The lighting battles of old were fought with crude tools. Today’s battles are more sophisticated with weapons called TM-21 and LM-79 instead of yesterday’s tool of choice—price. Our industry is much better served when we compete on features and benefits and not the lowest landed cost.
A new battle is shaping up in lighting and it is a sign of a greater war coming. Today’s skirmish is fought on the MR-16 front.
The first salvo was fired by Cree in an article published in the October edition of LED’s magazine. The article, LEDs ready to displace halogen in MR16 lamps by Mark Youmans and Mark McClear. The article states, “GaN-on-GaN LED technology… allows for relatively higher drive currents and therefore higher current densities…” The article further states, “Even though the GaN-on-GaN chips may perform well under these stressful conditions, the phosphor system and the LED package still have to reliably survive these conditions or the MR16 lamp cannot maintain the performance of halogen lamps over time.” This was a shot directed at Soraa, the leader in GaN-on-GaN.
A second, more penetrating shot, comes later when the writers state, “The LM-80 data for the GaN-on-GaN LED described here has not been published, but Cree has tested some of these MR16 lamps in real-world scenarios (open air, Ta = 22°C; closed fixtures, Ta = 55°C) and found that the L70 for at least a small sample of these MR16 lamps can be as short as about 4,000 hours — barely exceeding typical halogen rated lifetimes.”
The issue is that Cree and many other companies publish LM80 and TM21 summaries. Soraa does not—at least not yet.
Mike Kraemes, CTO of Soraa, took issue with the Cree article and posted a few comments on his blog. He stated, “At Soraa we have harnessed the power of GaN-on-GaN in a holistic design approach to MR16 lamps which provide up to 65 Watt equivalent, single-beam intensity, full visible spectrum performance including CRI and R9 (deep red) rendering levels of 95, true halogen-like whiteness rendering (not possible with blue-based LEDs), and enclosed-fixture use, in products rated for > 25,000 hrs and warrantied for three years.”
He went on to say, “Incumbent technologies always fight back against the newer, better approaches – it’s just the natural gasping for breath that portends demise. We’ve seen this pattern time and again in history, from vacuum tubes to transistors, from CRT tubes to LCD TV, and now from GaN-on-X LEDs to GaN-on-GaN.” This was a shot at Cree.
Maury Wright, Editor of LEDs Magazine, tried to sort everything out in his response on 29 OCT. She commented, “Soraa’s statement that its MR16 lamps being rated for 25,000 hours in an enclosed fixture presumably addresses the issue of reliability and longevity.” Really? Wright assumes Soraa’s statement simply settles that argument. We disagree.
Cree’s argument is logical. The phosphor system and packaging must survive the higher drive currents that GaN-on-GaN affords and there is no public data to prove Soraa’s claim. I assume this lack of public data prompted Cree to publish their findings of their competitor’s product.
Your humble editor spoke to Mike Krames at Soraa. Mike emphasized that GaN-on-GaN utilizes higher drive current densities, not necessarily higher currents. In some cases the total power may be slightly higher, but Soraa uses proprietary packaging materials not available to the rest of the industry. I pressed him to publish his summary LM-80 numbers and he pushed back pointing out that longer time data is needed for TM-21 extrapolations of lifetime. He reminded me that data at 6000 hours looks much more stable than at 2000 hours. He also sent a publicly-available Philips Lumileds chart supporting this fact. Mike explained that they are in testing to qualify for Energy Star and Soraa expects Energy Star certification in due course.
I asked about the data Cree published on GaN-on-GaN. He responded that the data was troubling and totally inconsistent from Soraa’s internal test. He cautioned that there was no telling where that device came from and that it could have been an early prototype, and emphasized that it is not clear from the article whether it is even from Soraa.
We went back to Mark McClear who co-authored the original article. He sent the following response, “Six or seven years ago, during the real wild west period of the LED industry development, every LED manufacturer had their own internally developed model for lifetime and reliability. IES LM-80-2008 and IES TM-21-2011 leveled the playing field and provided a means for customers to compare lumen maintenance claims from the various suppliers on an apples-to-apples basis. These standards give the Lighting industry confidence in this new technology and we would encourage all LED manufacturers to support these IES standards.”
A lot of people got shot in the Wild West and the bullets have just started to fly. President Reagan did not first speak the term, “Trust but Verify,” but he made it popular. In the case of LED, the term is very appropriate.