How to Verify a Finishing Vendor’s Coating Thickness

A coating thickness reading is only as good as the method behind it. A vendor may report an average of 150 microns, but that number can hide a 50 micron void on the inside corner of a flange. As a buyer, you need to confirm that the coating meets your spec across the entire part, not just at the points the vendor chose to measure.

This guide walks through the practical steps to verify a finishing vendor’s coating thickness. It covers instrument selection, sampling strategy, and documentation requirements.

Choose the Right Instrument for the Coating

The first decision is the measurement method. Different coatings and substrates respond differently to various instruments.

Instrument	Best For	Limitations
XRF (X-ray Fluorescence)	Paint, plating, e-coat	Requires calibration for specific coating stacks
Eddy Current	Non-ferrous substrates with metallic coating	Not effective on non-conductive coatings
Manual Micrometer	Thick coatings, non-metallic substrates	Destructive or semi-destructive
Pin Gauge	Porous coatings, thick films	Can damage the surface if misused

XRF is the most common choice for painted and plated parts. It works by firing X-rays at the surface and measuring the fluorescence emitted by the coating. The machine calculates the thickness based on the energy loss of the X-rays as they penetrate the coating and return from the substrate. It is fast and non-destructive. This speed allows operators to check multiple points on a large panel during a production run without stopping the line. However, XRF requires accurate knowledge of the substrate material. If the substrate is steel but the instrument is set for aluminum, the reading will be incorrect.

Eddy current is preferred for aluminum or copper substrates where the coating is metallic, like nickel or zinc. It measures the change in the electrical resistance induced in a coil near the surface. Because it relies on conductivity, it works well on metal-on-metal coatings. It struggles with non-conductive layers such as epoxy or powder coat.

Manual methods are slower but useful for thick films or when electronic instruments are not available. A micrometer clamp is effective for checking the total thickness of a bracket where the coating is applied over a non-metallic core. A pin gauge is a small, spring-loaded tool with a sharp tip. You press it into the coating; if it breaks through, you measure the depth of the penetration. This is a destructive test, so it should be used at the end of the inspection sequence or at sacrificial areas.

Selecting the wrong instrument leads to false readings. For example, an eddy current gauge may not work accurately on a non-ferrous substrate if the coating is too thick. The signal weakens as the distance increases, and the instrument may hit its measurement limit. Always confirm the instrument is calibrated for your specific coating stack. If you are buying a part with a zinc plating over a nickel barrier, ask the vendor which instrument they use and how they account for the multi-layer structure.

Define Sampling Points and Limits

Before measuring, agree on where the vendor will measure the coating. A single average reading is not enough. Specify a pattern of points on the drawing.

For a flat panel, measure at the center and each corner. For a complex part, measure at the inside corners, outside corners, and along the centerlines. The inside corners are often the thinnest areas because the coating flows away from the tightest radius. In dip coating, the liquid drains from the part as it is lifted. The liquid level rises slightly above the part, but the film thins as it drains. The tightest internal angles shed the most liquid, resulting in the thinnest film.

Set minimum and maximum limits in the purchase order. If the spec calls for 100 microns, define the acceptance window. Is it 95 to 110 microns? Or is the minimum 100 microns with no upper limit? Clarify whether the average or the minimum point matters. A part with one spot at 98 microns and five spots at 150 microns may fail if the spec requires a minimum of 100 microns at all points.

Write these requirements into the work instruction. Ambiguity here leads to disputes later. For instance, if you do not specify the number of points, a vendor might measure only three points per part. If you do not define the acceptance logic, they might report an average that passes while a critical spot fails. Be explicit. State the number of points per part, the location of each point, and the pass/fail criteria for each individual reading.

Verify Calibration and Equipment

A gauge that is out of calibration gives a false sense of security. Ask the vendor for their calibration records. The equipment should be calibrated at regular intervals, often annually or before each project.

Check the calibration certificate. It should list the date, the standard used, and the person who performed the check. If the vendor cannot provide this, request a witness test. Bring your own calibrated reference block and have the vendor measure it. If the reading matches your standard, the gauge is likely in tolerance.

Calibration drift is a common cause of failed coatings. A gauge that reads 10 microns low may pass a part that is actually below spec. This is why calibration verification is a non-negotiable step in any finishing vendor audit.

Consider the environment as well. Dust and debris on the probe head can affect XRF readings. Eddy current probes need to be kept clean to ensure proper coupling with the part. If the vendor operates in a dusty environment, ask how they clean the probes between measurements. A dirty probe can register a higher thickness than reality because it is measuring the coating plus the layer of dirt.

Perform the Measurement

Once the instrument and sampling points are set, the vendor performs the measurement. The operator should record each reading. For XRF, the machine often prints a report. For manual gauges, a technician writes the number on a log sheet.

Look for consistency. If the readings jump wildly across a flat area, the coating may be uneven. If the readings are too uniform, the operator may be measuring the same spot repeatedly. A good operator moves the probe slightly between readings to average out surface texture. Surface roughness can cause variations in the reading, especially with thin films. Moving the probe helps average these micro-variations.

For e-coat or powder coat, check for holidays. A holiday is a pinhole that exposes the substrate. XRF cannot always detect a holiday if the coating is thick enough. The X-ray energy may penetrate the coating and hit the substrate, but the instrument might not flag the defect if the surrounding area is thick enough to compensate. Use a pin gauge or a visual inspection with a high-magnification lamp to check for voids.

When using a pin gauge, apply consistent pressure. Too much force will push the needle through, but too little may not penetrate. The operator should use the same technique for every part to ensure fair comparison. Record the location of any holiday found. This information is critical for rework planning.

Document the Results

The measurement data must be documented. A verbal confirmation is not enough. Request a certificate of conformity that lists each measurement point and the reading.

The certificate should include the part number, batch number, date of inspection, instrument used, and the name of the inspector. It should also state the acceptance limits and the result for each point.

Keep this document in your quality records. If the part fails in service, you will need to trace the issue back to the finishing process. The certificate is your proof of what was measured and by whom.

Include the raw data if possible. For XRF systems, the machine often stores the data in a file. Request a copy of this file. It shows the exact readings before any software averaging or filtering. This protects you if the vendor later claims the data was manipulated.

Review and Approve

After the vendor submits the data, review it against your spec. Check the minimum points, not just the average. Look for trends. If the inside corners are consistently lower than the outside, the process may need adjustment.

If the data meets the spec, approve the lot. If it fails, request a rework or a replacement. Do not accept an average that is within spec but has a failed minimum point. The part will not meet the functional requirement.

A final verification step is to spot check a small number of parts with your own instrument if possible. This adds a layer of independence to the vendor’s data. If you do not have the equipment, rely on the calibration witness test and the raw data file.

Common Mistakes

• Relying on a single average reading. This hides thin spots and voids.
• Skipping the calibration check. A drifted gauge gives false confidence.
• Not defining sampling points. The vendor will measure where it is easiest, not where it is most critical.
• Ignoring holidays. XRF may not detect pinholes. Use a pin gauge for thick coatings.
• Accepting verbal confirmation. Always require written documentation.

Final Verification

The final step is to close the loop. Keep the certificate of conformity and the raw data in your project file. If the part is critical, consider a third-party inspection. An independent lab can re-measure a sample and confirm the vendor’s results.

Coating thickness verification is not just a number. It is a system of controls. Get the instrument right, define the points, check the calibration, and document everything. This approach protects your product and your supply chain.