CAP-XX Limited (LSE:CPX), developer of thin-form supercapacitors, has published an updated study comparing flash solutions for camera phones - xenon, standard LEDs powered by a battery, and high-current LEDs powered by a supercapacitor using the company's BriteFlash power architecture. The study tested each solution's ability to deliver the light energy needed to take digital-still-camera-quality pictures in low-light conditions, and also compared shutter requirements, ease of design-in, safety and size.
The original report from October 2006 compared light power and energy using 1.3 to 3.2-megapixel camera phones. The new report includes data from 5-megapixel camera phones released in the last year, and also considers advancements in camera sensors, xenon flash units, high-power white LEDs (WLEDS) and LED flash drivers.
Tests again showed that the LED BriteFlash approach delivers more light energy than most xenon flashes in a thin form factor suitable for slim camera phones and digital cameras.
Clear pictures in dim environments require sufficient light energy - the total amount of light received by each pixel in the camera sensor - during image-capture time. "People often wrongly assume that light power, which is the brightness or intensity of the flash, is the key because it's what draws our attention, but it's really the light energy that counts," said Pierre Mars, CAP-XX vice president of applications engineering.
To calculate light energy, one would multiply light power (in lux) by the duration of the flash exposure (in seconds): Light power (lux) x flash exposure time (sec) = light energy (lux.sec). Ten to fifteen lux.sec of light energy is ideal for high-resolution pictures:
- Xenon flash tubes driven by electrolytic storage capacitors deliver higher light power, but over a very short flash exposure.
- High-current LEDs driven by a supercapacitor deliver lower light power, but over a longer flash exposure to generate more light energy.
Flash solutions tested:
- Xenon: SonyEricsson K800, LG KU990, Nokia N82 and Samsung G800, all with 5-megapixel cameras but with varying size electrolytic storage capacitors.
- Standard battery-powered LEDs: Nokia N73 (3.2-megapixel) and N96 (5-megapixel)
- Supercapacitor-powered LEDs: To demonstrate the BriteFlash approach, CAP-XX used a small, thin (20mm x 18mm x 3.8mm thick), dual-cell supercapacitor to drive a two-LED array of Philips LUXEON PWM4s at 2A each or 4A total during the flash pulse.
"BriteFlash maximizes performance from our LUXEON Flash LEDs so that cell phone users get superior image quality," said Michel Zwanenburg, product manager, LUXEON Flash, Philips Lumileds.
A photo detector measured on-axis illumination, while a digital storage oscilloscope captured light power over time at 1 and 2 meters from the source. The areas under the power curves were integrated to measure the light energy at the detector as a function of time.
The supercapacitor-powered BriteFlash example (two-LED array powered at 2A per LED), using a 15-frame-per-second rolling shutter over a 67-millisecond flash exposure, delivered more light energy than the xenon flashes.
From 1 meter, the BriteFlash LEDs delivered the best of all cases with 21.7 lux.sec, 37 percent more than the best-performing xenon, which was the SonyEricsson K800 with 15.8 lux.sec. The standard battery-powered LED flash unit in the Nokia N73 delivered only 1.71 lux.sec with 1 LED, and 3.45 lux.sec in the Nokia N96 with 2 LEDs.
From 2 meters, the BriteFlash LEDs delivered 7.0 lux.sec, approximately 60 percent more light energy than the 4.45 lux.sec from the best-performing xenon, which was again the SonyEricsson K800. The standard battery-powered LED flash unit in the N73 delivered only 0.43 lux.sec with 1 LED, and 0.86 lux.sec in the N96 with 2 LEDs.
The BriteFlash example over a 33-millisecond flash exposure from 1 meter, which is short enough for crisp images without using anti-handshake image-processing software, delivered comparable light energy to two of the xenon examples: 11.2 lux.sec compared to 11.5 lux.sec from the Samsung G800 and 10.2 lux.sec from the Nokia N82. Image-processing software is readily available for designers to use to correct for hand movement that may blur images captured over a longer, 67-millisecond exposure.
For complete light energy test results, see: http://www.cap-xx.com/news/photogallery.htm#FlashTable
The study also compared ease of design-in, safety and size. Supercapacitor-enabled LED flash units are smaller and thinner than xenon solutions (2 - 4mm thick, occupying less than 2cc, compared to 3.8cc and 7mm thick for the K800 xenon flash unit), use a lower voltage (5V compared to a 330-V electrolytic storage capacitor), recharge quicker between flashes (two seconds compared to eight in the K800), do not require a mechanical shutter to achieve correct exposure, and can handle other peak-power needs in the phone, including the RF power amplifier and audio amplifier. The result is longer talk time and better photos and audio quality. For a more complete comparison, see: http://www.cap-xx.com/news/photogallery.htm#DesignTable
Stuart Robinson, director of handset component technologies for Strategy Analytics, explained, "Consumers want camera phones that rival digital still cameras. We estimate high-powered LED flash will grow significantly, reaching 30% of all camera phones by 2012. Supercapacitors are an enabler for this market and we have seen them power WLEDs to produce clear pictures in low-light conditions."