In fields such as construction, bridges, and anti-corrosion engineering, the thickness of protective coatings applied to cement substrates directly affects their service life and performance. As a non-destructive testing tool, coating thickness gauges are widely used both on-site and in laboratories. However, one thing is often overlooked: the surface roughness, porosity, and water absorption of cement substrates are completely different from those of metal substrates, which sometimes makes traditional electromagnetic or eddy current thickness gauges inadequate. Because ultrasonic coating thickness gauges offer material tolerance, they have become a solution for this type of demand in recent years. However, there are some minor pitfalls when applying, so let's dive deeper below.
Principle overview
The core of the ultrasonic coating thickness gauge is to emit high-frequency sound pulses that pass through the coating, reach the substrate interface, and then reflect back. The instrument calculates the round-trip time of sound waves and combines it with preset sound speed values to determine the thickness. The formula can be written as:
Thickness = Speed of sound × Time ÷2
Written in HTML style, it is:d = v × t / 2。 Where d is the thickness, v is the speed of sound, and t is the round-trip time.
This principle is mature for metal substrates, but the challenge arises for cement substrates: cement itself is not a homogeneous dense material; it contains micropores and aggregate particles inside, causing sound waves to partially scatter and attenuate. Additionally, the coating may be paint, asphalt, epoxy, or polymer mortar, with significant differences in the speed of sound for each material. More importantly, the surface of the cement substrate often has floating ash or unevenness, affecting coupling. To put it bluntly, if the ultrasonic probe doesn't fit tightly, the data tends to float.
The special nature of cement substrates
Cement substrates are roughly divided into concrete, mortar, and cement boards. What they have in common is that their surface roughness is usually in the tens of microns to millimeters, unlike metals that are smooth. For example, a common concrete sidewalk slab has a surface that feels like sandpaper. In this case, if the dry-coupled probe is used directly, it is difficult for the ultrasonic signal energy to be effectively delivered to the coating. Many technicians are accustomed to using coupling agents (such as glycerol or gel) to fill the gaps, but this introduces another layer of medium that affects the wave propagation path.
Additionally, the acoustic impedance of cement substrates is about 7-10 MRayl, while coating materials (such as epoxy) have 2-3 MRayl. The difference between the two is significant, and the reflected signal is strong, which is advantageous. However, the aggregates inside the cement base (such as gravel) can form irregular reflections, interfering with signal interpretation. To be honest, some ultrasonic thickness gauges mistakenly interpret the undulations or internal bubbles of the base layer as interfaces when measuring the coating on cement. The solution is usually to adjust the gain or use a probe with filters.
For example: an outdoor steel structure has epoxy paint on the surface and concrete underneath. Inspectors found large fluctuations in the data, later discovering that the small holes on the concrete surface were not fully sealed, with small gaps under the coating. A magnifying glass makes it clearer: these gaps are "pinholes" that have not been filled with primer. After repolishing, the readings became much more stable.
Testing methods and steps
Here, I won't memorize the manual, but I'll share a few practical experiences. First, clean the surface. Blow away the dust with a brush or compressed air, and wipe with alcohol if necessary, but be careful not to dissolve the coating. Second, select the right coupling agent. It is recommended to use high-viscosity glycerin or specialized ultrasonic coupling adhesive, applying it evenly and not too thickly. I've seen someone apply a layer of butter, but the ultrasonic waves are completely absorbed by the butter, and the reading drops to zero. Then, set the speed of sound. Different coating materials have different speed of sound: polyurethane about 2400 m/s, epoxy about 2650 m/s, asphalt about 1800 m/s. It's best to calibrate a sample of the same material with known thickness—commonly referred to as a "piece" or test block.
During detection, the probe should be pressed vertically and held steady for about 3 seconds. For rough surfaces, pressure can be appropriately increased, but local deformation of the coating under pressure should be avoided. Here's a little trick: measure points at the edges or corners of cement slabs, because the substrate there is denser and the reflected signal is cleaner. But don't rely solely on these points; the specification requires at least 5 points to be averaged. Some standards, such as ISO 2808 or ASTM D6132, also mention similar practices.
Factors affecting accuracy
Several key points: First, temperature. Cement substrates can experience a 1-2% change in sound speed in winter and summer, and the same coating may harden at low temperatures, increasing the speed of sound. Second, the coating thickness range. Ultrasonic methods are usually suitable for coatings ranging from 20 microns to a few millimeters; coatings that are too thin (e.g., less than 10 microns) or too thick (over 8 millimeters) are prone to errors. Third, the moisture content of the grassroots layer. The sound speed of wet cement substrates is about 5% lower than that of dry substrates, which requires special attention in humid environments. In one test, the same curing board showed a coating thickness of 300 microns in wet conditions, but after two days of drying, it was 285 microns—a difference of 15 microns. This difference was because the moisture content of the substrate affected the wave velocity.
Another easily overlooked stage: the curing stage of the coating. The freshly applied coating is not yet fully hard, and some of the energy as sound waves pass through is absorbed by viscoelasticity, resulting in higher measured values. It is best to inspect after the coating has fully cured. But if rapid evaluation is needed on site, a layer of polyester film can be applied to the coating surface to assist coupling—don't joke, some people have actually done this, and the results are quite good.
Case analysis
Case 1: Anti-corrosion coating for a certain bridge pier. The coating consists of two layers of epoxy + topcoat, with a total design thickness of 350 microns, and the base material is C40 concrete. During detection, two types of probes were used (one at 5MHz, one at 2MHz). The 5MHz probe reads stably in the smooth region, but the data oscillates significantly in the exposed coarse aggregate region. Although the 2MHz probe has strong penetration, its resolution is insufficient, resulting in large errors when measuring thin coatings. Based on the data from the 5MHz probe in the polishing area, the average thickness measured was 375 microns, slightly above the design value, which is still acceptable.
Case 2: Testing of fireproof coatings for cement boards. The thickness of fireproof coating must be 2mm, and ultrasonic thickness measurement is used. Because the coating is loose and porous, causing severe sound attenuation, after multiple attempts, a lower-frequency probe (1MHz) was switched to and the amount of coupling adhesive increased. The measured data ranged between 1.8-2.2mm, and compared to the destructive method sampling profile, the deviation was within ±0.1mm. However, operators need to note that sometimes fibers are added to fireproof coatings, which can cause sound wave scattering—those anomalies were later found to be caused by fiber bundles.
Case 3: A layer of epoxy self-leveling on an indoor cement floor. The standard requires a thickness of 1.5mm. The inspection started smoothly, but at the corner of the baseboard, the data suddenly dropped to 1.1mm. Upon re-examination, it was found that the corner coating had bubble interlayers, and the ultrasonic waves reflected both the upper surface and the bubble interface—this was actually a measurement trap. Later, a wet film thickness gauge was used for further verification before confirmation.
FAQs
Problem 1: Unable to read the numbers. Possible causes: poor coupling, coating too thin, or probe not suitable for high-attenuation materials. Solution: Replace the coupling, increase the gain, or use a low-frequency probe.
Problem 2: Large fluctuations in readings. It may be due to surface roughness, internal interlayer delamination, or probe tilt. Besides repeatedly testing several times to remove abnormal points, sometimes testing from a different direction can also improve the situation. I remember once, my colleague pressed the probe diagonally, and the thickness readings fluctuated up and down, as if drunk—but later it straightened up and became normal.
Question 3: The coating has similar ultrasonic characteristics to the substrate. For example, when a cement-based polymer is coated on a cement substrate, the difference in sound speed or impedance is small, and the reflected signal is weak. At this point, you can try using a higher-frequency probe, increase signal amplification, or simply switch to magnetic or mechanical methods for verification.
Compared to other methods
Compared with traditional magnetic thickness measurement methods, the advantage of ultrasonic methods lies in their non-relying on the magnetic or electrical conductivity of the substrate. The magnetic method is only suitable for ferromagnetic metals and cannot work at all on cement substrates. Although eddy current can test non-magnetic coatings, the substrate must be a conductive metal; cement is non-conductive, so it is also not acceptable. The ultrasonic method is suitable for almost any solid substrate, but it is more difficult to operate and has greater environmental impact. Also, ultrasonic equipment is usually a bit more expensive, but a single device can test multiple substrates, offering good cost-performance across diverse scenarios.
Compared to destructive testing (such as microscopy or micrometer measurement), ultrasonic methods do not damage coatings and are suitable for batch sampling. However, the destruction method can be used as a semi-quantitative calibration method, especially when the coating's speed of sound is unknown. If you want to save money, you can simply take a vernier caliper and measure the total thickness at the coating edge minus the substrate thickness—but this only works for smooth substrate edges and not for most irregular templates.
Ultrasonic coating thickness measurement is not impossible on cement substrates, but it is more troublesome than with metals. The key lies in understanding the substrate properties, selecting the right probe, and setting parameters. Some users prefer one-touch foolproof operation, but cement substrates often require some manual experience and patience. In the future, as probe technology and signal processing advance, the instrument may automatically compensate for these differences. But at least for now, testing several times, changing angles, occasionally checking the surface with a magnifying glass is not supersensible. Sometimes, these trivial experiences are much more useful than manuals.