison and testing of precision machined workpieces. During the night shift, a manual adjustment program was executed without single segment confirmation, and the probe component with TP20 triggered the probe to collide with the unreleased pressure plate on the fixture, causing the controller to alarm and shut down.

The operator cleared the alarm and reset it to zero without recording the scene. The reset was successful, and then continued to attempt normal measurement. The first piece result falls within the tolerance, while the second piece begins to exceed the tolerance, and the direction of the deviation is inconsistent at different clamping positions.

6.2 The process of grading judgment

On site, it was judged in five levels: Level I found slight breakage of the stylus ball head and slight bending of the shaft, while the rest of the appearance was normal; Level II three-axis operation has no abnormal noise but is slightly “dull” compared to normal state; Level III has successfully returned to zero three times in a row, but there is a slight difference between the second stop and the other two; The single point range of level IV is significantly greater than the pre collision archive; The deviation direction of different positions at level V is inconsistent.

Neither level IV nor level V passes, and the direction of level V is inconsistent and points to changes in geometric relationships. The complete six steps must be taken. The first round of replacing the stylus, executing the needle replacement frame Part 1/Part 2, redoing calibration and mastering, and retesting: the single point range returned to near the level before the collision, but there were still small deviations in direction at different positions. Based on this, it is determined that the problem lies in the measurement framework layer, and self-study should be stopped to switch to professional evaluation; Assess and confirm that there is a gap in the hinge point area, and dispose of it according to the original factory path.

After the disposal is completed, execute the six steps again, and all three verifications pass, entering the observation period. During the observation period, daily retesting was conducted, and several batches remained stable, followed by a return to regular frequency.

The three most reusable points in this case 6.3

The first point is that the immediate recording after a collision is more important than any technical action.In this example, the operator first cleared the alarm and returned to zero. During the collision, all clues such as the platform posture and the presence of abnormal sounds were lost. Although the problem was still identified through classification, the path was determined to be much longer.

The second point is that ‘first piece within tolerance’ is not evidence of restoration.The first piece of this example is qualified, which is likely due to its point being located in an area with a small deviation; Systemic offset requires multiple points, locations, and needle positions to be exposed.

The third point is that completing the six steps does not necessarily mean success, failure is also a valid message.After the first round of six steps in this example, there is still a deviation with inconsistent directions, which is precisely the most valuable – it eliminates the problem from “the benchmark has not been reconstructed well” to “the geometric relationship has changed”. If the deviation is’ suppressed ‘at this step parameter, the problem will only be masked.

Frequently Asked Questions (FAQ)

Q1: After the collision, the machine can move normally and return to zero. Do we still need to do geometric reconstruction?

Need. Being able to move and return to zero only indicates that the electrical and motion links are connected, belonging to the first two layers of the four layer structure. Whether the geometric reference layer and the comparison reference layer are still present depends on two levels of checks: single point repeatability and spatial accuracy. Special attention should be paid to the combination of “good repeatability but spatial deviation”: if a collision leaves a fixed small deviation at the hinge point, the machine will steadily hit the same wrong position every time, and the repeatability data is good, but the measurement results have systematic deviations. The judgment method is to compare the saved data before the collision with multiple points and locations, rather than just looking at the result of one point.

Q2: What should be done first after a collision, power on or check?

Turn off the power first for inspection, then turn on the power. The sequence is: confirm that the main power is disconnected, wait for the controller and driver to discharge; Visually inspect for debris, deformation, and oil stains around the stylus, probe housing, strut, and hinge point area; Confirm that there are no workpieces or tool residues stuck on the platform; Then gently test the platform to determine if there is any hard obstruction. After these are completed, power on, clear the alarm, and jog at low speed. Doing the opposite – powering on first, resetting to zero first, and running for a while first – will cause the damaged hinge point to bear another cycle of load, or it may force the stuck mechanism to move and expand the damage.

Q3: Why is the positioning of the needle changing rack ranked before calibration and mastering in the six steps?

Because the reconstruction order of geometric benchmarks is from physics to logic. The positioning of the needle changing rack solves the problem of “the position of each needle in the machine coordinate system”, which is a physical and geometric quantity; The calibration solves the correspondence between the nominal value of the program and the actual measurement of the machine; Mastering addresses the relationship between machine coordinate systems and standard components. If the position of the needle changing rack has not been confirmed yet, calibration and mastering are both done on a platform with unreliable needle position coordinates, and the results will be contaminated by needle changing errors. The official installation guide includes the positioning of the needle changing rack (Part 1/Part 2) as a necessary process after moving or changing the rack, and collisions are classified as events of the same level.

Q4: Is it okay to test again every other day after mastering?

Not recommended. The first prerequisite given by the official for the uncertainty of ± 0.002 mm comparison is’ measure immediately after re mastering ‘. The logic behind this requirement is that mastering establishes anchor points based on the environment and structural state at that time. After a period of time, the environmental temperature, structural thermal state, and even clamping state may change, and the effectiveness of the anchor points cannot be guaranteed. The reconstruction after collision should follow this rule: once mastering is completed, it should enter retesting verification, and no other tasks should be inserted in between. If there must be intervals due to production arrangements, the mastering should be arranged before the start of the measurement, rather than before the shift the day before.

Q5: If a single point passes 10 times with extremely poor results, can production be resumed?

Not enough. The extremely poor qualification only proves consistency, not correctness. It is also necessary to retest the standard parts to confirm that the comparison benchmark is not biased, and to confirm the consistency of needle replacement to ensure that each needle position is aligned. The most easily overlooked item is the “comparison with the baseline before collision”: if the deviation direction is consistent, it means that a benchmark has not been completely reconstructed; If the deviation directions are not consistent at different positions, it indicates that the geometric relationship has changed, and this situation cannot be suppressed by adjusting parameters. Only when all three criteria are met and comparable to the baseline, are the conditions for resuming production met. After recovery, it is recommended to keep an observation period for encrypted retesting.

Q6: If you see oil stains or gaps in the strut or hinge area, can you tighten it yourself?

No. The Hooke hinge, hinge point seat, and positioning surface of the parallel constraint mechanism at both ends of the strut belong to the metrological framework, which determines the geometric relationship between the three struts. All that can be done on site is to observe and record: where the oil stains appear, in which direction the clearance is, and how much empty space there is when lightly pushing the platform. Tightening and adjustment require knowledge of pre tightening requirements and geometric relationships for rechecking, both of which are not available on site. The occurrence of open space itself is a signal that the measurement framework may fail, and the movement should be stopped and transferred to professional evaluation. Repeated attempts will only widen the gap.

Q7: The accuracy has decreased after the collision. Can we first adjust the software parameters to compensate for it?

It is not recommended to adjust parameters before confirming the source of damage. What software compensation can fix is’ repeatable fixed deviation ‘, which is effective if the deviation does indeed come from parameter settings; However, the geometric changes caused by collisions often manifest as “deviations that vary with position and attitude”, which cannot be compensated for. Forceful compensation will make some positions more accurate and others worse. The correct sequence is to first locate the layer according to the five level judgment, then reconstruct the benchmark in six steps, and then look at the direction and shape of the residual deviation. If there are still residues of inconsistent directions after completing the six steps, it is evidence that the geometric relationship has changed and should be transferred to structural evaluation instead of continuing parameter tuning.

Q8: How should I choose between sending the crashed plane for professional maintenance or returning to the factory?

See which layer the damage falls on. When it falls on the electrical layer (controller, driver, interface, cable, fan filter), board level and component level repairs can be carried out by qualified professional maintenance providers, provided that the user is informed that non original factory repairs will void the original factory warranty for the component, require re verification after repair, and still follow the original factory path if it involves original calibration or permission migration. If it falls on the measurement framework layer (casting cracks and deformation, grating body damage, hinge point and constraint point gap), the original factory path should be the main approach, as this layer requires the re establishment of geometric relationships and factory calibration conditions. The judgment is based on the residual deviation after the five level inspection and six step reconstruction, not on the size of the collision sound.

Data source explanation

The technical parameters of 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 Comparison Instrument User Guide H-5504-8500, Equator Installation and Controller Guide (ID 132816/107657), Equator Controller Technical Instructions H-5504-4000-01-A, Software Suite 2. x Guide (ID 134769).

The working temperature of 300 is different in caliber between English 06-A and Chinese 04-A, and has been listed in parallel; The difference in raster resolution between official documents is not quoted value by value, and the main text is treated as’ subject to random files’; The needle changing rack is not named EQR-6 in the 300/500 data sheet, and it has been indicated that the order configuration shall prevail. The five level damage grading, stop loss rules, and observation period frequency are industry practices and are not official specifications. Please distinguish them from official data sheets when quoting; The case data in the article 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.