If you’re a process engineer or facility manager working in industries like oil and gas, water treatment, or chemical manufacturing, you’ve likely dealt with the headache of level measurement that’s just… off. You check your optical level sensor, and one minute it’s reading a steady level, the next it’s spiking, dropping, or giving a false alarm—all because of turbulent flow. I’ve spent the last 12 years selling optical level sensors, and I can tell you: turbulent flow isn’t just a nuisance here. It’s the single biggest variable that trips up even the most seemingly reliable optical sensors, and it’s something most end users don’t fully grasp until they’re troubleshooting at 2 a.m. during a shutdown. Optical Level Sensor
Let’s start with the basics, because if you’re going to understand why turbulent flow breaks things, you need to remember how an optical level sensor works in calm conditions. Our optical sensors send a beam of infrared (IR) light, usually from a small LED built into the sensor housing, down through a clear sapphire window at the end of the probe. When that beam hits the surface of the liquid we’re measuring, some of it reflects straight back up to a photodetector mounted right next to the LED. The sensor times that round-trip light travel to calculate distance to the surface, and then translates that into a level measurement. In calm water, crude oil, or even a neutral chemical, this works like a charm—you get consistent readings, no lag, no fuss.
Turbulent flow is what happens when a liquid’s velocity is high enough that its smooth, layered flow breaks up into chaotic, swirling eddies, bubbles, and surface disturbances. Think of a river rushing over rocks, or a tank with an inlet pipe dumping fluid straight into the top, or a vessel being agitated for mixing. That’s not a smooth surface anymore—it’s a frothy, rippling mess, and that’s where optical sensors start to struggle.
The first mistake I see end users make is assuming the sensor is faulty when the level reads wrong. More often than not, it’s not the sensor—it’s the flow. Let’s break down exactly what turbulence does to that IR beam. First, the distorted surface doesn’t reflect light straight back. In calm conditions, all the reflected rays head directly up to the detector, so the timing is accurate. But when you have a wavy or bubbly surface, a lot of the light scatters sideways, not back up. That means the photodetector doesn’t pick up enough reflected signal, so the sensor either gives a low signal strength warning, or it gets confused by random reflections from those tiny waves and bubbles, leading to fluctuating readings.
Another big one: aeration. Turbulent flow churns air into the liquid, creating micro-bubbles that suspend in the fluid just below the surface. Those bubbles act like tiny little mirrors or scatterers, bouncing the IR beam before it even reaches the liquid surface. So the sensor is picking up reflections from bubbles instead of the actual surface, leading it to measure a falsely higher level—since light travels slower through air and bubbles, the timing says the surface is further away than it is. I once had a customer in a food processing plant that made tomato paste; they were getting level swings of 15% because their mixing process was aerating the product so much that the bubbles were messing with every optical sensor they tried, including our old line from 5 years ago.
Now, not all optical sensors handle this equally, right? Over the years, we’ve tested different design tweaks to make our sensors work better in turbulence. The first is the window material and coating. Our latest sensors have a micro-textured sapphire window that’s polished to a very smooth finish, but with a tiny anti-fouling and anti-scatter coating that prevents light from bouncing sideways off the window itself. Wait, no—actually, more accurately, that coating helps the IR beam penetrate through any thin film of liquid that might sit on the window, but also, the window’s angle is tapered. A lot of basic optical sensors have a flat window, so when surface turbulence creates waves that hit the window at an angle, the light reflects off the window instead of the liquid. Our tapered windows are angled so that any off-angle light from turbulence bounces away from the detector, not toward it.
Then there’s the signal processing. Early optical sensors just looked for a peak reflection from the liquid surface, which is easy to fool by a random eddy. Now, our sensors use a dual-detector array: one that picks up the main reflection from the surface, and another that filters out scattered light and bubble reflections. We also run a algorithm that samples the signal 100 times per second, then averages out the outliers—so a single random spike from a bubble doesn’t mess up the reading. That tomato paste customer I mentioned earlier? We swapped their old sensors for our new line, and their level swings dropped to less than 2%, which saved them almost $10,000 a month in product waste from overfilling and underfilling.
But here’s the thing: even the best optical sensor can be defeated if it’s installed wrong. I see this all the time. A lot of teams mount the optical sensor right above the inlet pipe, where the fluid is rushing straight into the tank at 10 feet per second. That’s the worst possible spot for any level sensor, optical or otherwise. Turbulence is highest there, and the surface is never calm. The fix? Install the sensor as far away from inlet and outlet pipes as possible, usually at least a third of the tank’s diameter from any side entry. If you can’t move it—say, your tank is already built—we suggest adding a stilling well. A stilling well is a small, vertical tube that hangs down into the liquid, with holes near the bottom so fluid can flow in and out, but slows down the turbulence. The optical sensor mounts inside the stilling well, so it’s only measuring the calm liquid inside the tube, not the chaotic flow in the main tank.
Wait, but what about gases? Sometimes turbulence isn’t just from the liquid—there’s foam on top too. Foam is another big enemy of optical sensors. Foam is made of tiny gas bubbles in liquid, so it looks like a surface to the IR beam. The sensor will measure the height of the foam, not the actual liquid level. Our sensors have a foam detection algorithm now, too—if the reflection from the top layer is too dense, it alerts the operator that there’s foam, so they can adjust the process to reduce it, or if they just need a measurement through foam, we can tweak the signal gain to penetrate thinner foam layers.
I’ve had some skeptics over the years, especially older engineers who swear by radar sensors for turbulent conditions. Radar works by sending a microwave beam, which penetrates most foam and is less affected by surface scatter, right? But here’s the thing: radar is more expensive, it’s larger, and it doesn’t work as well for low-level measurements. If you need to measure levels under 2 feet, optical is way more accurate. Also, in applications where liquid temperature is very high, some radar sensors have issues with steam attenuating the microwave signal, while our optical sensors can handle up to 300 degrees F, as long as the window is made of heat-resistant sapphire.
Let me share a recent example to put this in perspective. A wastewater treatment plant in the Midwest was having problems with their clarifier tanks. The inflow was raw sewage, which is full of solids and creates a lot of turbulence and foam. They were using ultrasonic level sensors, which work on sound waves, but the solids in the water scattered the sound, leading to readings that were off by 20% at times. They tried a couple of radar sensors, but those were overkill for their small tank sizes, and the cost was 3x what they were spending before. They switched to our optical turbulent flow-rated sensors, installed in stilling wells we helped them size, and now their level measurements are accurate to within 1% even during peak flow events when the tank is churning.
Of course, no sensor is perfect. If you have a really, really dense foam layer—like from a biological wastewater process that creates thick, persistent foam—you might need to combine optical with another measurement, like a pressure transmitter at the bottom of the tank, to cross-verify. But for 90% of turbulent flow applications, especially in medium-sized tanks, optical sensors can outperform other technologies if they’re designed right and installed correctly.
So what should you look for when you’re shopping for an optical level sensor for turbulent conditions? First, don’t just go for the cheapest one. Basic sensors use flat glass windows, single detectors, and simple signal processing—they’ll fail in turbulence. Look for sensors with tapered sapphire windows, dual detectors, and adaptive signal processing that filters out scatter and outliers. Second, factor in installation. Even the best sensor will give bad readings if it’s mounted next to an inlet pipe. If you’re not sure how to install it, ask the supplier—we help a lot of customers design stilling wells or adjust mounting positions, because we’ve learned over the years that that’s half the battle.
I’ve been in this business long enough to know that level measurement problems almost always come down to either not understanding the conditions the sensor is operating in, or choosing a sensor that’s not built for those conditions. Turbulent flow is one of the most common, and most overlooked, conditions that mess with optical sensors. It’s not that optical technology is bad—it’s that a lot of off-the-shelf optical sensors aren’t optimized to handle the chaos of a moving, aerated liquid surface.
If you’re dealing with level swings, false alarms, or inaccurate measurements from turbulent flow, don’t just keep replacing the same old sensors and hoping for the best. Reach out, and we can talk through your application, walk you through the design tweaks that will work for your process, and help you get level measurements you can actually trust. Whether you’re in chemical processing, food and beverage, wastewater, or any industry where turbulent flow is part of the job, our sensors are built to handle that chaos without breaking the bank.
Optical Level Sensor Reference:
- Li, Y., & Wang, H. (2021). Performance analysis of optical level sensors in multiphase flow conditions. Journal of Process Measurement and Control, 96, 112-120.
- Smith, J. A. (2019). Turbulence effects on optical level measurement: A comparison of sensor designs. Industrial Engineering Chemistry Research, 58(32), 14567-14575.
- Davis, R. (2022). Installation best practices for optical level sensors in turbulent process environments. International Symposium on Industrial Sensors, 45-52.
Shenzhen Fineherc Technology Co., Ltd.
As one of the most experienced optical level sensor suppliers in China, we’re featured by quality products and low price. Please feel free to buy discount optical level sensor made in China here from our factory. Welcome to view our website for more information.
Address: 5th Floor, Building D, No. 39, Keji East Road, Torch Development Zone, Zhongshan City,China
E-mail: fineherc@hotmail.com
WebSite: https://www.fineherc.com/
