LED Lighting Technology | LED Programming

Ford Kerosene Car Lamp

Calcium Carbide Car Lamp

Original automotive headlights burned kerosene, had a chimney, needed to be lit and extinguished.  Headlight innovations around the same time gave calcium carbide generators, with a controlled water drip, produced Acetylene Gas, which, when burned, produced intense light, greatly improving a car's nighttime vision. 

Today we don't have to re-fuel our headlights. Instead we enjoy LED lamps rated at 50,000 hrs, or 5 years of continual operation. In the past 2-3 years the industry has made great strides in integrating high performance LEDs in all vehicles. From side markers, to taillights, to the interior LEDs are becoming the standard. Like the LED lightbulb replacing standard bulbs, all auto lamps are headed to obsolescence.

The "holy grail" of automotive lighting has been to replace headlight lamps with an LED solution. A legitimate LED swap has been completely unheard of until now. The doors have flung open with the recent milestone in LED brightness output of 100 lumens per Watt, and further with some manufacturers demonstrating 200 lm/W in laboratories. Cars need over 800 lumens for proper headlight usage, and these LEDs are up to the task. For example, a single Cree XLamp XP-G Cree XLamp XP-G series LED can provide 400 Lumens at 1A and over 130 lm/W at 350mA. Driving the LED at 1A produces the maximum amount of heat, the number one issue with LED systems.

However, price has completely blocked the wide automotive adoption of LEDs. Just a few years ago, a single LED would cost $5.00 USD, in the past year has dropped to $1.00 USD. The sweet spot of price and ease of manufacturing should be hit in the coming 12 months. Despite initial cost, overall lifetime cost and environmental impact will over pay for itself. A regular halogen lamp, single beam, requires 55 watts (240 watts on high beam). The more efficient Xenon lamp needs 35 watts. Average power input need for LED lamps comes in at 28 watts per beam. The overall efficiency is only getting better. Volkswagen plans to introduce OSRAM Joule JFL2 LED systems into their headlamps promising only 19 watts per beam. Environmental impact also pushes adoption. On conventional cars, with LED lamps, 196 grams of carbon dioxide per 100Km (62 miles), compared to 768 grams per 100Km with halogen bulbs.

Plug-in and Hybrid vehicles benefit the most from LED lamp adoption. Up to 6 miles is added to a hybrid car's electric only range with the use of LED headlights. 19-28 watts is a far cry from the 55 - 240 watts needed with halogen bulbs. Toyota/Lexus was the first to offer LED lamps as an option, followed by Audi, Cadallac, and 2011 has most car manufacturers promising LED options. Most LEDs options are used for daytime running lights at the moment, but full beam implementation is close. As brightness levels increase and modules prices decrease, adoption for full headlight usage will become market standard. As of 2011, 3rd generation Toyota Prius, Nissan Leaf, and Mitsubishi i-MiEV will have full beam LED headlight options.

Efficiently driving and regulating an array of LEDs in a headlight is the main design difficulty. A few ICs are available now to drop right into the application. Linear Technologies has a DC/DC converter, LT3956, designed to operate as a constant-current and constant-voltage regulator, ideal for high brightness/high driving current LEDs. In particular, a 25 W white LED headlight lamp can be configures with 18 of the Cree XLamp XP Cree XLamp XP (130lm/W at 350mA) series elements. The LT3956 LT3956 outputs a PWM signal ranging from 100kHz to 1MHz, giving the user a dimmable ratio of 3000:1. Due to its tunable features, the LT3956EUHE#PBF LT3956EUHE#PBF is up to a 94% efficient. Drives LEDs in Boost, Buck Mode, Buck-Boost Mode, SEPIC or Flyback Topology. However the 6% in efficiency, or 1.5W (25W*0.06), is dissipated as heat, even with the UHE package (5mm x 6mm) heat-sinking is inevitable.

Another option is the Renesas LED headlight driver, μPD168891, which can provide constant-current control for up to 12 LEDs in series. Renesas's IC has built in overcurrent protection, diagnostic functions, emergency shutoff to prevent current issues, and is in a 48-pin VQFN package to reduce heat generation.  A few others ICs exist, but no one chip is the end-all option in this fledgling industry. Choosing the right components is based on individual design requirements, and should be chosen accordingly.

A mere 100 years after the kerosene headlight, the automotive lighting has grown to be an over 10 Billion USD industry. Electrically efficient high lumen LEDs made it possible, and more of them are coming every quarter. With few components available in this relatively new automotive LED Lighting industry, the brave and the willing to get up to speed in a very short amount of time. Very few times in technological advances can someone start at the leading edge of an industry without extensive training. The building blocks are here, they just need someone to put them together.

Cabe

Personal automotive lighting design story:

I started in the automotive industry designing a side (turn) lamp systems for a future hybrid.  What I discovered is nothing more than V = I*R is needed for me to begin that career path. At first I was meticulous and cautions about power regulation, but eventually it came down to just matching a resistor to a LED ( R = V/I ). A later incarnation of my circuit used a driver IC like mentioned above, all I had was a datasheet. The prototype was finished and on its way to being part of the LED Auto Revolution. I will say with my first hand  experience, the industry needs more engineers.

Image sensors as used in cell phones are partially color-blind, due to their coating, which prevents UV light from passing through. Because of this, CMOS chips have not been suitable for spectroscopy. However, a new production process makes the coating transparent, and the sensors suitable for special applications, and may allow for use in spectroscopy after all. CMOS image sensors are no longer only used in cell phones and digital cameras. The automotive industry, for instance, has discovered the potential of optical semiconductor chips and is increasingly using them in driver assistance systems, from parking aids and road lane detection to blind-spot warning devices. In special applications, however, the sensors that convert light into electrical signals have to cope with difficult operating conditions, such as high temperatures and moisture. For this reason, CMOS devices are covered with a silicon nitride coating. This chemical compound forms hard layers which protect the sensor from mechanical influences and the penetration of moisture and other impurities. The protective coating is applied to the sensor in the final stage of CMOS semiconductor production. The process is called passivation, and is an industry requirement. Unfortunately, up to now this passivation has entailed a problem: the silicon nitride coating limits the range of optical applications because it is impermeable to light in the UV and blue spectral range. CMOS sensors for high-performance applications, used in special cameras are therefore partially color-blind. Scientists at the Fraunhofer Institute for Microelectronic Circuits and Systems IMS in Duisburg have found a solution to this problem. “We’ve developed a new process step that allows us to produce a protective coating with the same properties but which is permeable to blue and UV light. This reduces the absorption of shortwave light,” said Werner Brockherde, head of department at Fraunhofer IMS. In simplified terms, the new coating material will absorbless light of an energy higher than blue light, which means the sensor becomes more sensitive at the blue and UV range. With this process development the experts have expanded the range of applications for CMOS image technology. This could revolutionize UV spectroscopic methods, which are used in laboratories around the world, significantly improving their accuracy.

Eavesdropper

Researchers at Princeton have developed a rubberized material that can be places over the surface of objects and detect cracks in the surface. How? An organic laser is deposited on the surface of the sheet of rubber. When the rubber stretches, the color of light emitted changes, allowing structural engineers to more readily observe any changes in the structure. Researchers have been focused on using sensor arrays for such applications; however, a stretchable laser would cover more area than wires or fiber optics. The device is made from a specially prepared sheet of polydimethylsiloxane (PDMS) that is made to have a wavy surface and which is covered with organic molecules. The lasing effect occurs when the material is exposed to an ultraviolet laser. Once photons are emitted, the material’s surface acts like a diffraction grater, amplifying the signal. The stretchable sensor was constructed by Sigurd Wagner, professor of electrical engineering at Princeton, with Patrick Gorrn, a researcher at Princeton. Wagner says that his prototype still needs to be fine-tuned. While the PDMS sheets can stretch a great distance, the organic layers sheer off when they're extended too far. Fixing this problem will likely come down to testing different types of light-emitting molecules and finding a way to better affix them to the PDMS. “We know the experiments to do,” he says. “We just haven't found the magic recipe yet.” If only this material covered all space-shuttles and rockets, many catastrophes could have been avoided.

Eavesdropper

Ostendo Technologies, Inc. and Technologies and Devices International, part of the Oxford Instruments Group, have recently announced that LED structures grown on their semi-polar GaN wafers have resulted in more than 2.5x the emission intensity of the c-plane GaN based LED structures. Ostendo & TDI had entered into an Information Exchange Agreement with Palo Alto Research Center (PARC) in 2008, pursuant to which they agreed to make semi-polar GaN wafers available for PARC to grow LED and Laser Diode structures on the supplied wafers, and independently validate and report the achieved results.  As part of their validation, PARC has grown MQW LED structure on our semi-polar GaN side-by-side with a reference c-plane LED structure in the same MOCVD run.  Some of the key results verify the following: The LED structure grown on their semi-polar GaN achieved more than 2.5x more emission intensity than the reference LED structure grown on c-plane GaN. The new semi-polar GaN allowed for higher indium (In) incorporation resulting in longer peak wavelength of ~25 nm for the structure grown. “This is an excellent validation of our work in the semi-polar GaN area for the last two and a half years as it verified the main advantage of our semi-polar GaN and should help encourage LED makers to start considering it for future LED brightness improvements,” said Dr. Hussein S. El Ghoroury, CEO of Ostendo. Earlier in 2010 Ostendo and TDI announced the availability of semi-polar (11-22) GaN layer on sapphire substrate wafers using Ostendo’s proprietary design and TDI’s proprietary Hydride Vapor Phase Epitaxy (HVPE) technology.  This joint development now provides the opportunity to leading High Brightness Light Emitting Diode (HBLED) and Laser Diode developers to increase optical efficiency significantly compared with structures grown on c-plane GaN substrates.

Zero