POST OF THE DAY
Cree
Cree's ASM

Format for Printing

Format for printing

Request Reprints

Reuse/Reprint

By Imshaken
October 30, 2002

Posts selected for this feature rarely stand alone. They are usually a part of an ongoing thread, and are out of context when presented here. The material should be read in that light. How are these posts selected? Click here to find out and nominate a post yourself!

By now I'm sure most of the developments of the ASM have been posted but I'm going to add my few cents worth. As has happened in the past I met a few new people who struck me as quite intelligent from whom I could learn quite a bit. This year we had a friendly, if more sober and grounded crowd than in years past. No doubt a sign of the bear market. :o))

No great surprises. Swoboda gave a polished presentation. He described a few of Cree's challenges, where Cree has been and hopefully where Cree's going. He described Cree's product line segments and each of their potential market segment potentials. (Anticipated annual value of markets in five years is projected to be: LED @ $6 billion, RF device @ $1 billion, power device @ $3 billion, laser @ $2 billion) Swoboda didn't offer any guidance of what Cree's projected market share of any of these segments would be in five years. He commented that there are variable paths Cree can take to achieve the billion dollar sales mark. I guess the bottom line is it was a nice presentation, but it was a bit like showing someone a map of the U.S. but then not describing which route we're taking from point "A" to point "B". Still to be fair, if Cree doesn't know how their market segments are going to develop I would prefer they don't blow smoke. So this was OK.

I noticed, and am sure others noticed that Cree's visibility is noticeably reduced from past years. Cree is now in the same boat with the rest of the semiconductor industry. Thus prognostication about where Cree's revenue / profits will be in six months, much less next year or in five years was rather muted. I think there might actually be an upside to this development. That is.... reduced expectation. If the market continues to respond to Cree's financial results AFTER earnings instead of running up INTO earnings, as we witnessed this quarter, perhaps the stock price will experience lower levels of volatility. The benefit of that would be less momentum short sellers and cleaner message boards. :o))

But I digress...

Cree demonstrated a few headlight / spotlight devices configured with nine Powerchips (3x3). These were blindingly bright. To my layman eye it appeared bright enough for auto headlights. However, a brief discussion with an amiable Cree exec whom Cree swiped from Osram a while back... (I don't know if he wants his name here) indicated headlight applications are still probably a few years off yet. He indicated last year a headlight would have required eighty Cree LEDs, this year about fifty Cree LEDs. (I didn't ask him to specify what type LEDs, I'm assuming XBright). He indicated when twenty five LEDs become sufficient to power a headlamp auto manufacturers may be inclined to make the switch which itself would take a couple years. Thus, we shouldn't hold our breath. Sounds like four years to me.

Also, there's a rumor Cree's LEDs are in Honda vehicles. wink wink. I'm going to test drive a 2003 Honda this week. If there's an exposed LED I'm going to put an end to speculation. Although after today, I have no doubt. I might just have to purchase a Honda Pilot if Cree's inside. :o))

Dr. Palmour, a budding film star, was kind enough to offer a tour of power switching devices and offer a couple comments. Cree's current Power Switching device design win is in a power factor correction circuit. I understand this to mean that the function of the circuit the SiC schottky diode is in is intended to bring current and voltage which is out of phase, back into phase. Evidently induction loads change the current/voltage phase relationship. (It's been twenty some years since I had any electronics so the cobwebs be thick). Anyway.... Cree's SiC schottky having almost no reverse recovery permits the removal of a "snubber circuit" which requires several components. Thus, the footprint of the entire power factor correction device can be made smaller, with less components. Thus, the price level that must be achieved to begin getting design wins must be less than the cost of the components that can be omitted because of the SiC schottky. (You could also factor in the cost of installation of the components as well IMHO. Although that number might be harder to determine than individual components' costs.) Additionally, since the SiC schottky operates without the reverse recovery it operates at lower temperatures. Thus in rooms where there are hundreds of thousands of these circuits (telecom switching rooms) the air conditioning savings alone might justify the SiC schottky design wins.

Dr. Palmour also indicated Cree's working with end use customers to try and educate them about SiC schottkys advantages in order to gain design wins. It's going to take time. Both Swoboda and Dr. Palmour (word has it... the next Clark Gable) indicated that Infineon being a competitor in the SiC schottky market is a benefit and validates the material / device. I've heard Cyndy Merrell say the same thing in the past. It seems logical enough. Still I think once the orders are rolling in for SiC schottkys, Cree should stop selling SiC to Infineon just as a little joke. :o))

Also stated was that a large opportunity awaits Cree if they can [get] 1200V SiC schottkys to market. Dr. Palmour indicated that these devices will be about four times the size of a 600V SiC shottky of the same current value. This will require better substrates since the devices will be quite large. (Whether he was referring to material having to be better than Cree's new improved three inch wafers (mentioned below) isn't clear to me. Since Cree hasn't released these devices yet some hurdles evidently remain.)

The upshot is, the power device market is a tough nut and like all things worth doing it's going to take some time.

Also, at one point in his presentation Swoboda indicated Cree had significantly improved the quality of their three inch wafers. Improved to the extent that they were better than the two inch wafers made just a year ago. Dr. Palmour, the Erol Flynn, put an exclamation point on the description. He indicated the three inch wafers today are WAY better than the quality of Cree's two inch wafers of a year ago. Evidently Cree's made considerable progress. This should help commercialization of SiC schottkys and SiC RF devices as quality of the substrate is / was (?) one of the hurdles to achieve larger area devices.

While trying to get any decent information out of the fellow who was showing off Cree Microwave's "wares" I kept getting "I can't comment on that", or "I can't say", or "I can't comment" or something similar. He indicated Cree's starting to work on LDMOS 9. (He didn't /wouldn't share the differences other than more linear, more efficient.) So evidently there's at least one more iteration of LDMOS to come. (Iteration is that a great word or what!?) I heard Redfox use it today and I thought great word so I thought I would share that here. :o)) Now, back to the program... FWIW: When asked if SiC RF devices would be penetrating the market, say in five years, one of Cree's finest indicated that Yes, SiC will be commercialized within five years. (Or something very similar to that.)

Personally, at this point my feeling on Cree's RF device line is that if and when it happens it will be a pleasant surprise.

Swoboda also mentioned the 30mW laser now passed a life expectancy of 1000 hours. He indicated Cree's research team results with the 30mW laser had surpassed even the most optimistic forecast of even their most aggressive researcher. This is welcome news. It's nice to know these guys can bang something out when the chips are on the table. I wonder what price was paid in human capital, with respect to burn out? (You know the pressure was on, maybe even more intense than at Cree Microwave when they had to get LDMOS 8 out or else...

They also played a nicely done Cree commercial. Maybe they should play a modified version of it on a national TV campaign to educate the general public about Cree. (Just kidding Libra) :o))

In the end I came away having been reinforced that Cree's SiC IS going to ultimately change the world in many ways. However, significant hurdles remain, and there are going to be plenty of bumps in the road.


Become a Complete Fool
Join the best community on the web! Becoming a full member of the Fool Community is easy, takes just a minute, and is very inexpensive.