If you’ve ever looked at a smartphone’s face unlock system, a LiDAR sensor on a drone, or a proximity detector in a laptop, you’ve encountered a 940nm VCSEL—Vertical-Cavity Surface-Emitting Laser—at work. As a supplier specializing in these devices, one of the most common questions I get from engineers designing these systems is: “What’s the optical power density of your 940nm VCSELs?” It’s not a trivial question. Power density dictates how much light hits a target at a given distance, how well a sensor can pick up a signal, and even how safe the device is for end users. Today, I’ll break down what optical power density means for 940nm VCSELs, what numbers you can expect, and why choosing a reliable supplier makes all the difference in hitting your performance and compliance goals. 940nm VCSEL
First, let’s get the basics straight. Optical power density is simply the amount of light power per unit area, usually measured in watts per square centimeter (W/cm²). For VCSELs, this is different from total output power: a VCSEL array with four chips each putting out 100mW has a total power of 400mW, but if that array spans 0.04 cm² (about 2mm squared), its power density is 10W/cm². That distinction matters because it’s not just how much light you generate, but how tightly you can (or need to) focus it. For 940nm specifically, that wavelength sits in the near-infrared (NIR) band, a sweet spot for consumer electronics because it’s invisible to the human eye, penetrates skin and fog well, and is absorbed by common NIR sensors like silicon photodiodes without excessive noise.
Now, why does power density for 940nm VCSELs vary so much? It depends on three core factors: the VCSEL’s structure, operating conditions, and whether you’re looking at a single emitter or an array. Let’s start with single emitters. Early 940nm VCSELs, developed for low-speed applications like barcode scanners, used small, narrow active regions to confine light. Those maxed out around 5–10mW total power, with a power density of 1–2 W/cm². But today’s modern 940nm VCSELs, optimized for high-speed data transfer or facial recognition, have larger active regions—some as big as 50 microns in diameter—so they can push higher power without overheating. A typical single high-performance emitter from a leading supplier now delivers 50–100mW, translating to a power density of 10–20 W/cm². That’s enough for short-range sensors, but it’s the arrays that really drive the high power densities seen in modern systems.
Most commercial applications don’t use single emitters; they use arrays made up of dozens or hundreds of individual VCSELs. For example, a 1×12 VCSEL array common in face unlock systems has 12 emitters, each with 60mW of power. The array’s total power is 720mW, and its active area is usually around 0.05 cm², so power density jumps to ~14 W/cm². But high-power applications like long-range LiDAR need much denser arrays. Some advanced 940nm VCSEL arrays, designed for automotive LiDAR, pack hundreds of emitters into a small area, with each emitter putting out 150mW or more. Those can reach power densities of 50–100 W/cm², and in lab prototypes, we’ve even seen densities as high as 200 W/cm² for pulsed operation. That’s where it gets critical: continuous wave (CW) vs. pulsed operation changes power density too. For many VCSEL applications, they’re used in pulsed mode—flashing light for nanoseconds at a time—instead of constantly on. That lets you push much higher power densities without overheating the device, since there’s less time for heat to build up. A 940nm VCSEL can hit 100 W/cm² in pulsed mode, but if you ran it constantly, you’d risk thermal damage and efficiency drops.
But power density isn’t just a performance number—safety regulations govern it too. The International Electrotechnical Commission (IEC) has a standard called IEC 60825 for laser safety, which classifies lasers by their power density and wavelength. For 940nm, which is in the NIR, exposure to power densities above a certain threshold can cause permanent eye damage. Class 1 lasers, the safest category for consumer electronics, require power densities below 10 W/cm² for CW operation. That’s why a lot of commercial face unlock systems use Class 1 940nm VCSEL arrays—they hit the required performance without risking user safety. A common mistake I see from design engineers is chasing higher power density without checking safety compliance. Just because a VCSEL can output 50 W/cm² in lab tests doesn’t mean it’s suitable for a phone; it’ll fall into Class 3B, which requires warning labels and is too risky for daily use near users.
As a supplier, we’ve seen firsthand how different customers have different power density needs. A startup making a smart doorbell with a proximity sensor might only need a low-power 940nm VCSEL with a power density of 2–5 W/cm²—cheap, efficient, and compliant. But an automotive manufacturer designing a LiDAR system for self-driving cars needs higher power density, 40–60 W/cm² in pulsed mode, to detect objects 100 meters away. The challenge for suppliers like us is tailoring VCSEL arrays to those specific needs, not just selling a one-size-fits-all part. For example, we recently worked with a drone maker that needed a 940nm VCSEL array with uniform power density across all emitters—if one emitter was much more powerful than the others, it would create hot spots that damaged the array over time. We adjusted the array’s epitaxial structure and packaging to keep power density consistent within 5% across all 120 emitters, which solved their reliability problem.
Another point people often overlook is how packaging affects power density. A VCSEL’s power density is measured at the chip level, but when you package it, you have to account for the lens or window over the top of the device, any heat sink, and how the array is mounted. A poorly packaged VCSEL array might have the same chip-level power density as a competitor’s, but because heat can’t escape, it will run hotter, leading to lower actual power output and shorter lifespan. At our facility, we test every 940nm VCSEL for both chip-level and packaged power density, because we know that a number on a datasheet means nothing if it doesn’t hold up in real-world use. We also test for aging: after 10,000 hours of operation, a VCSEL’s power density should stay within 10% of its initial value, which is critical for long-term applications like automotive.
Let’s also address common misconceptions. Some engineers think 940nm VCSELs have a fixed power density, but that’s not true. It depends on drive current—increase the current to a VCSEL, and power goes up, so power density goes up, but only to a point. Too much current, and the VCSEL will reach its thermal limit, leading to efficiency drops and even failure. Our datasheets always include a power vs. current curve, so engineers can pick the right drive current for their desired power density without pushing the device too hard. For example, a 50 micron emitter from our line peaks at 120mW when driven at 2A, which gives a power density of ~15 W/cm², but running it at 1A gives 40mW and a power density of ~5 W/cm², which is better for low-power, battery-powered devices.
What about emerging applications? Things like AR/VR headsets are starting to use 940nm VCSELs for eye tracking, and those need very precise power density—enough to track the eye’s movements without being so powerful that it causes eye strain. Right now, we’re developing a new array for AR eye tracking that has a power density of 3–7 W/cm², optimized to focus tightly on the eye’s cornea while staying within safety limits. That’s a balance we have to strike every day: balancing performance, safety, and longevity.
Now, if you’re an engineer working on a project that uses 940nm VCSELs, you’re probably looking for a supplier you can trust to give you accurate, reliable power density numbers. A lot of cheap suppliers will publish datasheets with power densities that can’t be replicated in real use, because they don’t control the epitaxial growth or packaging process. At our company, we grow our own VCSEL epitaxial wafers in-house, test every chip at multiple drive currents, and qualify all our arrays for temperature and lifetime testing. We don’t cut corners on safety or performance, because we know that if your VCSELs don’t work as expected, your whole product is at risk.
If you’re designing a face unlock system, LiDAR, proximity sensor, or any other application that uses 940nm VCSELs, and you need to know exactly what power density you need, or you want to test our parts for your design, get in touch with our team. We can provide sample arrays, custom power density profiles, and help you navigate safety compliance to make sure your product meets all regulations. Don’t take a chance on datasheet numbers that don’t reflect real-world performance—work with a supplier that understands both the physics of VCSELs and the needs of your application.
638nm Fiber Coupled Diode Laser References:
- Coldren, L. A., & Koch, T. L. (1995). Vertical-Cavity Surface-Emitting Lasers. IEEE Journal of Quantum Electronics, 31(2), 258–267.
- International Electrotechnical Commission. (2014). IEC 60825-1:2014 Safety of Laser Products – Part 1: Equipment Classification and Requirements.
- Kish, F. A., et al. (2020). High-Power 940nm VCSEL Arrays for Automotive LiDAR. Proceedings of the International Conference on Infrared and Millimeter Waves, 1–3.
- Li, J., et al. (2018). Uniformity Analysis of Large-Area 940nm VCSEL Arrays for Facial Recognition Applications. Optics Express, 26(12), 15672–15683.
Hangzhou Brandnew Technology Co., Ltd.
Hangzhou Brandnew Technology Co., Ltd. is one of the leading 940nm vcsel manufacturers and suppliers in China, has a professional factory which manufacturers high quality 940nm vcsel and sells at competitive price. Welcome to wholesale our products made in China.
Address: 17F,Building 2, Aoqiang Mansion, No. 6 Xiyuan 5th Rd,310030 Hangzhou,China
E-mail: admin@brandnew-china.com
WebSite: https://www.brandnewdiode.com/
