Guide
Why do coating thickness measurement readings disagree, and which method should the drawing require for each finish?
In receiving inspection, coating thickness measurement is the check that a plated, anodized, painted or powder-coated part carries the film the drawing asked for, made with a gauge or a lab method that suits that particular coating on that particular base metal. Readings disagree mostly because the two sides used different methods, calibrated on different references, or measured different spots, so the fix is to name the method, the reference and the measuring locations on the drawing before the first lot ships.
By Saola Manufacturing engineering team · Published · Updated
This guide is for quality engineers, design engineers and buyers who have a finishing certificate that says one thing and their own gauge that says another. It explains why the readings split, how to find out which one is right, which test method fits each common finish and substrate, and the drawing and purchase order wording that keeps the argument from coming back. The thickness range itself for powder coat is covered in the powder coat thickness guide, and the zinc plating classes in the ASTM B633 guide; this page is about the measurement.
On this page
What does a coating thickness measurement dispute look like?
It usually shows up as a finisher’s certificate that reports the lot inside the callout while your receiving gauge, or your customer’s, reads the same parts thin or thick. Less often, two gauges in the same building disagree, or a cross section from a lab contradicts both.
The pattern matters because it points at the cause. If every reading from one side is offset by about the same amount, suspect calibration or the zero reference. If the averages agree but the minimums do not, suspect where each side measured, because plated and sprayed films are rarely even across a part. If the disagreement only appears on one feature, such as a small boss, an inside corner, a thin flange or a curved shaft, suspect the geometry limits of the probe. And if the two sides used different kinds of method, for example a magnetic gauge against X-ray fluorescence, they may each be right about a different quantity.
Before anyone argues, write down four facts for each set of readings: the method and instrument, how it was calibrated or verified and on what reference, exactly where on the part each reading was taken, and how the readings were combined into the reported number. Most disputes resolve once those four facts sit side by side.
Why do coating thickness measurement readings disagree?
Ranked by how often they explain a dispute, the causes are: measuring different locations on an uneven film, calibrating on a reference that does not match the part, using a method that does not suit the coating and substrate pair, part geometry the probe cannot handle, and different rules for averaging and reporting. Genuine lot-to-lot variation is real but comes after these.
- Different locations. Electroplate builds up on edges and outside corners and runs thin in recesses, holes and inside corners, and sprayed powder behaves in a similar way around Faraday-cage areas. A supplier that measures a flat face and a buyer who measures inside a pocket will both be correct and still disagree. That is why plating specifications apply their minimum thickness to significant surfaces, and why the drawing has to say which surfaces those are.
- Mismatched calibration reference. Eddy-current gauges respond to the electrical conductivity of the base metal, and the method requires the conductivity of the substrate in the calibration standard and the coated part to be the same (ASTM B244, section 5.2). Zeroing on 6061 and measuring anodize on a cast alloy, or zeroing a magnetic gauge on mild steel and measuring zinc on a heat-treated alloy steel, shifts every reading.
- Wrong method for the pair. The magnetic method should not be used for electrodeposited nickel on steel, because nickel is itself magnetic (ASTM B499, section 5.2). Electroless nickel with more than 8 % phosphorus is treated by ASTM B499 as nonmagnetic enough to measure this way (ASTM B499, Note 1), but only when the reading is taken before any heat treatment; after a bake, or on a lower-phosphorus grade, the magnetic reading cannot be relied on.
- Geometry. Probes need a minimum flat area, distance from an edge, base-metal thickness and radius of curvature, and readings near those limits drift. Small turned parts, threads and thin stampings are where this bites.
- Surface texture. On blasted or rough steel a magnetic gauge reads to a magnetic plane somewhere between the peaks and valleys of the profile, so a zero taken on smooth steel and an adjustment made with a shim laid on the blasted bare surface give different numbers for the same coating. Gauge makers advise against zeroing on a blasted surface and recommend the shim adjustment instead (DeFelsko, calibration, verification and adjustment), and the paint-gauge practice requires the user to verify accuracy before, during and after use (ASTM D7091-22, section 4.2).
- Different quantities. X-ray fluorescence measures mass per unit area and converts it to thickness with an assumed coating density (ASTM B568), so an alloy coating such as zinc-nickel, or a wrong density setting, gives a number that differs from a cross section of the same spot. A micrometer before and after anodizing measures dimensional growth, not film thickness, because a large part of an anodic film grows into the metal; the anodizing dimensional change guide covers that split.
- Averaging and reporting. One side reports the mean of a sample, the other reports the lowest single reading, or one side follows the paint practice SSPC-PA 2, which sets how often to measure based on the size of the coated area and groups gauge readings into spot measurements, while the other records and judges every single gauge reading.
How do you diagnose a coating thickness disagreement between your lab and the supplier?
Measure the same marked spots on the same parts with both instruments, each verified against a reference on the same base metal, and compare spot by spot rather than average to average. If they still disagree, settle it with a destructive referee method on one sacrificed part.
A practical sequence that works for plating, anodize and organic coatings:
- Swap parts, not reports. Mark three to five locations on a few retained parts, including the significant surface and the worst recess, and have both sides read those exact spots.
- Check the zero. Read an uncoated part of the same alloy, temper and surface finish. A gauge that does not read close to zero on bare metal of the same kind will be offset on coated parts too. On blasted steel, skip the zero and check with a shim laid on the blasted bare surface instead.
- Verify with foils or traceable thickness standards in the thickness range of interest, laid on that same bare base metal, not on the reference block that came with the gauge.
- Confirm the method suits the pair using the table in the next section. A magnetic reading of electroplated nickel on steel, or an ordinary (amplitude-sensitive) eddy-current reading of a conductive coating such as electroless nickel on aluminum, is not evidence either way.
- Check geometry. Move the probe away from edges and onto the flattest area; if the reading changes noticeably, the original spot was outside the probe’s working limits.
- Go to a referee method. Cross-section one part at the disputed spot. Microscopical examination of a cross section measures local thickness directly and, under good conditions, can reach an absolute accuracy of about 0.8 µm, a figure the standard itself calls a guideline that depends on microscope setup and sample preparation (ASTM B487-24, section 1.2). For metallic coatings, the coulometric method strips the coating over a small spot of about 0.1 cm² and is limited to parts that can tolerate that bare spot or be destroyed (ASTM B504-90(2023), section 4.2).
| What you see | Most likely cause | Check that settles it | Fix |
|---|---|---|---|
| Every reading offset by a similar amount | Zero or calibration on a different base metal, alloy or surface texture | Read bare parts of the same alloy and finish; verify with foils on that metal | Name the calibration reference on the PO; retain bare reference parts |
| Averages agree, minimums do not | Different measuring locations on an uneven film | Both sides read the same marked spots | Mark measuring locations and significant surfaces on the drawing |
| Disagreement only on small, curved or thin features | Probe edge, curvature or base-metal thickness limits | Repeat on a flat area; compare with a cross section | Specify a flat measuring area or test coupons processed with the lot |
| Nickel on steel reads impossibly thick or erratic | Magnetic method used on a magnetic coating | Re-measure by XRF, cross section or coulometric method | Name a suitable method on the drawing |
| XRF and cross section differ on an alloy coating | Density or composition setting in the XRF program | Confirm the program matches the coating alloy; check against a standard | State the referee method for disputes |
| Anodize growth by micrometer is less than the gauge reading | Comparing dimensional growth with total film thickness | Eddy-current reading on a flat face of the same alloy | Specify film thickness and the dimension after coating separately |
Which coating thickness measurement method should the drawing require for each finish?
Match the method to the coating and base metal: magnetic gauges for nonmagnetic coatings on steel, eddy-current gauges for nonconductive coatings such as anodize and paint on nonferrous metals, X-ray fluorescence for metallic coatings where gauges cannot work, and a cross section or coulometric test as the referee. Conversion coatings are too thin for these gauges and are specified by class instead.
The ASTM guide to measuring metallic and inorganic coatings exists because each method has limits tied to the coating type and thickness (ASTM B659), and the table below applies those limits to the finishes that show up most on machined, cast and stamped parts. The eddy-current method states outright that chemical conversion coatings are too thin to be measured by it, so a chem film or a chromate is controlled by the specification class and, where needed, a coating-weight test, not a gauge reading.
| Coating | Base metal | Routine method | Referee in a dispute | Watch out for |
|---|---|---|---|---|
| Zinc or zinc-nickel electroplate | Steel | Magnetic (ASTM B499) or XRF (ASTM B568) | Cross section (ASTM B487) or coulometric (ASTM B504) | Thick edges, thin recesses; XRF alloy and density setting for zinc-nickel |
| Electrolytic nickel, copper-nickel-chrome | Steel | XRF (ASTM B568); ASTM B530 for nickel | Cross section or coulometric | ASTM B499 magnetic method not suitable for nickel on steel |
| Electroless nickel, high phosphorus (above 8 % P) | Steel | Magnetic before heat treatment, or XRF | Cross section | Only before heat treatment; ASTM B499 covers only deposits above 8 % P |
| Electroless nickel | Aluminum | XRF | Cross section | Coating is conductive, so the eddy-current method in ASTM B244 does not apply; phase-sensitive eddy-current instruments are an alternative |
| Anodic oxide (Type II, Type III) | Aluminum | Eddy current (ASTM B244) | Cross section | Calibrate on the same alloy; gauge reads total film, not dimensional growth |
| Chem film, chromate or trivalent conversion | Aluminum, zinc | Not measurable by these gauges | Per the coating specification | Control by specification class, not thickness |
| Powder coat or paint | Steel | Magnetic gauge per ASTM D7091 | Agreed on the PO; cross section of a coupon | Blast profile (adjust with a shim on the blasted bare surface); measurement frequency per SSPC-PA 2 |
| Powder coat or paint | Aluminum or zinc, including die castings | Eddy-current gauge per ASTM D7091 | Agreed on the PO; cross section of a coupon | Calibrate on the same alloy; anodize or pretreatment under the paint adds to the reading |
What should the drawing, RFQ and PO say so coating thickness readings stop disagreeing?
Give the thickness as a minimum or a range on named significant surfaces, name the measurement method and the referee method, and say how many readings are taken, where, and how they are combined. Then say whether the thickness applies before or after any topcoat, seal or heat treatment.
A workable set of notes, to adapt rather than copy:
- Specification and class. Cite the coating specification with its class or type, so the thickness is not invented separately; the nickel plating guide and the B633 guide show typical callouts.
- Significant surfaces. Mark the faces where the minimum applies, and say that holes, threads and recesses outside them are exempt or carry a lower minimum. Without this, the thinnest recess sets the rejection.
- Measurement method. “Thickness by magnetic method per ASTM B499” or “by eddy current per ASTM B244”, chosen from the table above.
- Referee method. “In case of dispute, cross section per ASTM B487 governs,” with the number of parts that may be sacrificed.
- Calibration reference. Ask for calibration or verification on uncoated parts or coupons of the same alloy and surface condition, and keep a few bare parts for that purpose.
- Measuring plan. Number of parts per lot, number and location of readings per part, and whether the requirement applies to each reading or to the mean. A ballooned sketch of the measuring spots removes most ambiguity.
- Sequence. State whether thickness is measured before or after chromate, seal, topcoat or bake, since each changes what a gauge sees.
- Report. Ask for the individual readings, the instrument, and the verification record, not only a pass statement.
For small or complex parts that cannot be read reliably, specify test coupons of the same material processed in the same load, and say so on the PO so both sides measure the same thing.
How is coating thickness measurement handled on a Saola order?
On a Saola order, parts are machined at our lead CNC plant and coated at a partner finishing plant we manage, and the quote names the production site and country of origin for each step. Thickness is checked against the method, locations and acceptance rule on your drawing and purchase order, so the more of the notes above you include, the less room there is for a dispute.
CNC turning and milling run at our lead CNC plant, T.K Viet Nam Precision Company Limited, in operation since 2017, as described on the CNC milling page. Plating, anodizing and powder coating are carried out by partner plants we manage, as listed on the surface finishing page; those partners are not named and are not necessarily in Vietnam, which is why each quote states where the work is done.
In practice, send the coating callout with the significant surfaces, measuring method, referee method and measuring plan, and say whether you want thickness readings reported with each lot. We pass those requirements to the finishing plant as written and ask for the readings in the form your purchase order sets. If your readings disagree with a report, send the spots, instrument and calibration reference you used, and we take it up with the plant that did the work using the checks in this guide. ISO 9001 certification is in progress, not held. Send the drawing and finish notes through the quote form.
Frequently asked
Which coating thickness measurement method is most accurate?
For local thickness at one spot, a well-prepared metallographic cross section is usually treated as the reference, which is why it is the common referee method. It is destructive and measures only where it is cut, so routine acceptance still relies on nondestructive gauges or XRF verified against it.
Can one gauge measure both zinc plating on steel and anodize on aluminum?
Many combination gauges carry both a magnetic and an eddy-current mode and switch automatically on the base metal. Each mode still needs its own verification on the right base metal, and neither mode suits electroplated nickel on steel or conductive coatings such as electroless nickel on aluminum; those need XRF, a phase-sensitive eddy-current or nickel-specific instrument, or a cross section.
Why does the plating thickness look fine on the certificate but threads do not gauge?
The certificate usually reports a flat significant surface, while plate builds up on thread crests and edges. Thread fit after plating is a separate requirement; the threads after anodizing and plating guide covers allowances and how to call them out.
Should the drawing specify plating thickness as a minimum or a range?
Corrosion protection depends on the minimum on significant surfaces, so plating specifications are written as minimums. Add a maximum only where fit demands it, and then make sure the fit tolerance allows for the uneven build of the film.
Who pays when a referee cross section proves the lot was in specification?
Agree this on the purchase order before the first lot: one arrangement is that the party whose readings were wrong carries the cost of the referee test and any sacrificed parts.
Sources
- ASTM B659-90(2021), Standard Guide for Measuring Thickness of Metallic and Inorganic Coatings (scope and significance), ASTM International, 2021
- ASTM B499-09(2021)e1, Measurement of Coating Thicknesses by the Magnetic Method: Nonmagnetic Coatings on Magnetic Basis Metals (scope and significance), ASTM International, 2021
- ASTM B244-09(2021), Measurement of Thickness of Anodic Coatings on Aluminum and of Other Nonconductive Coatings on Nonmagnetic Basis Metals with Eddy-Current Instruments, ASTM International, 2021
- ASTM B568-98(2021), Measurement of Coating Thickness by X-Ray Spectrometry, ASTM International, 2021
- ASTM B487-24, Measurement of Metal and Oxide Coating Thickness by Microscopical Examination of Cross Section, ASTM International, 2024
- ASTM B504-90(2023), Measurement of Thickness of Metallic Coatings by the Coulometric Method, ASTM International, 2023
- ASTM D7091-22, Nondestructive Measurement of Dry Film Thickness of Nonmagnetic Coatings Applied to Ferrous Metals and Nonmagnetic, Nonconductive Coatings Applied to Non-Ferrous Metals, ASTM International, 2022
- Calibration, Verification, and Adjustment of coating thickness gages, DeFelsko Corporation, accessed October 2026
- Coating thickness measurement methods (magnetic induction, amplitude- and phase-sensitive eddy current, XRF, coulometric), Helmut Fischer, accessed October 2026