9.2 Fixed Gages, Optical Comparators, and CMM Systems

9.2 Fixed Gages, Optical Comparators, and CMM Systems

Fixed Limit Gaging and Taylor's Principle of Gaging

While variable hand measuring tools (micrometers, calipers, indicators) provide continuous numerical readings, high-volume manufacturing often requires rapid, binary screening to determine whether a part lies within allowable dimensional limits. This inspection methodology utilizes fixed limit gages—also designated as Go / No-Go gages.

Fixed limit gages do not measure an exact numerical dimension; rather, they establish whether the part's boundaries conform to the engineering tolerance boundaries known as the Maximum Material Condition (MMC) and the Least Material Condition (LMC):

• Maximum Material Condition (MMC): The state of a manufactured part where it contains the maximum amount of material within the stated tolerance limits (e.g., maximum shaft diameter or minimum hole diameter).
• Least Material Condition (LMC): The state of a manufactured part where it contains the minimum amount of material within the stated limits (e.g., minimum shaft diameter or maximum hole diameter).

Taylor's Principle of Gaging (William Taylor, 1905)

The design and application of fixed limit gages is governed worldwide by Taylor's Principle of Gaging (standardized in ASME B89.7.3.1 and ISO 1502):

[!IMPORTANT] Taylor's Principle of Gaging:

1. The GO Gage: Must be designed to check the feature at its Maximum Material Condition (MMC), and it must check all interrelated dimensions and geometric form simultaneously (size, roundness, straightness, and perpendicularity) over the full length of engagement.
2. The NO-GO Gage: Must be designed to check the feature at its Least Material Condition (LMC), and it must check only one dimension at a time (isolated point or two-point contact) to detect localized out-of-roundness or taper.

TAYLOR'S PRINCIPLE ILLUSTRATION (OVAL / OUT-OF-ROUND HOLE):

       DEFECTIVE OVAL HOLE: Major axis exceeds USL, but Minor axis is undersize!
       
                Minor Axis (Undersize)
                      |
                 +----+----+
                /     |     \
   Major Axis  |      +      |  Major Axis (Oversize: Exceeds LMC)
   (Oversize)   \     |     /
                 +----+----+
                      |
                      
   INCORRECT FULL-FORM NO-GO PLUG GAGE:    CORRECT TWO-POINT NO-GO PIN GAGE:
   - Cylindrical plug binds on minor axis!  - Checks major axis independently!
   - Fails to enter!                       - Enters major axis!
   - FALSE PASS: Accepts defective part!   - TRUE REJECT: Correctly flags defect!

The Operational Danger of Violating Taylor's Principle

Consider an out-of-round (oval) hole where the major diameter exceeds the specification limit (past LMC), but the minor diameter is undersize. If an inspector uses a full cylindrical NO-GO plug gage, the gage will bind across the narrow minor axis and fail to enter the hole. The inspector would mistakenly conclude that the hole is acceptable, when in reality the major axis is severely oversized! By designing the NO-GO gage with two-point pin contact, the gage checks individual radial orientations, immediately detecting the oversized major axis and rejecting the nonconforming part.

---

Plug Gages, Ring Gages, Snap Gages, and Thread Gages

Cylindrical Plug Gages (Internal Features)

Cylindrical plug gages inspect internal bores and drilled, reamed, or ground holes:

• The GO Plug Member: Ground to the minimum hole diameter (MMC). A conforming hole must allow the GO plug to slide smoothly through the entire depth of the bore under its own weight or light finger pressure without binding.
• The NO-GO Plug Member: Ground to the maximum hole diameter (LMC). A conforming hole must not allow the NO-GO plug to enter the hole (or enter no more than the chamfer depth, typically $< 1$ to 2 turns).
• Physical Identification: The GO member is distinctly longer in length than the NO-GO member. The NO-GO member is also marked with a red identification groove around its aluminum handle.

Cylindrical Ring Gages and Snap Gages (External Features)

• Ring Gages: Cylindrical rings used to inspect external cylindrical shafts. The GO ring checks maximum shaft OD (MMC), while the NO-GO ring checks minimum shaft OD (LMC). The NO-GO ring features an annular groove knurled into its outer circumference for instant tactile identification.
• Snap Gages (Caliper Gages): C-shaped rigid frames carrying two pairs of precision-ground carbide anvils in a single progressive tool. The front anvil pair is set to the GO dimension (MMC), and the rear anvil pair is set to the NO-GO dimension (LMC). In production inspection, the operator passes the shaft through the front jaws: it must pass through the GO jaws, but stop against the NO-GO jaws.

Thread Gages: Functional Pitch Diameter Verification

Thread gaging checks the complex geometry of screw threads, including pitch diameter, lead, flank angle, and major/minor diameters:

• Thread Plug Gages: Check internal tapped threads. The GO thread plug features a full thread form with complete flank engagement over the full thread length. The NO-GO thread plug features truncated crests and wide root clearances, isolating and checking pitch diameter alone.
• Thread Ring Gages: Used in matched pairs (GO and NO-GO) to inspect external screw threads.

Gage Maker's Tolerance Classes (ANSI/ASME B89.1.5)

Fixed limit gages are manufactured to standardized precision classes based on nominal size:

| Gage Maker's Class | Relative Precision Level | Typical Application | |---|---|---|| | Class XX | Highest precision ($0.000020"$ / $0.5,\mu\text{m}$) | Reference master standards for setting optical comparators, bore gages, and CMMs | | Class X | High precision ($0.000040"$ / $1.0,\mu\text{m}$) | Close-tolerance inspection gages and calibration masters | | Class Y | Medium precision ($0.000070"$ / $1.8,\mu\text{m}$) | Standard quality control laboratory working gages | | Class Z | Commercial precision ($0.000100"$ / $2.5,\mu\text{m}$) | Production shop-floor inspection gages with broad tolerances | | Class ZZ | Utility grade ($0.000200"$ / $5.0,\mu\text{m}$) | Rough machining checks and toolroom layout |

[!NOTE] Gage Wear Allowance: Because the GO member enters every acceptable part, it experiences continuous frictional abrasive wear. Standard practice applies a small manufacturing wear allowance to the GO gage (making it slightly larger for plugs or slightly smaller for rings), allowing the gage to wear toward the blueprint limit throughout its service life without violating part tolerances.

---

Optical Comparators (Profile Projectors)

An optical comparator (or profile projector) is a non-contact optical inspection system that projects a magnified two-dimensional shadow silhouette or surface reflection of a workpiece onto a ground-glass viewing screen.

OPTICAL COMPARATOR OPTICAL PATH:

   [ Profile Light Source ] 
              |
              v (Collimated Parallel Light Beam)
         [ Workpiece ]  <-- Mounted on X-Y Micrometer Stage
              |
              v (Transmitted Shadow Silhouette)
    [ Telecentric Projection Lens ] (10x, 20x, 50x, 100x)
              |
              v
       [ Turning Mirror ] 
              |
              v
   [ Ground-Glass Viewing Screen with Crosshairs / CAD Mylar Overlay ]

Illumination Modes

Optical comparators feature two distinct illumination systems:

1. Profile (Diascopic / Transmitted) Illumination:
   • The light source is located behind or below the part staging table, directing collimated parallel light rays across the workpiece edge into the projection lens.
   • Produces an ultra-sharp, high-contrast black silhouette (shadow) against a bright green or white background.
   • Primary Application: External profiles, outside diameters, radii, chamfers, punch contours, screw thread profiles, and gear tooth forms.
2. Surface (Episcopic / Reflected) Illumination:
   • The light source directs intense illumination through half-mirrors directly onto the front face of the workpiece.
   • The reflected light passes back through the lens system to form a full-color or grayscale image of surface features.
   • Primary Application: Blind cavities, internal step depths, stamped alphanumeric date codes, surface engravings, printed circuit board traces, and blind holes.

Telecentric Projection Lenses

Standard photographic lenses create perspective distortion: features closer to the lens appear larger than features further away. Optical comparators utilize telecentric optics, where the entrance pupil is located at infinity. This ensures that collimated light rays strike the part parallel to the optical axis, meaning magnification remains completely constant even if the part is slightly out of focus or positioned at varying depths along the stage.

• Standard interchangeable lens magnifications include $10\times$, $20\times$, $50\times$, and $100\times$.
• If a stamping has a burr of $0.002"$, viewed under $50\times$ magnification it appears on the screen as $0.002" \times 50 = 0.100"$, making microscopic defects instantly measurable with a standard rule!

Staging, Readouts, and Overlays

• X-Y Precision Stage: Mounted on crossed roller bearings, driven by lead screws equipped with linear optical glass scales. The stage connects to a Digital Readout (DRO) that computes geometric constructions (circle diameters from 3 points, intersection points, skew alignments, and center-to-center distances).
• Chart Overlays: Clear Mylar or glass screens placed over the comparator face. Common overlays include precision radius grids, concentric circle charts, thread pitch profiles, and $10\times/20\times$ custom master CAD engineering drawings for direct contour overlay comparison.

Advantages of Optical Metrology

• Zero Measuring Force: Ideal for flexible, compliant, or soft materials (elastomeric O-rings, rubber seals, micro-springs, thin metal foils) that would deform elastically beneath the touch of a caliper or micrometer spindle.
• Speed: Simultaneously verifies multiple complex geometric features (radii, angles, intersections) in a single visual inspection.

---

Coordinate Measuring Machines (CMM)

A Coordinate Measuring Machine (CMM) is an advanced 3D electromechanical measuring system that senses discrete spatial coordinates on physical part surfaces using a contact or non-contact probe moving along three mutually orthogonal Cartesian axes ($X, Y, Z$).

Structural Architectures of CMMs

Probe Technologies

1. Touch-Trigger Probes (Kinematic Probes):
   • Internal construction features a three-point kinematic seating mechanism (spring-loaded triangular array of ball and cylinder electrical contacts).
   • When the ruby stylus tip contacts the workpiece from any direction, the stylus pivots, breaking electrical contact across one of the seats. This instantaneous resistance change sends a trigger pulse to the controller, latching the optical scale positions along $X, Y, Z$ within microseconds.
   • Extremely repeatable ($< 0.5,\mu\text{m}$), highly durable.
2. Continuous Contact Scanning Probes:
   • Rather than retracting after each point, the probe stylus maintains continuous sliding contact with the part surface, traveling along complex contours.
   • Internal inductive (LVDT) or optical deflection sensors continuously monitor probe tip deflection, capturing thousands of spatial coordinates per second.
   • Essential For: Form evaluation, including circularity (roundness), cylindricity, straightness, and complex aerospace airfoil profile inspections.
3. Optical & Laser Line Triangulation Probes:
   • Non-contact laser scanners capture dense 3D "point clouds" (millions of spatial points per second) for CAD surface deviation heat-mapping and reverse engineering.

Part Coordinate System (PCS) Setup & The 3-2-1 Alignment Rule

A physical part clamped to a CMM granite table is randomly oriented relative to the machine's internal Machine Coordinate System ($MCS$). To inspect the part against its engineering CAD model, the technician must establish a Part Coordinate System (PCS) based on blueprint datums.

Every rigid unconstrained body possesses six spatial degrees of freedom (DOF):

• Three Translational Movements: $T_x, T_y, T_z$ (movement along the $X, Y, Z$ axes).
• Three Rotational Movements: $R_x, R_y, R_z$ (rotation about the $X, Y, Z$ axes: pitch, roll, and yaw).

The 3-2-1 Datum Alignment Rule mathematically constrains all six degrees of freedom:

THE 3-2-1 DATUM ALIGNMENT RULE:

   1. PRIMARY DATUM (Planar Leveling: 3 Points)
      - Constrains 3 DOF: 2 Rotations (Pitch, Roll) + 1 Translation (Tz)
      - Defines primary reference plane (Z = 0)
         [ Point 1 ]       [ Point 2 ]
                  \       /
                   [ Point 3 ]
                   
   2. SECONDARY DATUM (Axis Orientation: 2 Points)
      - Constrains 2 DOF: 1 Rotation (Yaw) + 1 Translation (Ty)
      - Defines secondary reference axis (X-axis alignment)
         [ Point 4 ] ------------ [ Point 5 ]
         
   3. TERTIARY DATUM (Origin Location: 1 Point)
      - Constrains 1 DOF: 1 Translation (Tx)
      - Establishes coordinate origin (0, 0, 0)
         [ Point 6 ]
         
   TOTAL CONSTRAINED: 3 + 2 + 1 = 6 DEGREES OF FREEDOM (Completely Fixed!)

1. Primary Datum (Leveling Plane - 3 Non-Collinear Points):
   • Probing three points on the primary datum surface defines a flat plane.
   • Constrains 3 degrees of freedom: two rotations ($R_x, R_y$) to level the plane perpendicular to the vertical axis, and one translation ($T_z$) establishing the vertical datum level ($Z = 0$).
2. Secondary Datum (Orientation Line - 2 Points):
   • Probing two points along a perpendicular edge or feature defines a straight line in the leveled plane.
   • Constrains 2 degrees of freedom: one rotation ($R_z$, yaw) to align the part axis parallel to the CMM coordinate axis, and one translation ($T_y$) establishing the $Y$ datum position ($Y = 0$).
3. Tertiary Datum (Origin Point - 1 Point):
   • Probing a single point against an orthogonal stop or hole center establishes the final datum plane.
   • Constrains the final 1 degree of freedom: translation along the remaining axis ($T_x$), establishing the coordinate origin: $(X = 0, Y = 0, Z = 0)$.

Vision Systems / Video Measuring Machines

Video CMMs replace the contact stylus with a high-resolution CCD/CMOS digital camera and motorized zoom telecentric lens. Programmable top-light LED rings, coaxial through-the-lens lighting, and bottom backlights illuminate the workpiece. Proprietary edge-detection software detects sub-pixel intensity transitions, capturing hundreds of dimensional points simultaneously. Video systems excel in micro-electronics, semiconductor packaging, medical catheter inspection, and stampings where physical contact would distort the component.

---

Surface Roughness Measurement and Profilometry

Dimensional inspection verifies macro-geometry (size, position, orientation). However, functional performance—such as lubricating oil retention in engine cylinders, seal leakage prevention, bearing fatigue life, and cosmetic paint appearance—depends directly upon micro-geometry: surface texture.

SURFACE TEXTURE DECOMPOSITION:

  Combined Profile:    /\  /\  /\  /\  /\  /\  /\  /\  /\  /\  /\  /\
                      /  \/  \/  \/  \/  \/  \/  \/  \/  \/  \/  \/  \

  1. Roughness:       /\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\  (Tool marks, grit: High freq)
  
  2. Waviness:        /¯¯¯¯¯¯¯¯¯¯\          /¯¯¯¯¯¯¯¯¯¯\          (Machine vibration, chatter)
                                  \________/            \________
                                  
  3. Form Error:      /-------------------------------------------\  (Straightness, flatness error)

Surface Texture Terminology

Per ASME B46.1 and ISO 4287, surface texture encompasses three overlapping components:

1. Roughness: The finest, high-frequency micro-irregularities produced by the cutting tool edge, feed marks, or abrasive grit tears.
2. Waviness: Intermediate cyclic irregularities with wider spacing, caused by machine tool vibration, spindle chatter, deflections, or heat-treat distortion.
3. Form Error: Macro-level departures from the intended geometric nominal shape (out-of-straightness, barrel taper, flatness warpage).

Contact Stylus Profilometer Operation

A profilometer glides a sensitive diamond stylus across the workpiece surface at a constant velocity:

• Stylus Tip: Precision cone with a spherical diamond tip radius of $2,\mu\text{m}$ (or $5,\mu\text{m}$) and an included cone angle of $60^\circ$ or $90^\circ$.
• Stylus Force: Ultra-light tracking force (typically $0.75\text{ mN}$, or approx. $0.00017\text{ lbf}$) to prevent the diamond from plowing or scratching soft aluminum or polymer surfaces.
• Skidded versus Skidless Tracers:
  • Skidded Tracer: The pick-up arm rests on a rounded metal skid that rides directly on the workpiece adjacent to the stylus. The skid acts as a physical mechanical high-pass filter: it establishes a local reference datum. Skidded gages can measure roughness ($R_a$), but cannot measure waviness because the skid rides up and down the waviness waves.
  • Skidless (Free) Tracer: The stylus is referenced to an internal, optically flat precision datum inside the drive unit. Skidless units measure both roughness and waviness simultaneously.

Cutoff Length ($\lambda_c$) and Evaluation Length

To isolate micro-roughness from macro-waviness, the profilometer processes the raw electronic signal through a digital high-pass filter characterized by its cutoff length ($\lambda_c$):

• Cutoff Length ($\lambda_c$): The spatial wavelength filter threshold. Surface wavelengths shorter than $\lambda_c$ are categorized as roughness; wavelengths longer than $\lambda_c$ are filtered out and categorized as waviness.
• Standard Cutoffs (ASME B46.1): $0.08\text{ mm}$, $0.25\text{ mm}$, $0.80\text{ mm}$ ($0.030"$), $2.5\text{ mm}$, and $8.0\text{ mm}$. $0.80\text{ mm}$ is the default industrial cutoff unless specified otherwise.
• Evaluation Length ($L_n$): The total distance over which the parameters are calculated, conventionally consisting of five consecutive cutoff lengths: $L_n = 5 \times \lambda_c$ For a standard $0.80\text{ mm}$ cutoff, the evaluation length is $5 \times 0.80\text{ mm} = 4.0\text{ mm}$.

Primary Surface Roughness Parameters

1. Roughness Average ($R_a$)

$R_a$ (historically known as Arithmetic Average (AA) or Center Line Average (CLA)) is the universally specified surface roughness parameter worldwide. It represents the arithmetic average of the absolute vertical departures of the roughness profile from the mean center line over the evaluation length: $R_a = \frac{1}{L} \int_0^L |z(x)| dx \approx \frac{1}{N} \sum_{i=1}^N |z_i|$

• Units: Micrometers ($\mu\text{m}$) or micro-inches ($\mu\text{in}$). $\mathbf{1\,\mu\text{m} \approx 39.37\,\mu\text{in}} \quad (\text{commonly approximated as } 40\,\mu\text{in})$
• Inherent Limitation of $R_a$: Because it averages absolute deviations, $R_a$ cannot differentiate between peaks and valleys. A surface with tall, sharp peaks has the exact same $R_a$ value as a surface with flat plateaus and deep lubrication pits, even though their functional tribological behaviors are completely opposite!

2. Average Maximum Peak-to-Valley Height ($R_z$)

$R_z$ measures the vertical distance between the highest peak and the deepest valley within each individual sampling length (cutoff), averaged across the five sampling lengths of the evaluation length: $R_z = \frac{1}{5} \sum_{i=1}^5 (R_{pi} - R_{vi})$

• Highly sensitive to individual deep scratches, porosity voids, or tool drag marks that would be completely buried and smoothed over by the averaging calculation of $R_a$.

3. Other Common Parameters

• $R_q$ (Root Mean Square Roughness, RMS): Standard statistical deviation of profile heights, $R_q = \sqrt{\frac{1}{L} \int_0^L z^2(x)dx}$. Typically $10%$ to $25%$ higher than $R_a$ for machined surfaces ($R_q \approx 1.11 R_a$).
• $R_t$ (Total Peak-to-Valley Height): The absolute vertical distance between the single highest peak and single deepest valley across the entire evaluation length.

---

Step-by-Step Worked Numerical Examples

Worked Example 1: Taylor's Principle Limit Gage Sizing

Scenario: An internal cylindrical bore is dimensioned on an engineering drawing as $\varnothing 1.250" \pm 0.003"$. Determine the nominal dimensions for the GO plug gage member and the NO-GO plug gage member, and specify how each member must engage the bore per Taylor's Principle.

Step 1: Calculate Specification Limits

$\text{Upper Specification Limit (USL)} = 1.250" + 0.003" = 1.253"$ $\text{Lower Specification Limit (LSL)} = 1.250" - 0.003" = 1.247"$

Step 2: Determine Maximum and Least Material Conditions for a Hole

• For an internal bore (hole), maximum material occurs when the hole is smallest: $\mathbf{MMC = 1.247"}$.
• Least material occurs when the hole is largest: $\mathbf{LMC = 1.253"}$.

Step 3: Size the Plug Gage Members

• GO Plug Gage: Must check MMC $\implies \mathbf{1.247"}$. Must be a full cylindrical plug checking diameter, roundness, and straightness simultaneously over the full bore depth.
• NO-GO Plug Gage: Must check LMC $\implies \mathbf{1.253"}$. Should ideally feature two-point pin contacts, checking one diameter at a time across multiple angular orientations.

---

Worked Example 2: Roughness Unit Conversion and Profilometer Evaluation

Scenario: A hydraulic cylinder bore has an engineering callout of $R_a \le 0.40,\mu\text{m}$. A quality technician tests the bore with an imperial profilometer, which reports an average roughness of $R_a = 18.5,\mu\text{in}$. Does the cylinder meet the specification?

Step 1: Convert Imperial Micro-inches to Metric Micrometers

Using the conversion factor $1,\mu\text{m} = 39.3701,\mu\text{in}$: $R_a (\mu\text{m}) = \frac{18.5\,\mu\text{in}}{39.3701\,\mu\text{in}/\mu\text{m}} \approx 0.4699\,\mu\text{m}$

Step 2: Compare Against Engineering Specification

$\text{Observed } R_a = 0.47\,\mu\text{m} > \text{Specification } 0.40\,\mu\text{m}$

Conclusion

The cylinder fails acceptance criteria. Even though $18.5,\mu\text{in}$ appears numerically small, it equates to $0.47,\mu\text{m}$, breaching the maximum allowable limit of $0.40,\mu\text{m}$ ($15.75,\mu\text{in}$). The technician must flag the lot for honing tool re-dressing.

---

Technician Inspection Scenarios & Common Exam Traps

Real-World Shop Scenario: Optical Comparator Thread Inspection

A technician inspects a high-strength aircraft fastener on an optical comparator using profile illumination. The technician aligns the thread chart overlay to the shadow silhouette, but notes that the thread flanks appear blurry and out-of-focus on one side. The technician mistakenly assumes the comparator lens is defective.

• Root Cause Analysis: Fastener screw threads feature a helix angle (lead angle). When parallel collimated profile light strikes the screw straight-on ($90^\circ$ to the axis), the light beam is clipped by the angled thread helix, casting a distorted, blurry shadow. To project a true, razor-sharp thread cross-section, the technician must helix the comparator workstage (swivel the stage table angularly by the exact thread lead angle, typically $2^\circ$ to $5^\circ$), aligning the optical axis parallel with the thread flanks.

Common Exam Traps for CQT Candidates

• Exam Trap 1: MMC versus LMC for Holes versus Shafts: For a hole, MMC is the smallest size (least metal removed). For a shaft, MMC is the largest size (most metal retained). Reversing these conditions is the single most common fixed gaging error on the exam.
• Exam Trap 2: Full-Form NO-GO Gages: Remember Taylor's rule: NO-GO gages must never check full form! A NO-GO plug must check isolated two-point diametral width to catch out-of-round lobes.
• Exam Trap 3: The 3-2-1 Primary Plane Constraint: A common exam question asks: "How many degrees of freedom does the primary datum plane constrain in a 3-2-1 CMM alignment?" Distractors often state 1 (just the Z axis) or 6. The correct answer is 3 degrees of freedom (two rotational tilts and one linear translation).
• Exam Trap 4: Profilometer Cutoff Misconception: Increasing the cutoff length $\lambda_c$ (e.g., from $0.25\text{ mm}$ to $2.5\text{ mm}$) allows wider surface wavelengths into the calculation, which almost always increases the reported $R_a$ value. If an exam question asks how to separate roughness from waviness, the answer is the cutoff length filter ($\lambda_c$).