# Metrology > Our thinking on the people, technologies and future of Metrology Public Ghost content for AI and LLM tooling. This file includes a bounded export of public pages first, then recent public posts. Append `.md` to any post or page URL to get the content in Markdown (for example, `/example-post.md`). ## Pages ### Special Applications URL: https://www.metrology.com/specials/ Last updated: 2026-08-09T01:13:04.000Z When your application isn't standard we design to your drawing. Every special is designed from your part or feature, spans 2–500 mm (0.079"–19.69"), repeats to 0.002 mm (0.0001"), and comes back from the design office as a proposed solution within 24 hours. #### Internal Threads A thread's pitch diameter hides between the crests, where no standard bore gauge can reach. Our thread heads carry profiled contacts that seat on the flank, reading pitch, major or minor diameter directly — 8–200 mm (0.315"–7.87"). Read our [Three Minute Guide to Thread Measurement ](https://www.metrology.com/metrology-guide-to-thready-meaujdsjdnd/)or contact our design office and we will get back with a design and solution in 24 hours. #### Spherical Bores A spherical bore is widest where a conventional gauge can't reach — past its own mouth. Our heads collapse to enter, then expand to read the true sphere diameter — 8–200 mm (0.315"–7.87"). Read our [Three Minute Guide to Spherical Measurement](https://www.metrology.com/spherical-bores-the-three-minute-metrology-guide/) or contact our design office and we will get back with a design and solution in 24 hours. #### Ball Screws A ball screw is measured where the balls actually run — on the track, not the crest. Our two- and three-ball contact heads seat in the groove exactly as the working balls do, built from your profile, pitch and ball diameter. Read our [Three Minute Guide to Ball Screw Measurement](https://www.metrology.com/ball-screws-the-3-minute-metrology-guide/) or contact our design office and we will get back with a design and solution in 24 hours. #### Internal Slots Internal slots defeat round gauges — the feature is wider than its opening. Our heads enter through the narrow opening and gauge true slot width, plain, recessed or special-profile. Read our [Three Minute Guide to Slot Measurement](https://www.metrology.com/slot-measurement-the-three-minute-metrology-guide/) or contact our design office and we will get back with a design and solution in 24 hours. #### Deep Bores Honed cylinders, barrels and metre-deep housings are beyond any handheld gauge. Extensions carry the head to the measurement point, transducer near the anvils to cancel temperature error — proven past 4 m (13 ft), 50–300 mm (1.97"–11.81"). Read our Metrology Guide to Deep Bore Measurement or contact our design office and we will get back with a design and solution in 24 hours. #### Internal Tapers A taper has a different diameter at every depth, so where you measure matters as much as what. Our heads gauge diameter at a defined datum, with a clamp-ring method to verify the angle itself. Read our [Three Minute Guide to Taper Measurement](https://www.metrology.com/taper-measurement-the-three-minute-metrology-guide/) or contact our design office and we will get back with a design and solution in 24 hours. #### O Ring Grooves A groove diameter is always larger than the bore hiding it. Our heads pass through the entry bore, then extend contacts into the groove to read its true diameter — 8–200 mm (0.315"–7.87"). Read our [Three Minute Guide to O-Ring Groove Measurement ](https://www.metrology.com/metrology-guide-to-o-ring-measurement/)or contact our design office and we will get back with a design and solution in 24 hours. #### Spline Bores A spline bore has no single diameter — pitch, major and minor each need their own contact geometry. Our heads match your tooth form, odd or even counts, reading PCD, major or minor directly. Read our [Three Minute Guide to Spline Measurement](https://www.metrology.com/metrology-guide-to-spline-measurement/) or contact our design office and we will get back with a design and solution in 24 hours. ### Standard Ranges URL: https://www.metrology.com/standards/ Last updated: 2026-08-09T01:12:52.000Z Twenty-four interchangeable heads. Two handles. Every bore from 6 to 200 mm (0.236"–7.87"). The Metrology three-point system self-centres in the bore, reads on the indicator of your choice, and repeats to 0.002 mm (0.0001") in any operator's hands — no special skill required. Choose the pistol-grip handle or the lever-action for vertical work; both accept the entire head range. Preset against a calibrated setting ring, measuring in seconds. Boxed sets ship complete: heads, handle, extensions, rings. 20 minutes of training to get consistent and repeatable results. #### 2 Styles of Horizontal And Pistol Grip Handles Two handles cover the entire system: the Trigabore pistol-grip and the Truebore lever-action for vertical gauging. One spans 6–16 mm (0.236"–0.63"), the second 16–200 mm (0.63"–7.87") — two handles, and you never need another. Both accept every head in the range, with single-handed trigger operation collapsing all three anvils for entry. The handle is insulated against body heat, so the reading is the bore's, not your hand's. #### 24 Standard Fixed Anvil Heads For 6mm - 200mm Twenty-four standard heads cover 6–200 mm (0.236"–7.87"), each spanning 2–10 mm (0.079"–0.394") of range. Every head repeats to 0.002 mm (0.0001"), with accuracy of 0.003 mm (0.00012") up to 32 mm and 0.004–0.005 mm above. Fixed anvils eliminate operator error; carbide contacts are standard above 10 mm. All heads interchange between both handles. #### 2 Micron Repeatability And 3 Micron Accurcy Repeatability is 0.002 mm (0.0001") on every standard head, in any operator's hands — self-centring three-point geometry does the work, not skill. Head accuracy runs 0.003 mm (0.00012") to 32 mm and 0.004–0.005 mm above; residual gauges resolve to 0.0001 mm. Truecal verification checks linearity at any point across the range, not a single diameter. Every instrument ships with calibration traceable to national standards. #### Wide Range of Digital Readouts And Transducers Any presettable indicator with an 8 mm or 3/8" stem fits including Metrology, Mitutoyo and Sylvac units as standard. Minimum resolution 0.001 mm (0.00005"), with 0.0001 mm gauges available that virtually eliminate digitising error. Typical indicator accuracy is 3–5 µm; bench transducers achieve ±0.001 mm (0.00004"). Bluetooth and cable output feed readings to PCs for data logging and SPC. #### Retrofit To Any Existing Heads Or Gauges Every head — standard or special — fits the Mitutoyo Borematic handle directly, and the Mitutoyo 468 readout accepts our heads across 16–200 mm. Adaptors suit effectively any other make of readout, so your existing instruments gain full special-application capability. Standard heads carry a 10 mm maximum range on retrofits, with setting rings supplied where needed. You keep the handle you own and extend what it measures. #### Pre Calibrated Setting Master Rings Every head presets against a calibrated master ring, so each reading is a deviation from a certified reference. Standard rings run 8–190 mm (0.315"–7.48"), one ring typically covering several heads. Rings above 100 mm (3.94") ship in individual boxes; smaller sizes are included in the boxed set. The Truecal system can replace large ring inventories — one unit calibrates any head from 16 to 200 mm using slip gauges. #### Modular Extensions For Deep Bores Standard 100 mm (3.94") and 150 mm (5.91") extensions join in series for deeper measuring points; we advise achievable depths per diameter. For serious depth, the pneumatic system runs 1 m (39.4") extensions — twenty joined on one application, proven past 4 m (13 ft) — at ±0.005 mm (0.0002") across 50–300 mm (1.97"–11.81"). All extensions interchange between any handle, head and digital readout. #### Calibration and Traceability Every instrument ships with a certificate traceable to national standards; UKAS-accredited (UK) and US-accredited certificates are available where your quality system requires them. Heads are verified against calibrated master rings which checks linearity across the full range. Recalibration is a return service, certified against the same serial for unbroken traceability. Rings and masters certify on the same cycle, so your kit renews together. #### 10 Year Manufacturing Defect Warranty Every standard instrument carries a 10-year swap-out warranty for defects and manufacturing faults — so a fault means a replacement, not a repair queue. Loan instruments are provided during repair or recalibration, so a gauge off the floor never means a measurement you can't take. Design engineers, not a call centre, handle support — typically within 24 hours. Instruments from thirty years ago are still in service and still supported. ### Metrology Governance URL: https://www.metrology.com/metrology-governance/ Last updated: 2026-07-21T07:35:42.000Z Increasingly operations run on an estate of thousands of sensors feeding software, robots and remote operators — decisions once made by a person looking at a gauge are now made by systems trusting one. Metrology Governance is the independent review of the instrument estate, the legal and contractual measurement obligations it must satisfy, and the evidence chain that drives corporate and regulator reporting and risk management #### Sensor Estate Quality-room gauges are registered; the thousands of sensors feeding control systems and dashboards usually are not. You cannot govern what you have never counted. The register comes first. #### Obligations Register Measurement duties are scattered across statutes, licences, reporting regimes and contracts — no single function sees them all. We compile them into one register, so "what must we measure, how well, and says who?" has a single answer. #### Evidence Chain A measurement you cannot evidence is an opinion. Every number should trace through unbroken, in-date, accredited calibration certificates to a national standard. We report where the chain holds and where it breaks. #### Automation Risk When a person read the gauge, the person was the anomaly detector. Software optimises on whatever it's fed, at machine speed. We assess where automated decisions now rest on instruments nobody verifies #### Board Attestation Directors sign statements resting on measured data, and courts expect boards to oversee mission-critical systems. Our quarterly attestation pack puts a defensible monitoring record on file — before the question arrives. ### We're building the operating system for dimensional metrology URL: https://www.metrology.com/home/ Last updated: 2026-08-09T00:53:01.000Z --- ### We draw on fifty years of craft in custom instrument making for aerospace, defence, automotive, medical, luxury, metrology and precision engineering. --- ### We design and build single instruments for small machining shops to integrated multi-year, multi-domain quality systems for global supply chains. --- ### We advise boards, management teams and financial sponsors on governance, strategy and corporate development across products and borders. --- ### We solve your specific problem from your drawing or description and our design engineers in our UK, US or Asia offices reply with a solution in 24 hours. ### Three Minute Metrology Guide URL: https://www.metrology.com/three-minute-metrology-guides/ Last updated: 2026-07-16T18:08:31.000Z News about metrology ### We design and build custom solutions for your specific problem URL: https://www.metrology.com/design/ Last updated: 2026-08-09T01:43:45.000Z --- ### We design instruments for your feature, to your tolerance and your access — self-centring where it matters, calibrated to a reference, repeatable in any operator's hands. --- ### We ship from stock when we can. When we can't we work from your drawing and our design engineers reply with a proposed solution within 24 hours. --- ### We provide laboratory certification and a 10 year defect warranty as standard. Accurate, traceable, and repeatable. ### Contact us URL: https://www.metrology.com/contact-us/ Last updated: 2026-07-18T07:34:04.000Z ## **Our Teams** --- **Technical Suppor**t Support@metrology.com --- **Design Office** Design@metrology.com --- **Everything else** Team@metrology.com --- ## **Our Offices** --- **Europe** Metrology Holdings Ltd, 128 City Road, London, EC2V 1NX **Asia Pacific** Metrology PTY LTD, 56 Campbell Street, Sydney, 2000 **Americas** Metrology LLC, 47 St Columbus Street, Delaware, 46700 ### Metrology Deals URL: https://www.metrology.com/metrology-deals/ Last updated: 2026-07-21T07:39:21.000Z Precision measurement is one of the last genuinely fragmented industrial sectors — hundreds of founder-owned instrument makers and calibration labs, each with a narrow moat, an ageing owner and no obvious successor. Capital has noticed: the roll-ups are underway and the multiples are moving. Metrology Deals is transaction support from operators inside the sector — the diligence, strategy and integration judgement that generalist advisers cannot read off a data room. #### Commercial Due Diligence A niche instrument business is either a moat or a museum, and the accounts rarely say which. We test what the data room cannot — the defensibility of the technology, the health of the installed base, the recurring calibration and service revenue, and whether the customer relationships survive the founder. Buy the moat; leave the museum. #### M&A Strategy In a fragmented sector, the multiple is made on the way in. We build the acquisition thesis — which niches consolidate, which capabilities compound, what the platform looks like at exit — so each deal is a step in a strategy rather than an opportunity that happened to ring. Serial acquirers are priced differently from serial buyers. #### Target Identification The best targets in this sector are not for sale; they have retirement dates. We map the founder-owned instrument makers, calibration labs and service businesses that fit your thesis, and open the conversations before a process starts — because the deals that create value are the ones that never reach an auction. #### Post Merger Support Deals are priced on synergies and lost in the first hundred days. In measurement businesses the value walks out the door as people — accreditations, inside knowledge, the founder's customer relationships. We plan and support the integration so capability, certification and customers survive the change of ownership the model assumed they would. ### Metrology Operations URL: https://www.metrology.com/metrology-operations-2/ Last updated: 2026-07-21T07:37:46.000Z Every operation runs on measurement — parts are accepted, processes adjusted, invoices raised and audits passed on the strength of numbers an instrument produced. Yet measurement is bought like consumables, scheduled like maintenance and staffed like overhead. Metrology Operations is the practical review of how a business specifies, buys, proves and defends its measurements — built for the management team that has to answer for them. #### Pre Audit Review The worst time to discover a measurement gap is when the auditor does. We run the instrument estate, calibration records and traceability evidence against the standard you will be audited to — ISO 9001, AS9100, IATF 16949 or your customer's own flow-downs — and close the findings before they become findings. #### Procurement Strategy Most instruments are bought on price and specified by habit. We work back from the tolerance to the measurement, and from the measurement to the instrument — so capital goes into resolution where the drawing demands it, not into precision nobody will ever use. The cheapest gauge is the one that never buys you a recall. #### Measurement Capability A tolerance your gauge cannot resolve is a guess with a certificate. We run gauge R&R across the measurements that accept or reject product, and report where the measurement system is consuming your tolerance band — before scrap, concessions and customer returns report it for you. #### Metrology Business Case Measurement investment competes for capital against machines that visibly make parts, and loses — because its returns sit hidden in scrap rates, concessions, disputed invoices and audits narrowly survived. Metrology Business Cases is the development of an investment case senior management can approve: the cost of poor measurement quantified, the options priced, and an independent view of a clear RoI. ## Posts ### The Metrology Calibrator - The Three Minute Metrology Guide URL: https://www.metrology.com/the-metrology-calibrator-the-three-minute-metrology-guide/ Last updated: 2026-08-23T01:36:34.000Z ## The issue A three-point bore gauge measures relative to zero, and that zero has to be set before every job against a known diameter. The standard method is a ring gauge machined to the exact nominal size being measured. ## The problems with the usual answers Buying a full set of rings scales badly. A shop covering a 6–100 mm range with class XX rings needs a separate ring for every nominal size in that span, and the bill for heads and rings together lands in five figures once seven or more sizes are covered. Ring tolerance is also tighter in name than in practice: many suppliers only certify a ring to ±1.5 µm of its labelled size, which erodes the working margin on any job specified below that figure. Rings drift as well as cost. Each one needs periodic recertification, and a shop holding seven or more rings is running seven or more separate calibration cycles, each with its own paperwork and its own downtime while the ring is away from the bench. ## The Calibrator advantage The Metrology Calibrator replaces the ring gauge with a single bench unit. Its 60° reference is not a fixed diameter but a tapered seat: three flat faces, each set at 60° to its neighbours, running the length of the unit. Because the faces angle inward along that length, the diameter presented to a three-point head changes continuously with how far the head is inserted — shallow positions present a larger diameter, deeper positions a smaller one. Setting a target diameter is a positioning problem rather than a matching problem. The slip block method fixes insertion depth by stacking grade 0 gauge blocks to a calculated height beneath the head, using the known 60° angle to convert block height into diameter. The transducer method reads insertion depth directly and performs the same conversion electronically, skipping the block build-up. Because the taper is continuous rather than fixed at one size, a single unit functions as an infinite ring: any diameter within its range can be set at any point along the taper, not just the handful of nominals a physical ring set happens to cover. - One unit covers the full 5–100 mm range, in place of a dedicated ring for every nominal size in that span - Diameter is continuously variable along the taper, so odd or in-between sizes are set as easily as round numbers - Removes the ongoing cost of storing and recertifying a ring for every size on the job list - Slip block and transducer methods both read off the same 60° reference, so either can be used depending on what is on hand ## The virtual infinite ring A physical ring gauge checks a three-point head at exactly one diameter. A shop holding one ring per nominal size gets one confirmed point on the range, and everything between those points goes unverified. The Calibrator's continuous taper removes that limit. Because any diameter along the 5–100 mm range can be generated by repositioning the anvil, the same unit effectively holds an unlimited number of ring diameters rather than a fixed set — a virtual infinite ring, generated on demand rather than machined and stored ahead of time. This matters because instrument error is not always uniform across a range. A head can read accurately at its nominal bottom point and drift out of tolerance further along its travel, and a single ring check has no way to catch that. Checking at several points along the taper — for example every 1 mm, using the 1.5 mm-of-slip-block-per-1 mm-of-diameter relationship — builds a full linearity picture of the instrument rather than a confirmation at one size, catching errors that a ring-based check would miss entirely. ## How to use Seat the head in the Calibrator's 60° reference and pick a method. For the slip block method, build the target diameter from a gauge block stack, bring the three contact points onto the stack through the 60° geometry, and zero the gauge indicator at that reading. For the transducer method, set the integrated transducer to the target diameter directly and zero the gauge against it, skipping the block build-up step entirely. Either route sets a three-point head to any diameter in the 5–100 mm range from one unit, in place of sourcing, storing, and recertifying a ring gauge for every size on the job list. --- # Operations Manual ## System overview The Calibrator system replaces the ring gauge with a bench unit for calibrating three-point bore gauges and micrometers with contact points at 120°. It comprises the Calibrator base unit and a certified 60° setting master, plus a ring gauge or cylindrical block used to set the head's nominal bottom point before calibration begins. The unit works on a 60° included-angle principle. Two fixed angle blocks, side A and side B, form a 60° V that a sliding anvil moves along. Moving the anvil changes the diameter presented to the three-point head, and the relationship between axial movement and diameter change is fixed: D2 − D1 = L / 1.5, where D1 and D2 are the diameters generated at two points along the V and L is the axial distance between them. ![](https://storage.ghost.io/c/97/2a/972a49a2-bf56-480d-bfc6-da726d4924be/content/images/2026/08/MET-GDE-010-truecal-geometry.svg) ## Equipment required - The Metrology base unit, with fixed angle blocks and a sliding anvil - A matched set of slip gauges (calibrated for the slip gauge method, nominal value only for the transducer method) - A ring gauge or cylindrical block matching the nominal bottom point of the head's range - A micrometer or dial gauge, or a transducer and digital readout for the transducer method ![](https://storage.ghost.io/c/97/2a/972a49a2-bf56-480d-bfc6-da726d4924be/content/images/2026/08/ChatGPT-Image-Aug-6--2026--12_00_19-PM.png) Typical slip gauge method calibration ## Slip gauge method The example below covers a 50–60 mm head; the same steps apply across the range, with slip block sizes selected to suit. 1. Set the bore gauge to the nominal bottom point of its range, using a 50 mm ring gauge or a 50 mm cylindrical block. 2. Set the micrometer to nominal zero using a 20 mm slip block against sides A and B, with the sliding anvil E in position. 3. Locate the three-point gauge on the calibrator between sides A and B. Clamp the slip block in position with knob D, then move anvil E until it generates the target diameter. 4. Search for the smallest reading, adjusting the position of anvil E. Finding this reliably takes practice — allow up to 15 minutes when new to the system. 5. Clamp the slider firmly at the nominal lowest point and zero the micrometer. 6. Repeat the measurement to confirm the instrument reads zero within its repeatability (0.002 mm overall, 0.001 mm resolution). Reset to zero if necessary. 7. For each subsequent reading, reduce the slip block size and reclamp. A reduction in slip block size of 1.5 mm produces a 1 mm increase in the generated diameter, so a 15 mm reduction generates a 10 mm increase. 8. Record each point in a deviation table, working through the block sizes for the range being checked — for example, steps of 20, 17, 14, 11, 8, and 5 mm. | Slip block (mm) | Calibrated setting R2 (mm) | Micrometer reading R1 (mm) | Deviation R2 − R1 | | --------------- | -------------------------- | -------------------------- | ----------------- | | 20 | 0.000 | 0.000 | ±0.000 | | 17 | 2.000 | | | | 14 | 4.000 | | | | 11 | 6.000 | | | | 8 | 8.000 | | | | 5 | 10.000 | | | ## Transducer method The transducer method uses an electronic length standard in place of the slip gauges, which simplifies the check at each point. ![](https://storage.ghost.io/c/97/2a/972a49a2-bf56-480d-bfc6-da726d4924be/content/images/2026/08/ChatGPT-Image-Aug-6--2026--12_00_45-PM.png) Typical transducer method calibration 1. Fit a transducer with a range at least 1.5 times the head's range — a 10 mm head needs a 15 mm transducer, a 25 mm head needs 38 mm. 2. Slip gauges in this method are used at nominal value only, since their role is to locate each check position rather than act as the length standard. 3. At each position, take the readout reading Ro and divide by 1.5 to get R2, the reading at that calibration point. 4. Deviation is R2 − R1, where R1 is the micrometer reading at the same position. 5. Reduce the slip block size and repeat at each subsequent point, following the same block sizes as the slip gauge method. | Slip block (mm) | Readout Ro (mm) | R2 = Ro / 1.5 (mm) | Micrometer R1 (mm) | Deviation R2 − R1 | | --------------- | --------------- | ------------------ | ------------------ | ----------------- | | 20 | 0.000 | 0.000 | 0.000 | ±0.000 | | 17 | | | | | | 14 | | | | | | 11 | | | | | | 8 | | | | | | 5 | | | | | ## Calibrating the Calibrator The unit's own accuracy depends on the 60° setting master and the alignment of blocks A and B. Both need periodic checking rather than a one-off setup at build time. The certified setting master must be within its stated tolerance, with the 60° angle central to its two locating pins. To align the Calibrator to it, clamp angle blocks A and B against a matched set of four slip gauges using the clamp screw, then release and press the blocks against the calibrated master before reclamping — all four matched slip blocks must show contact for the alignment to be valid. Correctly set, blocks A and B align to the setting master within ±00° 00′ 05″. Squareness of the base to A, B, and the slider is held to 0.004 mm, and straightness of the angle block faces is held to 0.002 mm over any 25 mm length. The angle of sides A and B is calibrated to 60° ± 00° 00′ 40″, which works out to a systematic error of ±0.0028 mm over a 25 mm range, or ±0.0014 mm over 12.5 mm — this can be calculated and applied as a correction where required. ## Matching and beating ring-gauge accuracy The Calibrator's systematic error is not random scatter — it comes from a fixed angle tolerance on sides A and B, so it scales directly with how far the anvil travels from the last zero point. That gives two ways to bring it below the tolerance of the best setting rings, which run ±1.5 µm for 6–40 mm rings and ±2 µm for 40–110 mm rings. The simplest approach is to zero closer to the size being checked, rather than zeroing once and sliding a long way up the taper to reach the target diameter. Halving the travel from 25 mm to 12.5 mm halves the error from ±2.8 µm to ±1.4 µm, which sits inside the best ring tolerance rather than outside it. | Zero-to-check distance | Systematic error | Compares to best rings | | ---------------------- | ---------------- | ---------------------- | | 25 mm | ±2.8 µm | Wider than ±1.5–2 µm | | 12.5 mm | ±1.4 µm | At or inside ±1.5 µm | The second approach is to apply a correction rather than change technique. Because the error is calculable at any point on the range, it can be subtracted from the reading at the point of use. This removes the angle tolerance from the result entirely, leaving accuracy set by the unit's repeatability of 0.002 mm overall and 0.001 mm resolution, rather than by the ±00° 00′ 40″ angle tolerance on sides A and B. A third route changes the hardware rather than the procedure: tightening the angle tolerance itself. Roughly halving it, to around ±00° 00′ 20″, would bring the 25 mm-range error down to about ±1.4 µm without needing shorter zeroing steps or a correction step at all. ### From demo to factory - scaling robot production and the actuator bore gauging gap URL: https://www.metrology.com/from-demo-to-factory-scaling-robot-production-and-the-actuator-bore-gauging-gap/ Last updated: 2026-08-03T04:56:02.000Z ## The Demo-to-Factory Trap A robotic actuator is a joint module: a frameless torque motor, a precision reducer, an output bearing, two encoders and drive electronics, packed into one housing. Dimensionally it is a stack of concentric bores that have to agree with each other, since the motor sits in one, the reducer sits in another and the output bearing sits in a third. A humanoid carries 28 or more joints, so a $20,000 platform is running an $11,000 actuator budget at roughly $400 per joint, and actuators account for approximately 56% of the bill of materials. The tolerances on those bores are single-digit microns in places, because reduction ratios between 30:1 and 50:1 leave the designer almost no envelope to work in. Disagreement between the bores shows up as scrap, rework and unexplained runout on the largest line in the cost model. At twenty demo robots that is 560 actuators to get right, and at five hundred a month it is 14,000. > The punchline. Your ability to scale and meet customer and investor expectations means you need to get really good at measuring 4 concentric diameters to plus or minus a few microns. Most humanoid and legged programmes design the actuator and assemble it in-house, buying the reducer and motor as components, since catalogue joint modules arrive 30% to 50% heavier than the joint tolerates. First units get built on a bench, with bearings selected from a batch and shims cut until preload and runout come in. That is correct at twenty units, where formalising inspection against an unfrozen design wastes money on features that will change. It stops working at five hundred a month, because hand selection now needs 14,000 conforming parts and without an inspection gate the tail ships by default. Teams usually read the result as a firmware problem. Unit-to-unit variation in friction and torque ripple pushes control engineers into compensating in software, and the compensation only holds for the batch it was tuned on. A bore rejected at goods-inward costs the price of the part, while the same deviation found at end-of-line test costs the assembly plus teardown labour, a 10x to 100x multiple. Feedback time sets how many units are affected, since a CMM queue running hours to days runs while the machine keeps cutting. The punchline. Your ability to scale and meet customer and investor expectations means you need to get really good at measuring 4 concentric diameters to plus or minus a few microns. --- ## The supplier position makes things harder for start ups and scale ups Supply economics makes all of it harder. Harmonic Drive Systems held around 85% of the harmonic reducer market in 2023 and Nabtesco claims roughly 60% of medium-to-large joint reducers, and outside the vertically integrated Chinese programmes almost nobody makes their own strain-wave units. There is no practical second source, and 20 to 30 week lead times turn a rejected batch into a big issue. Rejecting a shipment carries real commercial cost, since the supplier has other customers and a small buyer has limited ability to escalate. Finding a problem at week 4 leaves the options to rework, resequence or requalify open, while finding the same problem at week 24 usually means a missed milestone. Further, as more "standard" assemblies appear the market moves toward integrated joint modules combining reducer, frameless motor, encoders and thermal sensors in one black box housing. That leaves the mounting register, the output flange pilot, the hollow shaft ID and your own housing bores. Most stack-up failures occur at that interface, and responsibility for them is regularly disputed between supplier and buyer. --- ## The Bores That Decide It The working list follows, with each feature described by the failure it produces and the gauge configuration that reads it. The first group covers the motor side, which is common to every joint module. The second covers the reducer, where the features depend on whether the design uses a strain-wave, planetary or RV unit. Three-point head sizes across our range run 2 mm to 40 mm, with anvil geometry configured to the feature rather than to the nearest standard size. ![](https://storage.ghost.io/c/97/2a/972a49a2-bf56-480d-bfc6-da726d4924be/content/images/2026/08/actuator-bore-gauging-callouts--1-.svg) ### 1) Motor side ### 1.1) Stator housing bore Robot joint motors run air gaps of 0.3 to 0.5 mm because torque density scales inversely with gap. A housing bore 20 µm out of round consumes 4% to 7% of the design gap on one side, producing unbalanced magnetic pull that appears as cogging torque, bearing side-load, acoustic noise and efficiency loss. Ovality does more damage than size error, since a bore uniformly 15 µm oversize shifts the gap while a bore 15 µm out of round modulates it once per revolution. **Gauge configuration.** Standard three-point head with extended reach, taking readings at both ends plus the midpoint to separate taper from ovality. In most designs this is machined in-house, so it is the first place to put a gauge at the machine. ### 1.2) Lamination stack internal diameter The shell bore and the stack bore are different features, and it is the stack ID that sets the air gap. A stack measured on the bench will read differently once it is pressed or bonded into the shell, because the laminations relax against the housing and take up its form. Measuring only the empty shell gives a number that is correct about the housing and wrong about the motor. **Gauge configuration.** Extended-reach head used after insertion rather than before, on 100% of units. This is one of the few features where the measurement has to happen at a specific point in the assembly sequence rather than wherever it is convenient. ### 1.3) Rotor press-fit ID and magnet ring bore The rotor ID that presses onto the shaft or the wave generator hub carries two failure modes at once. Ovality produces an unbalanced air gap on the running side and a dynamic imbalance that shows up as vibration at speed, and on bonded magnet rings the adhesive cure moves the geometry after the part was last measured. **Gauge configuration.** Standard head sized to the fit, measured before and after bonding. The second reading is the one that matters and the one most often skipped. ### 1.4) Encoder hub bore The encoder hub bore is usually 8 mm to 20 mm and its eccentricity converts directly into angular position error. A 10 µm offset on a 20 mm hub is arcminutes at the output, which the control loop cannot distinguish from real joint motion and will attempt to correct. **Gauge configuration.** Small-bore head with a dedicated setting ring. The tolerance is tight relative to the diameter, so the setting ring should not be shared with other features in the same size range. ### 1.5) Rear cover bearing seat and spigot register The rear cover carries the non-drive-end bearing seat plus the spigot that locates the cover to the shell, typically H6 and H7 respectively. Concentricity between the two sets rotor alignment at the rear, and error here loads the front bearing in a way that reads as a bearing quality problem. **Gauge configuration.** Blind-bore head, since the seat is usually closed at one end, measured on the same setting ring as the mating shell feature. ### 1.6) Hollow shaft internal diameter Cable routing through the joint means a deep through-bore, often at a length-to-diameter ratio above four. Two-point measurement is slow and unreliable in this geometry, because the operator has to rock the instrument searching for a maximum reading in a bore they cannot see into. **Gauge configuration.** Deep-bore head on an extension, with the display at the operator's end while the contacts work at depth. Three-point measurement matters most here, since the operator has no line of sight to the feature being measured. ### 2) Reducer side ### 2.1) Reducer mounting register and output flange pilot These are the interface features between the bought-in reducer and your housing, and they are the ones you can verify on receipt. Concentricity between the register and the output flange sets how much of the reducer's rated accuracy survives installation, and a 20 µm error here is indistinguishable in the field from a reducer that was out of specification when it shipped. **Gauge configuration.** Standard head matched to the register diameter, run against the same setting ring used on the mating housing bore so both halves of the fit sit on one datum. Measuring the two features on different instruments introduces a bias that is usually mistaken for an assembly fault. ### 2.2) Crossed roller bearing seat The output bearing seat is typically specified H6, which at Ø62 mm is a band of +0 to +19 µm. The seat provides both the interference fit and the geometric reference for the entire output flange, so ovality transfers almost directly into flange runout and is then multiplied by link length. On a 1.5 m arm, 30 µm of shoulder bore error becomes visible positional error at the wrist. **Gauge configuration.** Standard head sized to the seat, with shortened anvils where the seat is a narrow land rather than a full-length bore. On narrow seats a catalogue instrument will often contact the chamfer or relief groove and return a reading taken off the wrong surface. ### 2.3) Planet carrier bores Relevant where the joint uses a planetary first stage rather than a strain-wave unit throughout. A planet carrier holds each planet gear on a pin located in its own bore, and a 10 µm size spread across three of those bores routes a disproportionate share of torque through one planet. The fatigue failure that follows occurs at a load the design nominally survives, and in the field it is usually recorded as a design deficiency when the cause is dimensional. Bore-to-bore pitch matters equally and is a position measurement, so it stays on a CMM or a fixture while the gauge covers size and roundness. **Gauge configuration.** Blind-bore head, since carrier bores are frequently stepped or closed at one end. Blind-bore anvils measure to within a fraction of a millimetre of the bore floor, which is where the bearing sits and where the load is carried. ### 2.4) Cycloid disc and pin bores Relevant to RV and cycloidal designs rather than strain-wave. A mid-size RV unit carries 30 to 40 needle roller pin bores at 8 to 20 mm diameter, generally to IT5, a band of about 8 to 9 µm at those sizes. With the eccentric cam bores and the crankshaft bores through the carrier, one reducer presents more than forty critical internal diameters. Inspection volume per part is the practical constraint, since a feature count that high puts CMM inspection out of economic reach at rate. **Gauge configuration.** Small-bore heads in the 8 mm to 20 mm range, set up as a bench station with the setting ring alongside so an operator works through forty features without re-zeroing between each. Self-centring geometry supports that throughput, since there is no rocking or searching per feature. ### 2.5) Wave generator bearing seat The tightest feature in most actuators, frequently held to a total band of around 5 µm. The bearing is deliberately deformed in service, so the bore locating it has no margin left for manufacturing error and whatever geometry the bore contributes adds to the deformation the design already imposes. Loss of ratcheting margin presents as sudden torque loss under peak load rather than gradual degradation. **Gauge configuration.** Small-diameter head with carbide contacts and a dedicated setting ring at the tightest available resolution. This is measured during reducer manufacture rather than at goods-inward, since the feature is inaccessible once assembled. Buyers rely on supplier records for it, which is why the interface features above carry so much of the verification load. ### 2.6) O-ring grooves and internal recesses Sealed joints carry internal grooves whose diameter and width control seal compression, and a groove 30 µm out on diameter shifts squeeze enough to change seal life materially. These features are invisible to any instrument that cannot pass the entry diameter and then expand into the groove. **Gauge configuration.** Custom groove anvils designed to your groove specification. The parameters are the measured diameter, the entry diameter, the groove width, the depth to the groove, the clearance depth below it and the bottom clearance diameter, with the design check being that measured diameter minus entry diameter falls within head range. --- ## Why Two-Point Measurement Will Not Get You There A two-point instrument cannot detect odd-lobed form error. A shape with an odd lobe count holds constant diameter in every direction while being demonstrably not round, so the gauge reads it as conforming and the part passes inspection. Three-lobed form is the characteristic signature of three-jaw chucking and collet distortion, which covers the most common workholding methods in precision turning. A part clamped in a three-jaw chuck, bored, then released relaxes into a three-lobed shape whose diameter reads perfectly and whose bearing seat still produces runout. Honing tends to produce five-lobed and seven-lobed patterns for the same reason of periodic support, so the finishing operation does not remove the problem. Lobed form is the failure that carries straight through the transition to volume. At prototype scale a lobed bearing seat gets absorbed by selection and shimming; at volume it produces a population of joints with runout that the inspection records cannot account for, since those records show every bore in tolerance. Three contacts at 120° register the lobed condition rather than averaging through it. Self-centring geometry adds a second benefit, in that the instrument finds true diameter without rocking or searching. Against a 5 µm band, operator technique is frequently the largest single contributor to measurement uncertainty. --- ## Building the Regime: A 6 Step Practical Sequence The steps below run in order and are achievable inside a quarter. Total cost is less than a single scrapped batch of reducers. **Step one — write the critical bore list.** Take the joint module drawing and mark every internal diameter, then sort them by whether they gate function or only fit. Function-gating features get 100% inspection and everything else goes to sampling once capability is shown, and for most joint modules the first group runs to four or five features rather than forty. **Step two — select instruments against gauge R&R, not against diameter.** Convention holds that measurement system variation should consume no more than 10% of the tolerance band, with 10% to 30% acceptable under justification. Against a 5 µm band that means a system reproducible to 0.5 µm across operators and shifts, which most general-purpose instruments do not achieve in a working shop, so run the R&R study before you commit rather than after a customer rejection. **Step three — put the gauge where the metal is cut.** A CMM produces data on a schedule set by the queue in front of it, which in most shops runs from hours to days, while the machine keeps cutting. Gauging at the machine turns a process drift from a scrap event into a tool-offset adjustment. Where machining is outsourced, this splits in two: 100% inspection of critical features at goods-inward, plus an at-machine gauging requirement written into the contract shop's purchase order with the instrument and setting ring specified. A supplier measuring on your datum removes the bias that otherwise presents as an untraceable assembly fault. **Step four — set the sampling plan in writing and tie it to capability.** Start at 100% on the critical list and reduce only when Cpk supports it, with the rule written down rather than held by whoever runs the cell. Where the rule is not written down, sampling frequency tends to drift downward under schedule pressure with no record of when it changed. **Step five — establish calibration and recall.** Setting rings drift and instruments get dropped, so a gauge outside its verification interval carries no evidential weight whatever the display says. Quarterly verification against a traceable standard is a defensible default for actuator work, tightening to monthly where the gauge is in continuous production use, and the evidence chain matters as much as the calibration itself, since the first question in a supplier dispute is whether the buyer's measurement can be defended. **Step six — enforce thermal discipline.** Steel expands roughly 11.7 µm per metre per °C, so a Ø60 mm bore shifts about 0.7 µm per °C. A 7 °C difference between the gauge, the setting ring and the part consumes the entire tolerance band on the tightest features, which makes soaking parts and instruments together a hard requirement rather than good practice. --- ## Why the Tolerances Are This Tight The bands above are set by the design constraint rather than by convention. The governing metric for a humanoid joint is specific torque in newton-metres per kilogram, and it dominates because the actuator carries itself, unlike an industrial arm bolted to a floor where extra shoulder mass costs nothing beyond a larger base casting. Mass added at the elbow must be carried by the shoulder, then the torso, then the hip and the knee, so a 500-gram penalty at the wrist propagates into several kilograms of structure and battery by the time it reaches the ankle. Tesla, Figure, Apptronik, Agility, Unitree and Boston Dynamics all design their own actuators rather than buying modules for this reason. Chasing specific torque forces high reduction ratios, rising to 160:1 where static holding torque dominates. Each step up shrinks the envelope the designer has to work in, and the bore tolerances compress with it. High ratios also make the joint mechanically opaque, so the motor cannot sense external forces through the friction and inertia of the gear train. The design adds output-side torque sensing plus a software layer to synthesise the compliance direct-drive joints get for free, and once the joint depends on strain-gauge output rather than motor current, mechanical variation in the bore locating that element becomes control-loop error. --- ## The Supply Side, and Why It Gets Harder Before It Gets Easier McKinsey's humanoid supply-chain analysis ((as an example) identifies harmonic drives as the clearest constraint in the actuator stack, on the basis that production is precision-bound and capital-intensive, requiring dedicated tooling, long qualification cycles and the metrology infrastructure to support both. Reducer capacity cannot be commissioned in months the way electronics capacity can. Demand arithmetic explains the current capacity build-out. A single humanoid takes 20 to 40 harmonic reducers against six to twelve actuators for a conventional industrial arm, so one humanoid consumes the reducer content of three to five industrial robots. The International Federation of Robotics recorded more than 590,000 industrial robot installations globally in 2024, up 12.4%, while Omdia has humanoid shipments rising almost 480% in 2025 to 13,318 units with a projection of 2.6 million by 2035. > A buyer with a documented inspection regime and defensible measurement records is easier to deal with than one rejecting batches on a functional test failure it cannot localise. That projection implies between 52 million and 104 million precision reducers a year against current global capability measured in low single-digit millions. Capacity is being added, with Zhejiang Shuanghuan bringing on a Suzhou plant in 2026 described at 500,000 units annually and Leaderdrive growing 2025 revenue 47% to 570.7 million yuan, but the gap will not close within the timescale of a near-term production ramp. For a startup scaling now, lead times are likely to lengthen before they shorten, and allocation depends partly on how a supplier rates you as a customer. A buyer with a documented inspection regime and defensible measurement records is easier to deal with than one rejecting batches on a functional test failure it cannot localise. --- ## What It Costs The full regime for a joint module programme is a small number of gauges, a set of setting rings, a written plan and a calibration interval. Set against a single scrapped batch of reducers on a 20 to 30 week lead time, or a quarter of control engineering spent compensating for mechanical variation, the payback period is short. Teams that get through the ramp generally treat measurement as production infrastructure rather than as a quality department function. Write the critical bore list before you need it, buy instruments capable against the tightest band rather than the most common diameter, and put them where the parts are made. --- ## About metrology.com metrology.com designs and manufactures custom three-point internal bore gauges with standard heads from 2 mm to 40 mm and bespoke anvil geometry for features catalogue instruments cannot reach. The design office works from customer drawings, covering blind bores, recessed seats, O-ring grooves, deep hollow shafts and non-standard sizes, and returns a gauge specified against the actual feature. If you are scaling actuator production and have not yet written down which bores are critical, start with the list above. Send us the joint module drawing and the design office will come back within 24 hours with the features that warrant measurement at the machine and the head configuration that reaches them. ### The Metrology Governance gap cost Boeing $237.5 Million. A guide for directors and counsel. URL: https://www.metrology.com/metrology-governance-gap/ Last updated: 2026-07-29T14:59:45.000Z --- In 2012, a director who raised cybersecurity at a board meeting would have been thanked and redirected to the IT function. It was a technical matter. It had an owner and a budget line, and no natural place on a board agenda that was already full. People had enough to do as it was. And it wasn't a board issue anyway. Then Target lost 40 million card records to attackers who entered through a refrigeration contractor's credentials. The CEO "left". Directors were sued. Then the Equifax breach followed, then SolarWinds, where to make things more interesting the SEC pursued the company's own security officer personally. That tends to concentrate the mind. And a whole new world of risk (and director liability) came into view. Within a decade the category had a mandatory disclosure regime, a standing place on audit committee agendas, and a global spend that grew from roughly $23 billion in 2003 to about $272 billion today. What changed was not the technology. The threat had been there the whole time. What changed was a series of expensive and very visible failures that made it clear this was a board governance issue, not a management operational one. Metrology data is at the point cybersecurity occupied around 2012\. It sits with the quality function. It has an owner and a budget line, and no natural place on the agenda. In our view, and from the work we do, three forces are going to change that, and very quickly. --- ## First, the data already exists, at a huge scale most boards have never seen quantified A modern manufacturing plant is saturated with measurement. Inline gauging on the line. Coordinate measuring machines in the metrology lab. Vision systems on the inspection station. Condition monitoring on the bearings. Torque tools that record every fastener. A single certified facility runs to tens of thousands of instruments generating millions of data points per shift, and multiples of this a day. And that's just the sensor estate inside the factory. Every one of those readings is a (discoverable) record. Every record has a calibration certificate behind it and a tolerance band around it, governed by a retention policy that decides how long it survives. Every one of those measurements is linked to an instrument. Every instrument has a calibration and repair policy. And every instrument could potentially have many operators. In a company with ISO 9001 certification, the whole apparatus is documented by design. > The company holds vastly more evidence about its own operations than at any point in industrial history, and the board's view of that evidence has not materially changed since the 1980s. In our experience boards almost never see any of it. What reaches the board pack is a quality slide: cost of poor quality, customer satisfaction, audit findings raised against findings closed. These are lagging proxies, three or four numbers standing in for millions. The annual surveillance audit that certifies the system samples a fraction of one percent of the underlying records, and it was designed to satisfy an ISO registrar rather than a director. The result is an ignored asymmetry. The company holds vastly more evidence about its own operations than at any point in industrial history, and the board's view of that evidence has not materially changed since the 1980s. --- ## Second, sensors and robotics are about to multiply the volume The measurement estate is not stable. It is compounding. And in new ways. The International Federation of Robotics counted 4,664,000 industrial robots in operational use worldwide at the end of 2024, up 9% in a single year. Annual growth rates are increasing rapidly. Robots do not simply replace human labour. They instrument it. A robotic cell that used to be a manual station now generates torque curves, position feedback, cycle-time data, force readings and vision-system pass/fail records for every part that passes through it. Work that produced no data at all five years ago now produces a continuous stream. > Cyber became a board issue because the attack surface expanded faster than the ability to govern it. Measurement data is on the same curve, driven by the same underlying force: more connected devices, generating more records, faster than the governance around them adapts. Layer onto that the broader industrial Internet of Things build-out and the shift to automated inspection, alongside a growing use of machine learning to act on measurement data without a human in the loop. The trajectory is not ambiguous. A board that finds its current measurement estate difficult to oversee is looking at a problem that gets substantially harder every year, not easier. This is the part that makes the cybersecurity parallel more than an analogy. Cyber became a board issue partly because the attack surface expanded faster than anyone's ability to govern it. Measurement data is on the same curve, driven by the same underlying force: more connected devices, generating more records, faster than the governance around them adapts. We see this every day and the pace and complexity of change is increasing. --- ## Third, the liability has already started to land The third force is the one that will actually put this on the agenda, because it is the one that turns an operational matter into a personal one. The pattern in recent enforcement and litigation is consistent enough to be uncomfortable. A company held measurement data that identified a problem. The data did not reach anyone with the authority and the obligation to act on it. The failure surfaced later, at scale, in front of a regulator or a court. ![](https://storage.ghost.io/c/97/2a/972a49a2-bf56-480d-bfc6-da726d4924be/content/images/2026/07/exhibit1_what_happened_1200-1.jpg) **Exhibit 1 — What happened.* Eleven oversight failures, ranked by cost, 2023–2026\. Sources: US DOJ, EPA, NTSB, NHTSA, CPSC, CPUC, FAA, Delaware Court of Chancery, SEC filings and company disclosures.* The costs arrive through multiple channels. Regulatory penalties from bodies such as the EPA, the CPUC, the CPSC and NHTSA. Criminal resolutions with the Department of Justice. Civil verdicts brought by private plaintiffs. In Boeing's case, an $8.3 billion acquisition to regain control of a supplier whose quality data it could no longer trust. Most of that is corporate liability, paid by the company. One channel is different. Under *Caremark*, the Delaware doctrine governing a director's duty of oversight, a board member can be held personally liable for failing to ensure a system existed to surface material risks. Boeing's board settled such a claim for $237.5 million in 2021, the largest oversight settlement in Delaware history. D&O insurance absorbs most of that exposure in practice, but coverage carries limits and bad-faith exclusions, and no policy shields a director from a deposition, a named-defendant docket entry, or the years a derivative action takes to resolve. > If a company makes a physical product, the conforming quality of that product is almost certainly a mission-critical compliance domain. The test a court applies with hindsight is whether a system existed to bring measurement evidence to the board's attention. Of the eleven cases, only Boeing's produced an actual Caremark settlement. The other ten are the raw material from which such claims get built: documented failures to escalate data the company already held, sitting one derivative filing away from becoming the next Boeing. The doctrine has also been getting easier to plead. *Marchand v. Barnhill* (2019) allowed a claim against Blue Bell Creameries' board after a listeria outbreak, because food safety was the company's mission-critical compliance domain and the board had no committee covering it. In 2023, *McDonald's* extended oversight duties beyond directors to corporate officers. In 2024, the Delaware Supreme Court revived an oversight claim against AmerisourceBergen's directors over opioid distribution monitoring. The perimeter of personal accountability keeps moving outward. For a director of a manufacturer, the translation is direct. If the company makes a physical product, the conforming quality of that product is almost certainly a mission-critical compliance domain. The test a court applies with hindsight is whether a system existed to bring measurement evidence to the board's attention. --- ## The case that shows the mechanism Boeing repays close reading because nothing malfunctioned. The failure was governance. ![](https://storage.ghost.io/c/97/2a/972a49a2-bf56-480d-bfc6-da726d4924be/content/images/2026/07/exhibit2_boeing_anatomy--1--1.png) **Exhibit 2 — Anatomy of a metrology governance failure: Boeing.* Five-year sequence, cause, cost and board consequence. Sources: NTSB, US DOJ Criminal Division, FAA, Delaware Court of Chancery, company disclosures.* On 5 January 2024, a mid-exit door plug separated from a 737 MAX 9 at roughly 16,000 feet. The plug had been opened on the factory floor to repair rivet defects and reinstalled without its four retention bolts. No instrument gave a false reading. No sensor failed. What was missing was the record. The NTSB's investigation turned on the absence of documentation for a non-routine removal, and Boeing's own position was that if the removal was undocumented, there would be no documentation to share. Upstream of that sat a supplier estate that stopped at the factory gate while Spirit AeroSystems was failing 54% of its federal quality audits. The resulting NTSB finding — a failure to provide the training and oversight necessary for personnel to comply with the parts-removal process — is not a manufacturing finding. It is a governance finding about a manufacturing process. We think this is a crucial distinction that many boards miss. And drives the pressing point that resolution is not more measurement. Its more governance. Boeing's underlying problems run deeper than measurement. The production culture that followed the 1997 McDonnell Douglas merger contributed, as did the outsourcing of fuselage work and sustained pressure on production rates. Better metrology governance would not have fixed Boeing. It is the layer at which a problem of that kind becomes visible to a board before it becomes visible to everyone else. --- ## What a board actually needs Quality management is an operational function. It runs the machinery, for the product, under management. What the pattern above demands in our experience is different: a board-level system that can answer one question with evidence — **can the board demonstrate that it knew what its instruments knew?** We call this metrology governance. It stands to the quality management system as cyber governance stands to IT operations. It does not run the equipment. It ensures the equipment's output reaches the people carrying the legal duty, in a form that survives scrutiny. We see this made up of Five key components: **1\. Sensor estate mapping.** Every measurement point touching product conformance or a regulatory obligation, ranked by criticality, extended to critical suppliers. This is often simpler than it appears, much of the envelope in our experience has already been shaped by a quality system. **2\. Obligations register.** This is less simple. Each measurement stream needs to be mapped to the specific duty it evidences: a certification requirement, a contractual clause, a safety regulation, an environmental limit. In our experience this is where many boundaries blur and expectations differ. Often this exposes the cultural differences of different teams, and sometimes our role is to ajudicate, as much as it is to record. **3\. Evidence chain review.** The trace from instrument to board pack. Calibration status and traceability. Data integrity in transit, including whether readings can be altered, suppressed or auto-deleted. The aggregation points where individual exceedances vanish into averages. The last one is often key. Most manufacturing environments exclude "bad" readings. And practically this works. Until the trend of bad readings is smoothed by their exclusion. Again, this in our experience speaks as much to culture as compliance and is often a key finding from a diagnostic. **4\. Automation and escalation risk.** Where automated systems act on measurement data without human review, and where escalation depends on individual judgement rather than a defined trigger with a named owner. This is increasing in importance and the use of inline sensing especially for quality control is a key area to monitor. **5\. Board attestation framework.** Committee ownership and reporting cadence, with a documentation standard built for a books-and-records demand rather than a registrar's audit. If boards and companies in Exhibit 1 had a Metrology Governance framework things may have been very different. ![](https://storage.ghost.io/c/97/2a/972a49a2-bf56-480d-bfc6-da726d4924be/content/images/2026/07/exhibit3_what_would_differ--1--1.png) **Exhibit 3 — What would have been different with a metrology governance program.* The same eleven cases, mapped against the four technical steps of the diagnostic. Empty cells indicate the step is not the primary gap for that case.* Two patterns stand out. The evidence chain carries seven of the eleven cases — most large oversight failures happen between the instrument and the boardroom, not at the instrument. And several cells stay empty, because a framework that claimed to prevent everything would deserve less trust, not more. --- ## What this does when a claim arrives General counsel will ask whether any of this stops a books-and-records demand. It does not. The credible basis standard is the lowest burden of proof in Delaware corporate law, and *AmerisourceBergen* (2020) held that a stockholder showing a credible basis to infer wrongdoing need not demonstrate the wrongdoing is actionable. A derailment or a consent decree supplies that basis on its own. The demand is the wrong battlefield. Oversight claims are won and lost at the motion to dismiss, and what decides them is what the production reveals. In *NiSource* (Del. Ch., June 2022), following a series of pipeline explosions, Chancery rejected the claim because the books and records the plaintiffs themselves obtained showed the board had established a system for monitoring the mission-critical risk. The production demonstrated the existence of a system rather than its absence. *SolarWinds* (Del. Ch., September 2022, affirmed May 2023) dismissed on similar grounds, as did *Centene* (Del. Ch., July 2024), where a compliance reporting system was in place and appeared to be functioning at least to some extent. *Marchand* is the mirror image: Blue Bell lost because there was no board-level committee and no reporting stream at all. The knife cuts both ways, and a board should understand that going in. A review that identifies gaps creates documented, board-level red flags. In *Brewer v. Regions Bank* (2025), an internal document identifying legal risk became the plaintiff's most powerful evidence, and Chancery rejected the directors' argument that they had responded, merely not as quickly as the plaintiff wanted, holding that delay can itself be a tactic. Findings without remediation are worse than no findings. That is the argument for structuring the work around a report on a process rather rather than a report for a drawer somewhere, "just in case" --- ## How to organise a Metrology Governance process In our experience a review of this kind should run the way a board already runs any serious risk assessment, not as a novel exercise invented for the problem. There are 3 key principles: **First, involve counsel from the outset.** The company's own counsel should scope the engagement, determine what protections attach to the work product, and advise on how findings are documented. Whether particular materials attract privilege is a legal question specific to the engagement and the jurisdiction. **Second, understand the discovery landscape.** Any board-level risk review can become relevant in later litigation or regulatory inquiry. That is true of cyber assessments and safety audits alike, and it is not a reason to avoid the review. Delaware's caselaw is consistent: boards are better served by having conducted a diligent, documented assessment and acted on it than by never having looked. **Third, produce action rather than a document.** The true value of a metrology governance program, of of any board driven governance programs, is to set a standard, communicate expectations and shape behaviours. ![](https://storage.ghost.io/c/97/2a/972a49a2-bf56-480d-bfc6-da726d4924be/content/images/2026/07/exhibit4_structure-1.png) **Exhibit 4 — An example best practice metrology governance diagnostic.* Indicative timings, tasks, team and deliverables.* The approach is deliberate. Mobilisation and board attestation form the legal and governance wrapper. The four technical steps sit inside it. Most of our work is in these steps, ensuring capability is transferred to the client team wherever possible. Mobilisation settles scope and reporting lines before any data moves. Board attestation closes the loop by designing the quarterly governance pack and setting committee ownership, then stress-testing the resulting record against the kind of demand that broke Boeing. --- ## The future of metrology governance Cybersecurity did not arrive on the board agenda because directors developed an interest in network architecture. It arrived because the cost of not having it there became obvious, publicly, to everyone at once. Measurement data is on the same path, and the formal crystallisation of expectations and reporting by boards means Metrology Governance programs will be inevitable. The volume is already enormous. Robotics and industrial sensing are compounding it at close to 10% a year. The enforcement pattern is established and the doctrine is expanding. The need to close the metrology governance gap has never been greater, and forward thinking boards are already taking steps to close these gaps. --- *Metrology governance reviews for boards and audit committees are conducted through metrology.com's advisory practice, scoped in coordination with client counsel. The standard diagnostic runs eight weeks. This article is general commentary and not legal advice; companies should consult their own counsel on privilege and disclosure questions specific to their circumstances. To discuss any aspect of our work or the themes raised above contact us at team@metrology.com* --- ### 1. International Federation of Robotics, *World Robotics 2025*, September 2025. 2. NTSB probable-cause finding and public hearing, Alaska Airlines Flight 1282. 3. Boeing statement to NTSB regarding removal documentation, March 2024. 4. In re The Boeing Company Derivative Litigation, C.A. No. 2019-0907-MTZ (Del. Ch.). 5. DOJ non-prosecution agreement, May 2025. 6. Philips Q1 2024 results; Respironics settlements and consent decree. 7. DOJ and EPA, Cummins Clean Air Act settlement, December 2023. 8. CPUC Administrative Consent Order, PG&E Dixie Fire, January 2024. 9. *Marchand v. Barnhill*, 212 A.3d 805 (Del. 2019). 10. *In re McDonald's Corp. Stockholder Derivative Litigation* (Del. Ch. 2023). 11. AmerisourceBergen Corp. v. Lebanon County Employees' Retirement Fund, 243 A.3d 417 (Del. 2020). 12. City of Detroit Police & Fire Ret. Sys. v. Hamrock (NiSource), 2022 WL 2387653 (Del. Ch. June 30, 2022). 13. Construction Industry Laborers Pension Fund v. Bingle (SolarWinds), 2022 WL 4102492 (Del. Ch. Sept. 6, 2022), aff'd May 17, 2023. 14. Bricklayers Pension Fund of Western Pennsylvania v. Brinkley (Centene), Del. Ch., 12 July 2024. 15. Brewer v. Regions Bank directors, Del. Ch., 2025. 16. *In re Caremark International Inc. Derivative Litigation*, 698 A.2d 959 (Del. Ch. 1996). ### 3 Point Bore Gauging Designed to Drawing URL: https://www.metrology.com/3-point-bore-gauging-designed-to-drawing/ Last updated: 2026-07-24T03:49:27.000Z MET-BRF-001 Design Office ## Send us the drawing. A design engineer responds within 24 hours. No drawing yet? You can describe the part instead. Start a Brief→ ### Uber for Robots just raised $1.7bn. The Metrology gap just got bigger. URL: https://www.metrology.com/uber-for-robots-just-raised-1-7bn-the-metrology-gap-just-got-bigger/ Last updated: 2026-07-23T05:19:36.000Z Travis Kalanick raised $1.7bn this week to build what he’s calling the “Age of Atoms” — industrial AI and robotics aimed at mining, heavy transport, construction and food production. Andreessen Horowitz led. Bain chipped in. Ben Horowitz joined the board. Uber, the company that fired Kalanick in 2017, put money back in. It is one of the largest private robotics raises on record, in a year that has already broken the annual funding record for the category twice over. --- ## Physical AI is not Digital AI. (Obviously) Atoms is a bet that AI is about to leave the world of bits and start operating reliably in the world of atoms. But there's a big metrology gap yet to be filled. Digital AI got to scale because its substrate is forgiving. A large language model that’s 0.1% wrong produces a slightly awkward sentence. Nobody dies, nothing seizes, no part fails to fit (also AI helpfully writes in triplets. Like this just did). Physical AI has no such grace. A robot arm placing a component, a mining system judging clearance, a food-production line sorting to spec — all of it depends on knowing the true dimension of a physical object, not a probability led best case guess. Dropping a pizza and dumping a 240T truck have very different consequences. Atoms don’t round.This is the part the “Age of Atoms” framing glosses over. Intelligence — the model, the planning layer, the language Kalanick’s deck undoubtedly uses — is the part venture capital already knows how to fund and scale. What it has never had to fund at scale before is ground truth: the sensing and measurement layer that tells a physical system what is actually in front of it, to the tolerance the task requires. You can have the best-trained robot on earth and it is worthless the moment it can’t tell a 20.02mm bore from a 20.00mm one on a part with a 0.01mm tolerance. --- ## We've seen this movie before Robotics has been here before. The last robotics boom, in the 1980s, ran into exactly this wall. The arms were capable; the vision and sensing were not. Factories automated the easy 80% — repetitive motion — and stalled on the hard 20%, which was almost always perception: knowing precisely where something was, what shape it actually held, whether it met spec. That 20% is where inspection, quality and metrology still live, forty years later, mostly untouched by the AI wave because it’s the part that’s hardest to fake.What’s different this time is that vision systems, sensor fusion and edge compute have genuinely closed some of that gap. But closing the sensing gap and closing the measurement gap are not the same achievement. A camera can locate a part. It takes a calibrated, traceable measurement system to certify that part’s dimension is correct — and that distinction is the entire difference between a robot that looks impressive in a demo and one an aerospace or automotive quality department will actually sign off on. --- ## The Metrology Gap Just Got Bigger The key dependency of the Atoms bet is a measurement infrastructure that lets a physical AI system trust what it’s looking at, and lets a human trust the AI. Gauges, setting rings, calibration chains, vision systems certified to a standard. Boring, pedestrian and vital. This is not a footnote to physical AI. It’s the substrate it’s built on.The Age of Atoms needs an Age of Measurement underneath it. Right now, almost nobody funding the former is thinking about the latter, because its not a problem yet. But will be, and the real picks and shovels opportunities in the age of Atoms are just coming into view. ### Elon needs half a billion calibration certificates. No one has them. Yet. URL: https://www.metrology.com/elons-in-the-market-for-half-a-billion-calibration-certificates/ Last updated: 2026-07-21T09:49:12.000Z **The line starts this month.** If Tesla's schedule holds, the first Optimus V3 units begin moving down a Fremont production line that was still building Model S sedans in early May. Musk told the Q1 earnings call that initial output would be slow and impossible to forecast, which is fair for a machine with 10,000 unique parts on an entirely new line. But the intent is not modest. The Fremont line is designed for a million robots a year. Site work at Gigafactory Texas is underway for a second line targeting ten million. Boston Dynamics is already shipping its electric Atlas into Hyundai plants, and more than a hundred Chinese manufacturers are racing down the same road. Argue about the timeline all you like; the direction is settled. Humanoid robots are entering factories at industrial volume, and a measurement liability is walking in with them that the quality profession has not yet been asked to price. Which few understand. And a whole new calibration industry about to wake up. --- **A robot is a measuring instrument that walks.** Strip away the theatre and look at the bill of materials. Optimus V3 carries 37 joints, each with position encoding. The hands alone run 22 degrees of freedom across 50 actuators with force feedback. Add the cameras, the inertial units, the torque and temperature sensing distributed through the frame, and a single humanoid is carrying hundreds of transducers. Now run the arithmetic behind the headline. A million robots a year, times hundreds of internal measurement channels per robot, is on the order of half a billion sensor channels entering production annually — each one feeding decisions at machine speed. When the robot seats a connector or torques a fastener, the signal confirming the job was done correctly is its own sensor reading. --- > "A million robots a year, times hundreds of internal measurement channels per robot, is on the order of half a billion sensor channels entering production annually" --- **To be fair: the arm gets calibrated.** ISO 9283 has defined performance characteristics for industrial robots since 1998 — pose accuracy, repeatability, path accuracy, drift, overshoot — and an entire service industry verifies against it. Laser-tracker firms will map a robot's kinematic errors on site, update the controller parameters and hand over a written report with before-and-after numbers, on a verification sheet traceable through an accredited laboratory to national standards. Well-run plants put their production robots on an annual calibration cycle exactly as they would a CMM. For the conventional six-axis arm bolted to a factory floor, the discipline exists, the paperwork exists and the accreditation chain behind it is real. There is no single, specialized ISO calibration standard exclusively for humanoid robots. The recently discussed ISO 25785-1 covrs some but really focused on safety and guardrails. Instead, manufacturers of humanoid robots rely on a combination of general industrial robot standards, datasets, and personal care standards: --- **But look at what that certificate actually certifies.** Kinematic calibration answers one question: where does the tool centre point end up in space? It verifies the skeleton. It says nothing about the senses. The force feedback in the fingers, the torque sensing in the joints, the cameras deciding whether a part is present and correctly seated — those transducers were calibrated once, at component manufacture, by their supplier, before being buried inside a working machine. From that day forward there is no recall interval, no periodic verification and no maintained traceability for any of them, even though those are precisely the channels making accept-and-reject decisions in production. And the gap widens for humanoids. ISO 9283 was written for a manipulator fixed to the ground, tested along defined paths inside a static cube. ISO 25785-1 covers outputs and making sure the robot hordes don't rise up and take us over, one toaster at a time. No published method covers a machine that walks, shifts its own base, and fuses hundreds of sensor streams to decide what it just did. No accreditation scope exists for verifying one, let alone a fleet. Half a billion channels a year, and not one document that would survive an audit. --- > ISO 9283 was written for a manipulator fixed to the ground, tested along defined paths inside a static cube. No published method covers a machine that walks, shifts its own base, and fuses hundreds of sensor streams to decide what it just did. --- **Every sensor drifts. That is physics, not pessimism.** A human machinist notices when a reading feels wrong; that intuition was the factory's original anomaly detector. A robot optimises on whatever its senses feed it, including the drift, at machine speed, around the clock. Aerospace learned where that road leads. A modern aircraft's sensors live under maintenance schedules and airworthiness directives because the consequences of unverified measurement were written into accident reports. Humanoid fleets are arriving with no equivalent, at a hundred times the unit volume. So the question writes itself. When a robot-assembled product fails in the field, who can evidence the measurement chain that accepted it? The kinematic certificate proves the arm went where it was told. It proves nothing about whether the hand felt what it claimed to feel. "The robot's sensors said it was fine" is not an answer any quality manager wants to give under oath. --- **The Datum.** Two conclusions, one for each side of the factory door. For manufacturers deploying humanoids: the instrument register must include the robots — not just as assets to be kinematically aligned, but as estates of embedded transducers, each one touching an accept-or-reject decision, each one needing an interval, a method and a record before deployment rather than after the first quality escape. For the calibration industry: the arm-calibration business is mature and contested; the sense-verification business does not yet exist. That is the largest unclaimed instrument population to enter factories since the coordinate measuring machine. The lab that writes the first defensible in-service verification procedure for a humanoid's senses will set the standard everyone else is audited against, and standards, once set, are very hard to take away. Elon is going to need those half-billion certificates from somebody. Right now, nobody is in a position to issue them — and that is the most valuable gap in the measurement business. ### Deep Bore Measurement - The Three Minute Metrology Guide URL: https://www.metrology.com/deep-bore-measurement-the-three-minute-metrology-guide/ Last updated: 2026-07-20T01:33:34.000Z --- ## **The issue** Depth is a key challenge for accurate bore gauging. Honed hydraulic cylinders, gun barrels, drive housings, heat-exchanger bores — the diameter that matters is metres from the operator's hand, and three separate problems compound on the way down: the gauge axis must stay aligned to the bore axis over the whole distance, the reading must survive the journey back, and — the failure mode nobody prices — the instrument's own temperature changes as it travels down a bore that isn't at 20°C. The wear you're looking for is real; whether your instrument can report it truthfully from four metres away is the actual question. --- ## **The problems with the usual answers** A conventional gauge on a long stick answers none of the three: it flexes, it cocks in the bore, and its indicator sits at the warm end of a cold column of steel, reading the temperature gradient as faithfully as the diameter. Air gauging at depth needs close-toleranced sleeves that fight the very bores most in need of checking. Borescopes show what the surface looks like — informative, and dimensionally silent: you can't borescope a diameter. None of these delivers a traceable figure at a stated depth, repeatably, from either end of a metre-class bore. --- ## **The three-point advantage** The deep-hole system solves each failure separately. A pneumatically actuated head collapses its anvils for insertion and releases them at depth — no mechanical trigger linkage running the full length. Spherical contact points tolerate small angular deviation between gauge axis and bore axis without corrupting the reading — the practical cure for alignment over distance. And critically, the long-range electronic transducer sits at the head, adjacent to the measuring point, so the temperature that matters is measured where it acts — not metres away at the readout. One-metre extensions join in series — up to twenty have been used on a single application — for system accuracy of ±0.005 mm (0.0002") across a 50–300 mm (1.97"–11.81") head range, proven past 4 m (13 ft) in service. For moderate depths, the standard Trigabore system with joined 100/150 mm (3.94"/5.91") extensions covers most work at 0.002 mm (0.0001") repeatability; the design office will tell you which side of that line your application sits. --- ## **How to specify** Four dimensions from your drawing decide the design — the diameter to be measured (MD), the smallest diameter the head must pass through (ED), the depth at which measurement is required (L, stated honestly), and the clearance beyond the measuring point (DC) — plus the tolerance on MD and the environment, since in-factory and field use drive different transducer and extension choices. Spherical contacts are recommended for deep holes; blind-bore and ball forms are available where the geometry demands. Send the drawing to the design office and a designed solution — head, extensions, transducer, readout — comes back within 24 hours. ![](https://storage.ghost.io/c/97/2a/972a49a2-bf56-480d-bfc6-da726d4924be/content/images/2026/07/MET-GDE-008B-deepbore-specification.svg) MET-BRF-001 Design Office ## Send us the drawing. A design engineer responds within 24 hours. No drawing yet? You can describe the part instead. Start a Brief→ ### Taper Measurement - The Three Minute Metrology Guide URL: https://www.metrology.com/taper-measurement-the-three-minute-metrology-guide/ Last updated: 2026-07-20T01:35:31.000Z --- ## **The issue** A taper has a different diameter at every depth — which means "measure the bore" is an incomplete instruction. Any reading is meaningless without an answer to *where*: at what axial position, referenced from which face. Spindle noses, valve seats, hydraulic cones and tooling bores live or die on two numbers — the diameter at a defined datum, and the included angle — and a degree-minute of angle error at the spindle becomes runout at the tool tip, discovered in the cut rather than at inspection. --- ## **The problems with the usual answers** A conventional bore gauge will happily read *a* diameter in a taper — self-centring guarantees it reads *somewhere* — but without datum hardware, that somewhere shifts with operator technique, and two shifts of the same shop will "measure" different parts as identical. Taper plug gauges answer fit by bluing: informative, subjective, and numberless. A CMM answers everything, slowly, in another room. None of these gives diameter-at-datum and angle as repeatable shop-floor values, referenced the way the drawing references them. --- ## **The three-point advantage** A taper-application head fixes the axial question mechanically: a clamp ring and calibrated C-washer locate the head at a defined datum position (LD) from the reference face, and the three-point contacts read the diameter (DD) at exactly that depth — repeatable to 0.002 mm (0.0001") in any operator's hands. Remove the C-washer of known thickness and re-seat, and the head reads at a second, precisely displaced position: from two diameters and a known displacement, the included angle follows directly — tan(θ) = (D1−D2)/2L1 — with the arithmetic on the design sheet, not left to the operator. Ball contacts are fitted at the datum for reliable seating on the inclined surface. The head presets against a calibrated setting ring; standard coverage 6–200 mm (0.236"–7.87"), specials from 2 to 500 mm (0.079"–19.69"). --- ## **How to specify** Six parameters from your drawing decide the design — entry diameter (ED), minor diameter (MD), taper length (L), datum position (LD), datum diameter (DD), and the included angle or taper ratio with its tolerance — plus the one detail that shapes the hardware: *which reference face the datum is called from*, because the clamp ring is dimensioned to your component, not adapted to it. Send the drawing to the design office and a designed solution — head, clamp ring, washers, arithmetic — comes back within 24 hours. ![](https://storage.ghost.io/c/97/2a/972a49a2-bf56-480d-bfc6-da726d4924be/content/images/2026/07/MET-GDE-006B-taper-specification.svg) MET-BRF-001 Design Office ## Send us the drawing. A design engineer responds within 24 hours. No drawing yet? You can describe the part instead. Start a Brief→ ### Slot Measurement - The Three Minute Metrology Guide URL: https://www.metrology.com/slot-measurement-the-three-minute-metrology-guide/ Last updated: 2026-07-20T01:35:54.000Z --- ## **The issue** Internal slots break the founding assumption of bore gauging: that the feature is round. A circlip groove that isn't a full circle, a keyway broached inside a bore, a recess machined behind a smaller opening — these are width measurements hiding inside diameter-shaped access, and the geometry is often deliberately obstructive: the slot is wider than the entry that leads to it. A retaining ring that seats in a slot cut 0.1 mm (0.004") off nominal doesn't fail at inspection — it fails in service, two years later, as a warranty claim with a clean paper trail. --- ## **The problems with the usual answers** Calipers can't enter; their jaws are the wrong shape for the opening and the wrong reach for the recess. Pin gauges only bracket the answer — between go and no-go lies everything you actually wanted to know. Feel-based telescoping methods produce a different number per operator. And the common last resort — sectioning a sample part — destroys the evidence to obtain it: one honest number, one scrapped component, and nothing you can do at production rate. None of these yields the slot width as a repeatable, traceable value. --- ## **The three-point advantage** A slot-application head presents opposed anvils sized to pass through the entry width, which then locate on the slot faces and read true width directly on the indicator. The instrument presets against a calibrated master — for slots, typically a certified slip-gauge stack or purpose-made master rather than a ring — so every reading is a deviation from a traceable reference: repeatable to 0.002 mm (0.0001"), single-handed, in seconds, at the machine. Three slot families are catered for: plain slots, recess slots wider than their opening, and special-shape slots with contacts machined to the profile. Standard coverage 6–200 mm (0.236"–7.87"); specials from 2 to 500 mm (0.079"–19.69"). --- ## **How to specify** Six dimensions from your drawing decide the design — slot width (W), measured length (ML), entry width (EW), depth from face (L), clearance below the slot (DC), and head clearance (HC) — plus the slot type: plain, recess, or special (attach the profile drawing). The governing check is EW − ML against the head's measuring range — the single most common reason an off-the-shelf attempt at slot measurement fails, and the first thing our design office verifies on your drawing, before manufacture rather than after delivery. Send the drawing to the design office and a designed solution — head, anvils, master — comes back within 24 hours. ![](https://storage.ghost.io/c/97/2a/972a49a2-bf56-480d-bfc6-da726d4924be/content/images/2026/07/MET-GDE-004B-slot-specification.svg) MET-BRF-001 Design Office ## Send us the drawing. A design engineer responds within 24 hours. No drawing yet? You can describe the part instead. Start a Brief→ ### Ball Screws - The 3 Minute Metrology Guide URL: https://www.metrology.com/ball-screws-the-3-minute-metrology-guide/ Last updated: 2026-07-20T01:36:57.000Z --- ## **The issue** A ball screw doesn't run on its thread crests — it runs on the track, the semicircular or gothic-arch groove where the balls actually roll. The only diameter that predicts performance, preload and backlash is the running diameter measured exactly where a ball sits. Measure the crest and you've measured the one surface the working balls never touch — with great confidence, and to no purpose. It's why ball nuts and internal ball tracks are among the most consistently mis-measured components in precision transmission work. --- ## **The problems with the usual answers** A standard three-point or two-point head reads the crest or the root — real numbers about the wrong surfaces. Pin and wire methods are external-thread techniques; inside a nut they have nowhere to stand. A CMM can trace the groove profile given a fixture, a program and patience — a lab exercise, not a goods-in check, and it still reports geometry the inspector must convert into an effective running diameter. Meanwhile the numbers the drawing actually specifies — track diameter over balls, at a stated ball size — go unverified from supplier to assembly. --- ## **The three-point advantage** The correct contact for a ball track is a ball — of the same nominal diameter as the working balls. Our ball screw heads carry two-ball or three-ball contact sets, sized to your ball diameter, that seat in the groove precisely as the running balls do and read the effective track diameter directly on the indicator. Three-ball contacts self-centre; two-ball sets suit accessible geometries. The head presets against a calibrated ring, minor diameter can be gauged with standard contacts on the same handle, and repeatability is 0.002 mm (0.0001") in any operator's hands. Standard coverage 6–200 mm (0.236"–7.87"); specials from 2 to 500 mm (0.079"–19.69"). --- ## **How to specify** Six parameters, plus the profile: major diameter (MD1), minor diameter (MD2), pitch (P), depth of thread (DT), ball diameter (BD), and hand (LH/RH) — and attach the thread profile drawing, because the contact set is machined to your groove form, not picked from a page. The governing check is 2 × DT against the head's measuring range; the design office verifies it on the drawing before anything is cut. Send the profile to the design office and a designed solution — head, ball contacts, setting ring — comes back within 24 hours. ![](https://storage.ghost.io/c/97/2a/972a49a2-bf56-480d-bfc6-da726d4924be/content/images/2026/07/MET-GDE-003B-ballscrew-specification.svg) MET-BRF-001 Design Office ## Send us the drawing. A design engineer responds within 24 hours. No drawing yet? You can describe the part instead. Start a Brief→ ### Spherical Bores - The Three Minute Metrology Guide URL: https://www.metrology.com/spherical-bores-the-three-minute-metrology-guide/ Last updated: 2026-07-20T01:37:17.000Z --- ## **The issue** A spherical bore is the geometry that defeats the standard drawer for one simple reason: the feature is larger than the hole that leads to it. Ball seats, hydraulic sockets, rod-end housings, valve interiors — the sphere's full diameter sits behind a smaller entry, at exactly the point no jaw, anvil or stylus naturally reaches. So the sphere gets inferred: from the entry size, from the machining program, from the tool that cut it. And an inferred diameter is an assumption wearing a measurement's clothes — right up until a seat leaks or a ball binds, at which point it becomes a warranty claim with a paper trail that says "pass." --- ## **The problems with the usual answers** Calipers and two-point bore gauges can't pass the entry, and if forced in at an angle they read a chord, not a diameter. Telescoping gauges rely on operator feel inside a cavity the operator can't see. A CMM with a star stylus can map the sphere — at the cost of a fixture, a program and a queue, turning a ten-second shop-floor check into an inspection-room event. Sectioning gives one honest number and one destroyed part. None of these puts the true sphere diameter, as a repeatable value, in an operator's hand at the machine. --- ## **The three-point advantage** A spherical-application head enters with its anvils fully collapsed — the trigger retracts all three contacts — passes through the entry diameter, and releases so the anvils expand and seat on the sphere at its widest point. Three contacts at 120° self-centre in the cavity: the reading is the diameter, not a chord, not a feel. The head presets against a calibrated setting ring, so every reading is a deviation from a certified reference — repeatable to 0.002 mm (0.0001") in any operator's hands, single-handed, in seconds. Spherical or ball contact points are fitted to suit access and surface. Standard coverage 6–200 mm (0.236"–7.87"); specials engineered from 2 to 500 mm (0.079"–19.69"). --- ## **How to specify** Four dimensions from your drawing decide the design — spherical diameter (SD), entry diameter (ED), depth from face to sphere (L), and bottom clearance (HC) — plus the tolerance on SD, which sets the contact form and indicator resolution. The governing check is SD − ED against the head's measuring range; deeply re-entrant spheres that exceed it take a special anvil design, resolved at the drawing stage rather than discovered at goods-in. Send the drawing to the design office and a designed solution — head, contacts, setting ring — comes back within 24 hours. ![](https://storage.ghost.io/c/97/2a/972a49a2-bf56-480d-bfc6-da726d4924be/content/images/2026/07/MET-GDE-002B-spherical-specification.svg) MET-BRF-001 Design Office ## Send us the drawing. A design engineer responds within 24 hours. No drawing yet? You can describe the part instead. Start a Brief→ ### Spline Measurement - The Three Minute Metrology Guide URL: https://www.metrology.com/metrology-guide-to-spline-measurement/ Last updated: 2026-07-20T01:37:49.000Z --- ## **The issue** Ask what diameter an internal spline is and you've asked three questions. Major, minor and pitch circle diameter each govern a different aspect of fit — and the PCD, the one that controls load sharing and backlash, is a geometric construction no surface of the part physically presents. Meanwhile the tool that cuts the spline is wearing from its first stroke: PCD drift is the earliest warning a broach gives, and in most shops nobody is positioned to hear it until the scrap bin does. --- ## **The problems with the usual answers** Measurement between pins with calculation is accurate and slow — a bench exercise, not a production check, and error-prone on odd tooth counts where the pins don't sit opposite each other. Gear-checking machines and CMMs give the full answer at the cost of a queue: the part leaves the machine, the number arrives an hour later, and the broach keeps cutting in between. Plug-style spline gauges answer go/no-go only — no value, no trend, no warning. None of these puts the PCD, as a number, in an operator's hand at the spindle. --- ## **The three-point advantage** A spline-application head carries contacts machined to your tooth form — line or point contacts seating in the tooth spaces at the pitch line (the best-wire position) for PCD, line contacts for major diameter, extended contacts reaching the root for minor. Three-point contact self-centres in the spline, odd and even tooth counts each get their own contact arrangement, and the head presets against a calibrated setting ring so every reading is a deviation from a certified reference — repeatable to 0.002 mm (0.0001") in any operator's hands, in seconds, at the machine. If your drawing states a measurement-over-pins figure, the contacts are designed to reproduce that exact convention, so your gauge and your customer's calculation agree by construction. Standard coverage 6–200 mm (0.236"–7.87"); specials from 2 to 500 mm (0.079"–19.69"). --- ## **How to specify** Send the spline profile drawing with full design details — it's the one application where the drawing isn't optional, because the contact geometry is machined to the tooth form. From it we take: number of teeth (N), pitch circle diameter (PCD), major (MAJ) and minor (MIN) diameters, and, if specified, best wire size (BW) and distance between pins (DBP) — plus which diameter or diameters you need to read. Odd counts, even counts, and combined-diameter heads on one handle are all standard design work. Send the profile drawing to the design office and a designed solution — head, contacts, setting ring — comes back within 24 hours. ![](https://storage.ghost.io/c/97/2a/972a49a2-bf56-480d-bfc6-da726d4924be/content/images/2026/07/MET-GDE-005B-spline-specification-1.svg) MET-BRF-001 Design Office ## Send us the drawing. A design engineer responds within 24 hours. No drawing yet? You can describe the part instead. Start a Brief→ ### O Ring Measurement - The Three Minute Metrology Guide URL: https://www.metrology.com/metrology-guide-to-o-ring-measurement/ Last updated: 2026-07-20T01:38:36.000Z --- ## **The issue** An O-ring is only as good as the gland it sits in, and every sealing-failure investigation eventually arrives at the same unglamorous place: the groove. Its diameter — always larger than the bore that hides it — carries some of the tightest tolerances on the drawing, routinely ±0.05 mm (±0.002") in aerospace and hydraulic work, on a surface nobody can see and most instruments cannot reach. The geometry is deliberately obstructive: the measured diameter (MD) sits behind a smaller entry bore (ED), exactly where a caliper's jaws and a standard bore gauge's anvils cannot go. Get it wrong and the consequences arrive later, in service: too shallow and the seal extrudes under pressure; too deep and compression is lost. Either way, the leak is diagnosed months after the part left inspection with a pass. --- ## **The problems with the usual answers** Telescoping and small-hole gauges depend entirely on operator feel — two inspectors, two answers, and no defensible number for the quality record. Moulded impression compounds add a casting step, its cure time, and its shrinkage to every reading, converting a dimensional measurement into a materials experiment. Sectioning a sample gives one accurate number and one destroyed part — acceptable for a failure investigation, useless for production control. A CMM with a star stylus can reach some grooves, but turns a ten-second shop-floor check into an inspection-room queue, and still struggles to establish the true diameter on a narrow radius-form groove where the stylus has almost nothing to land on. None of these belongs on a production line, and none produces the repeatable, traceable figure an auditor or a customer deviation request demands. --- ## **The three-point advantage** A purpose-designed groove head runs extended anvils sized to pass through the entry bore, which then locate outward into the groove and seat on its true diameter. Three carbide contacts at 120° self-centre in the groove — the reading is the diameter, not a chord, and not a feel-dependent estimate. The instrument presets against a calibrated setting ring, so every reading is a deviation from a certified reference, and the result appears directly on the indicator: single-handed, on the shop floor, in seconds, repeatable to 0.002 mm (0.0001") by any operator, with no special skill required. Square grooves, radius grooves and special profiles each take their own contact form, machined to your gland. Where the diameter alone isn't the whole story, the companion width system measures groove width, position from the face, and relative position between grooves — the full gland geometry, with no calculation asked of the operator. Standard-system coverage runs 6–200 mm (0.236"–7.87"), with specials engineered from 2 to 500 mm (0.079"–19.69"). --- ## **How to specify** Five dimensions from your drawing decide the design — measured diameter (MD), entry diameter (ED), groove width (W), depth to the groove (L), and the clearances around it (DC, HC) — plus the groove type: square, radius, or special profile (attach the profile drawing). The governing check is MD − ED against the head's measuring range; our design office verifies it before anything is made, so the limitation is engineered out on paper rather than discovered in the inspection room. ![](https://storage.ghost.io/c/97/2a/972a49a2-bf56-480d-bfc6-da726d4924be/content/images/2026/07/MET-GDE-007B-oring-specification--1--2.svg) MET-BRF-001 Design Office ## Send us the drawing. A design engineer responds within 24 hours. No drawing yet? You can describe the part instead. Start a Brief→ ### Thread Measurement - The Three Minute Metrology Guide URL: https://www.metrology.com/metrology-guide-to-thready-meaujdsjdnd/ Last updated: 2026-07-20T01:39:42.000Z --- ## **The issue** An internal thread has three diameters, and the one that governs fit — the pitch diameter — is the one no ordinary instrument touches. It lives between the crests, on the flanks that actually locate the mating part, routinely toleranced tighter than either diameter you can see. When a thread fails downstream — a fitting that leaks, a fastener that strips, a joint that loosens in service — the investigation traces back to a pitch diameter nobody ever measured as a number. The tap wore, the thread drifted, and every part in between shipped with a pass. --- ## **The problems with the usual answers** Go/No-Go thread plugs answer only a binary question: in or out of limit. They report no value, reveal no trend, and give a machinist no warning that a tap is wearing — "the plug passed it" is where most failure investigations dead-end. Three-wire methods are external-thread techniques; inside a bore they're impossible. A CMM can chase a thread flank with enough programming, but it turns a shop-floor check into an inspection-room queue and still struggles for a defensible pitch diameter on smaller forms. None of these produces what an SPC programme, an AS9100 auditor or a supplier dispute actually requires: the pitch diameter as a repeatable, traceable number. --- ## **The three-point advantage** A three-point head fitted with profiled thread contacts seats directly on the flanks at 120° spacing, self-centring in the bore, and reads pitch diameter as a value on the indicator — preset against a master calibrated thread ring, so every reading is a deviation from a certified reference. Different contact forms address each diameter: line contacts for pitch circle diameter, line contacts for major, standard blind-bore contacts for minor — Metric, UN and Whitworth forms, left or right hand, each with its own contact design. Repeatability is 0.002 mm (0.0001") in any operator's hands, single-handed, in seconds, at the machine. Standard system coverage runs 6–200 mm (0.236"–7.87"), with specials engineered from 2 to 500 mm (0.079"–19.69"). --- ## **How to specify** Six parameters from your drawing decide the design — major diameter (MD1), minor diameter (MD2), pitch diameter (PD), pitch (P), depth of thread (DT), and hand (LH/RH) — plus the thread form: Metric, UN or Whitworth. The governing check is MD1 − MD2 against the head's measuring range (typically 2–10 mm / 0.079"–0.394" depending on size); deep forms like some ACME and buttress profiles can exceed it, which our design office resolves at the drawing stage — before money is spent, not after a gauge arrives that can't reach the flank. If the thread is one feature among several, send the whole part drawing; we frequently combine features into one head. ![](https://storage.ghost.io/c/97/2a/972a49a2-bf56-480d-bfc6-da726d4924be/content/images/2026/07/MET-GDE-001B-thread-specification.svg) MET-BRF-001 Design Office ## Send us the drawing. A design engineer responds within 24 hours. No drawing yet? You can describe the part instead. Start a Brief→ ### 80mm for pole. Give or take a hair. URL: https://www.metrology.com/coming-soon/ Last updated: 2026-07-20T10:04:40.000Z Formula 1 runs on a rulebook, and in 2026 that rulebook was rewritten more completely than at any point in the sport's history. Buried in the new power unit regulations are exactly two clauses held to a tolerance of ±0.1mm — about the width of a human hair, and the tightest lines in the document — and both of them point at the cylinder bore. Its diameter: 80mm, plus or minus a tenth. Its centreline: through the crank axis, plus or minus a tenth. Five manufacturers now machine to those two lines, and the new era's first controversy is already being fought inside them. --- **The Change.** Everything in Formula 1 begins with the FIA's technical regulations — the document that defines what a car is before a designer draws one. For 2026 that document was rewritten across chassis and power unit simultaneously, which the sport almost never does. > Bob Bell, Aston Martin's executive director – technical, framed the scale plainly: "We haven't really had such a comparable change for a very long time." The power unit carries the biggest changes. The MGU-H — the exhaust-energy recovery device that made the 2014 hybrids brilliant and nearly unbuildable — is gone. The MGU-K nearly triples, from 120kW to 350kW, pushing the electrical share of total power toward half. --- **The new kids on the track.** Audi arrives as a works engine builder for the first time, Red Bull's in-house powertrains operation debuts with Ford, and Cadillac has entered the sport while conceding its own engine cannot come before 2028\. Christian Horner warned the resulting cars risked being "Frankenstein" creations; Adrian Newey predicted "an engine formula at the start" — a season decided by powertrains rather than aerodynamics. The opening races have not argued with either of them. --- **The Tolerance.** Read the 2026 power unit regulations with a metrologist's eye and a pattern emerges. The document is generous almost everywhere. Cylinder bore spacing gets ±2mm. The crankshaft's position in the car gets ±0.5mm. Engine mounting studs get ±0.2mm. Connecting rod length gets a full millimetre of window; valve head diameters get two. Most of the rulebook doesn't prescribe dimensions at all — it sets caps and floors, and lets designers exploit toward the limit. This is a choice the FIA has made. But twice, the document tightens to ±0.1mm. Article 5.6.1: the cylinder bore diameter must be 80mm, plus or minus a tenth. Article 5.6.4: each cylinder centreline must pass through the crank axis, plus or minus a tenth. The two finest tolerances in the entire power unit rulebook both belong to the same feature — the hole where combustion happens. That is not a drafting accident. It is the FIA declaring, in tolerance form, the one geometry it refuses to let anyone design around. --- **So what’s new ?** The 80mm figure itself is not new — it has anchored the formula since the hybrid era began in 2014\. What is new is everything around it. Five manufacturers now build to it at once, two of them for the first time, under a compression ratio recut from 18:1 to 16:1, burning a fuel that combusts differently, with every kilowatt of the smaller combustion budget suddenly precious. The dimension is twelve years old. The challenge inside it is brand new. --- **And the fight has already started.** The season's running controversy is, at heart, a dimensional dispute. The 16:1 compression ratio is verified after the working cycle — measured, in effect, on a cooled engine, because measuring a firing one is impractical. Mercedes and Red Bull Powertrains are reported to have selected connecting-rod materials whose thermal expansion raises the effective compression ratio at operating temperature: compliant at the point of measurement, worth real performance at the point of combustion. The bore itself is not in dispute — no rival has questioned anyone's 80mm. Because the bore's cross-section is fixed by regulation, the argument moves entirely to what the metal does at temperature within it — which is exactly why Gary Anderson, the veteran F1 designer, could put arithmetic on the game: over an 80mm piston, 0.1mm of height growth is worth half a cubic centimetre of chamber volume. Mattia Binotto, running Audi's programme, had already called the newcomer's position "a very difficult challenge" before the row broke. Toto Wolff's defence of Mercedes was a metrologist's defence rather than a lawyer's: "it's very clear what the regulations say" — the measurement was made as the rulebook defines it. Both men are right, and that is the point. Fix the geometry, and legality becomes a property of the measurement method; performance becomes a property of what happens between measurements. **The Instrument.** A modern engine programme measures its bores at every stage of a cylinder's life. Liners are gauged after machining and after coating. Bores are verified during and after honing, where the target is not just diameter but shape — a bore can sit legally inside 80mm ±0.1 and still be the wrong geometry, because distortion under head clamping and thermal load is what the piston ring actually experiences. Development engines are torn down after every dyno campaign and gauged again, because the wear map is data: where material went missing says how combustion behaved. The new formula sharpens every one of these measurements — the fuel shifts the thermal profile that drives distortion, and the higher pressures of a 16:1 chamber land on components already at their regulatory minimum weights. The working instrument for most of this life cycle is the three-point internal gauge — the principle behind our own [Trigabore](https://www.metrology.com/trigabore/) system. The scale of the job is worth stating plainly: the regulation allows about a hair's width; the performance lives in a tenth of one; the gauge must resolve a hundredth. Three-point contact self-centres in the bore, removing operator alignment from the result; micron-level repeatability is what lets an engine builder distinguish a real change from measurement noise. Temperature discipline decides the rest. A steel gauge and an aluminium block drifting apart by a few degrees will manufacture more apparent dimension than the tolerance being hunted — the same physics the compression saga turns on, run in reverse. The teams that measure at controlled temperature, against traceable standards, with setting rings certified close to working size, are running the same experiment as the FIA's scrutineers — earlier, and to tighter numbers. Zak Brown, McLaren's chief executive, admitted before the season that what unsettled him was the field: "it is just the unknown." The unknown, reduced to hardware, is two lines of ±0.1mm and what five engine builders managed to do inside them. **The Applications.** Where this case meets our bench — each linked to its page. *Cylinder bore verification, 80mm ±0.1.* Multi-depth readings down the bore map taper, ovality and barrel form during honing and after it — because a liner can pass the regulation and still leak power through shape. The extended-range [Trigabore head](https://www.metrology.com/trigabore/) covers the full regulation bore with three-point self-centring and micron-level repeatability. *(These are the two finest lines in the rulebook.)* *Torque-plate and assembled-state gauging.* A bore measured free is a different bore once the head is clamped on. Gauging under torque-plate load, in the distorted condition the piston ring actually sees, is where self-centring contact matters most — the distortion being hunted is a handful of microns riding on 80mm. [Special gauging enquiries →](https://www.metrology.com/design-office/) *Thermal growth characterisation.* The season's controversy lives in the gap between the cold measurement and the firing engine. Gauging components at controlled elevated temperature, against masters certified at known temperature, quantifies real expansion behaviour — the FIA measures cold, the performance happens hot, and the teams that quantify the difference own it. [Calibration and traceability →](https://www.metrology.com/calibration/) *Groove gauging in manifolds and valve bodies.* Fuel rails at 350 bar, brake-by-wire hydraulics and cooling galleries are dense with seal grooves inside bores — a groove diameter that must be measured through a smaller entry bore, which no standard gauge reaches. That is a drawn-to-order anvil, from a drawing or a description, in 24 hours. [Start a brief →](https://www.metrology.com/design-office/) *Small-bore precision, 2–10mm.* Injector bores, pump plungers, valve guides and tappet bores carry the tightest clearances on the car, in hard and coated materials. The same three-point principle runs from a 3mm guide to the 80mm liner — one instrument family across the whole engine. [The standard range →](https://www.metrology.com/trigabore/) **The Reading.** A rulebook reveals what it values by where it spends its precision. The 2026 regulations hand out millimetres freely and hold back tenths for one feature only — then five manufacturers, the sport's first genuine engine race in a decade, and a compression controversy all converged on that feature within months. None of it is coincidence. A specification does not remove measurement from competition; it concentrates competition into measurement, and it makes the legality of a championship-deciding design choice a question of when and how a dimension is read. The lesson travels far beyond the paddock, into every industry where a drawing is fixed and margins are contested. Find where the rulebook holds its tightest tolerance, and you have found where the game is really played. --- *Sources: FIA 2026 Formula 1 Power Unit Technical Regulations, Issue 4 (Arts. 5.3–5.7); Formula1.com; Mercedes-AMG F1; F1technical.net; ESPN; The Race; Motorsport.com; GPFans; RacingNews365\. Quotes as attributed. Tolerance survey verified against the primary document; claims scoped to the power unit regulations.* --- ## LinkedIn post copy Formula 1's new era is months old and the engine builders are already fighting. Mercedes and Red Bull Powertrains stand accused of gaming the 2026 rules; Audi's Mattia Binotto wants the FIA to act. The issue: compression ratio is verified on a cooled engine, and the accused reportedly chose conrod alloys that grow at temperature — compliant when measured, faster when running. It all traces to one dimension. The bore is fixed at 80mm ±0.1mm — a hair's width, the tightest tolerance in the rulebook. Freeze the hole, and the fight moves inside it: 0.1mm of piston growth is worth half a cubic centimetre. The gauge that polices this must resolve a hundredth of a hair. We make those gauges. New from The Metrologist: Pole Position Is 80mm, Give or Take a Hair.