the following info was compiled by a poster from another thread...
CREE OVERVIEW
Silicon Carbide (SiC) is a desirable compound because it's a "wide bandgap" material. This makes it a very good semiconducting material. By and large all electronics today are based on semiconducting materials. So, it's been observed that the uses for SiC in electronics is nearly limitless. Furthermore, SiC melts at 2300 degree's C. It can be glowing red hot and still be stable as a semiconductor. SiC can be used as a semiconductor at temperatures approaching 700 degrees C. SiC also has a very high coefficient of thermal conductivity. These characteristics are far superior to silicon, the primary semiconducting material used today in industry. Additionally, SiC based devices when being used in a semiconducting capacity aren't affected by radiation or the EM pulse from nuclear blasts. This is known as "hardened electronics". These characteristics raise the commercial value of SiC, in it's pure monocrystaline form SiC is a highly desirable material. SiC is the second hardest substance known to man, second only to diamond. One last note, SiC does not occur naturally on earth. It's primarily found in meteorites, a resulting from the combination of heat and vacuum. There are no reliable natural sources for this compound.
Here's some information about Cree Research Inc. Based in Research Triangle Park in North Carolina. Cree currently has two primary sources of revenue. They are: Contract revenue for SiC research and SiC product sales revenue.
RESEARCH
Cree performs contract research for government agencies, the Department of Defense (DOD), and private industry. Cree had contract revenues of $6.5 Million dollars in fiscal year 1999. In September 1999, Cree announced additional government contracts worth $6.8 million for microwave and powerswitching R&D, a $4 mill contract from the DOD to lower costs on 3" and 4" wafer development.
During this last fiscal year Cree purchased $5 Mill of MVIS stock in a private placement. In return MVIS is funding Cree's research on blue laser development. (Something Cree was going to do in any event.) MVIS is funding Cree for a year with $2.6 Mill. MVIS has an option to provide a 2nd year at $2.5 mill. This is a win win deal, the best kind of business and demonstrates Cree ‘s management's skills.
Cree has performed research for power switching devices for Kansai Power of Japan based on $3 mill research funding. This research has evidently started to pay dividends with new devices being released later this year by Cree.
More detail on research grants if you're interested:
Of these awards, $3.5 million is directed toward the development of component technologies to enable future generation Advanced Multifunctional Radio Frequency Systems (AMRFS). The U.S. Navy has reported that they intend to invest millions of dollars into new radar technology in coming years to develop a more capable radar that can search, detect, and track smaller and faster targets. However, research and development will be required to enable amplifier chips made from silicon carbide and gallium nitride to perform very wide bandwidth radar functions.
John Palmour, Cree's Director of Advanced Devices and a co-founder of the company stated, ``Future generation advanced radar systems will require higher power levels that we believe are attainable from silicon carbide and gallium nitride devices. Power levels required for these systems are not possible with current silicon or gallium arsenide technologies.'
AFRL has funded an additional award focused on high power gallium nitride microwave devices for airborne radar and space-based military systems, as well as a variety of commercial applications. Other programs from Defense Advanced Research Projects Agency (DARPA), through ONR and AFRL are focused on power switching devices for motor control and power conditioning applications.
Libra Sing observed the following, which makes sense to me; Government DoD contracts make good business sense as they allow Cree to advance even more quickly in specific areas. But there is another reason in my opinion (IMO). It confirms that Cree is a leader in their chosen field.
PRODUCTION
Cree produces high purity Silicon Carbide (SiC) wafers. By all accounts this is a very difficult thing to do. GE, Westinghouse, Siemens, have all tried to develop SiC and failed. (Siemens and GE have efforts underway via subsidiary's / partnerships) Cree has been very successful at producing these wafers to date. Currently Cree produces between 90 to 95% of the world's high purity silicon carbide. Cree would seem to a monopoly in the production of SiC at this time. Recently Neal Hunter, Cree's CEO indicated that the competition is currently about 6 years behind Cree, and Cree is extending their technological lead.
A couple of years ago Cree was producing SiC wafers less than an inch in diameter, today Cree is completing the installation of process equipment for 3" wafers and has demonstrated the ability to manufacture 4" wafers. The 4" wafers will be implemented into production over the net 12 to 18 months. The significance of the wafer diameter increases is that the area of SiC is increased exponentially by the square of the radius. So the doubling of diameter roughly increases actual SiC output by 4x. The increase of SiC wafer diameter has increased Cree's profit margin considerably in the last year or so. More about that later.
What does Cree do with the wafers? Several things: Cree sells wafers, and manufactures devices.
WAFER SALES
1. Cree sells SiC wafers to optoelectronic researchers. Cree recently achieved a price breakthrough and is currently selling 2" diameter wafers for $495 which evidently makes it competitive with sapphire as a substrate for use in optoelectronic devices. I understand from a statement in the last conference call that Cree's profit margin on the 2" wafer sales is about 50%. According to the 1999 annual report about 10% of Cree's wafers are sold to "external" companies, researchers.
2. Cree also sells SiC to a company called Charles and Colvard (CTHR). AKA...C3 (An external company). In the last year about 20% of Cree's total product sales have been sales of SiC product to C3. C3 takes the SiC and cuts, polishes and fashions it into gemstones. C3'S gemstones by eye are identical to diamonds except that Moissanite is if anything brighter with more "fire" than a diamond. C3 markets these gemstones as "Moissanite". C3 is trying to position Moissanite as a unique gemstone in it's own right. Thus far Asian and European markets apparently like this product. Marketing efforts in the US have been less successful. C3 also sells a testing device to enable jewelers to discern the difference between diamonds and Moissanite. C3 is in the process of a 4th quarter advertising campaign in the US to increase sales and retail dealers.
I understand C3 is contractually required to purchase at least 50% of SiC product produced by Cree until the end of this fiscal year (June 2000). C3 has stated that they will accept 100% of the output, and plan to continue this policy for the remainder of the contract. The SiC is grown in C3 owned growers but located at the Cree Research plant and operated by Cree personnel.
Neal Hunter stated in the last conference call that he doesn't think C3 will stop accepting Cree product after June 2000. However, if C3 sales don't increase, Cree has a plan to compensate for the loss of SiC sales to C3. Cree will convert C3 growers to produce SiC wafers for use in other Cree devices. (In this scenario I believe Cree might actually make more money).
IMHO the technical matters pertaining to production of viable gemstones for C3 seem to have been largely overcome. All that remains for C3 to become a profitable company is to find a successful marketing scheme. The vision to achieve that goal has yet to be displayed. The CEO is an engineering type, not a marketing person, so this may be a problem. C3 management is attempting to address their marketing and insufficient dealer outlet problems.
It's my guess that Cree's sales to C3 will continue until the contract ends, then shipments will be reduced significantly. After C3'S gemstone backlog is reduced, sales to C3 will begin to increase from the reduced levels. I'm guessing this will begin around the holiday season of 2001.
DEVICES.....YUMMY!!
Cree makes several devices and is attempting to develop additional devices.
LEDS
Cree manufactures LED die packets which are shipped to customers who then package the dies into a finished LED. Cree's LEDS are manufactured in a process called epitaxy. This is a process where the SiC wafers have a Gallium Nitride (GaN) layer deposited on the surface of the SiC. The GaN is the material that emits the blue or green light. The white LEDS are actually "blue LEDS with a phosphor coating that emits white light when energized by the blue light from the LED. These wafers are etched, polished, cut into led segments, have ohmic contacts attached, sent to quality control, then sent to shipping. (I'm not really familiar with this process so if I've made an error in the order or left out a process don't crush my stones, OK? :0)) )
Here's some additional technical information about Cree's LED fabrication from a message board poster:
The process is that doped SiC is the base, with a conducting buffer layer next, topped off with GaN. The conducting buffer layer is a trick and the jury is out as to how they do it. It provides three distinct advantages over sapphire/GaN LEDs.
1. Size. The LEDs that Cree makes can have a contact on the top (on the GaN) and one on the bottom (on the SiC). Sapphire does not conduct, and must therefore put two contacts on the top, which is bigger and more costly (many more steps involved which I won't go in to).
2. The buffer layer allows the so-called vertical structure of Cree's devices, and makes it very difficult to engineer around. The patent protection is thus more solid. The side to side devices (sapphire) can have numerous styles, and therefore can be re-engineered, whereas Cree's can't.
3. The buffer layer allows better GaN films because it more closely matches the lattice size of GaN. Sapphire and GaN are way off, and thus don't fit together well, and this creates defects, which kills your device. The buffer that Cree uses produces a better surface for GaN to grow on.
What is the buffer? It's probably AlGaN, but it may be straight AlN. The thing with AlN is that is doesn't conduct well, and therefore must be doped to conduct, and I don't think anyone has done that yet. As I said, the jury is out, but Cree has patent(s) on AlGaN, so that is a safe bet probably. A r_rh post describing why Cree's LED die construction is different and currently better than others.
All of the competing LED products have both electrodes connecting to one side of a chip, CREE's connect on opposite sides (which is the industry standard for packaging equipment and all other LED's)*
When we speak of top & bottom connections and side by side, we aren't talking about the visible electrodes coming out of the polymer package. Look inside an LED, (or in the CREE club photo album) at where the contacts nearly meet inside the LED. On top of the horizontal bar, there is a small rectangle with a nearly invisible piece of wire connected to the top which extends to the opposite post. The little rectangle is what CREE makes (and currently charges about 20 - 30 cents for). If you think about this for awhile, you will understand how valuable these wafers are and some of the talk that goes on around here.
* SiC LED's (CREE only) have a conductive substrate , all other blue/ green/white LED's are based on Sapphire which is non conductive and requires same side contact. The one exception known at this point is experimental devices out of Nichia. Nichia is removing the sapphire substrate on it's blue lasers leaving a GaN layer at the "bottom", or non firing end. Test LED's at Nichia also have the sapphire removed, supposedly making them technologicly equal to CREE's (opposing connections) but, the additional manufacturing steps to remove the sapphire make the product prohibitive. Shuji (the lead researcher at Nichia) is still working on GaN bulk crystal production.
rh
Cree makes the following LED products. Standard Brightness Blue, Green and White LEDS. High Brightness Blue and Green LEDS. Cree may produce High Brightness White LEDS at this time but I'm not aware of it. Cree's LED sales accounted for about 80% of production revenue last fiscal year. Cree during the last year become the worlds leading provider of nitride based LEDS. They accomplished this feat by driving down LED prices so that they are the lowest cost source of blue and green LEDS in the world. In March 1999, Cree shipped over 20,000,000 LEDS. Cree's blue LEDS are used in Volkwagon dashboards, and white LEDS are used Audi's new TT auto. Cree just signed an agreement to provide Nokia with white LEDS. I'm not knowledgeable about specific details of that agreement. Cree's High Bright Green LEDS are being installed in China's stoplight system. Cree expects that after brightness is increased through ongoing research the US stoplight market will be penetrated. Cree has announced a backlog of $111 Mill in LED sales. My guess is that the backlog will increase further before it gets better. :0))
MICROWAVE AND RF DEVICES
Cree has also developed RF high temperature transistors to be used in cellular telephone base sites. My understanding of this device is incomplete (what else is new), but I understand that it can tolerate about 5x the power of existing transistors, operate at much higher temperatures, and eliminates a dc-dc transformer required by existing technology. I believe I heard Hunter state during a conference call about 6 months ago that some samples had been sent to a cell site equipment manufacturer. This was done to evaluate the devices and get them into the manufacturers design process. Cree never stated who that manufacturer was. The Cree message board on Yahoo has had several posts stating that the company was "Erickson". A poster on the yahoo message board recently found an add for Cree's CRF-20010 in a trade magazine. The annual report states that prototypes of these devices will become available in FY 2000. The quantity of these devices Cree can manufacture is unknown to me, as is the likely average sales price (ASP), and profit margin. Needless to say, cell sites are an increasingly common site and becoming more so. I've read that Cree intends to create a complete product line for this type device. Cree indicated that because of the design in process it would be "several quarters" before revenue from this device is realized.
This is a description of how a device like Cree's CRF operates from a post on the Cree message board by AHZeeman. Note Cree's device may operate a bit differently because their device eliminates the dc-dc transformer required by other devices:
A MESFET is a Schottky type field effect transistor or a field effect transistor that uses metal for the gate instead of a semiconductor. Using metal increases the speed of the transistor. A field effect transistor is a transistor in which the resistance through the transistor, from source to drain is varied by the gate voltage. This is similar to squeezing a water hose to control the flow of water through it. Pinch the hose and you can slow or completely stop the water flow. Remove all pressure and the water flows freely. There are two main types of FETs; depletion mode and enhancement mode. The depletion mode is like the water hose, unless you pinch the hose, the water will flow freely. Enhancement mode is the opposite, no current will flow until a voltage is applied to the gate. There are also transistors that function between enhancement and depletion. The gate influences the channel resistance only by voltage. Unlike a bipolar transistor, no DC current flows through the gate. The gate is insulated from the source and drain by a very thin film of oxide. For a more detailed explanation, check out this link.
ece-www.colorado.edu
AHZeeman continues:
There are many applications where metal on semiconductor has superior benfits. The main example is in power rectifiers. A typical silicon diode has a 0.7 volt drop across it. If you are making a 3.3 volt power supply, dropping 0.7 volts across the diode is very inefficient. By using a metal-semiconductor or Schottky diode, the voltage drop can be as low as 0.2 volts which greatly increases efficiency. The switching speed is also much higher. The trade off is that the blocking voltage is much lower and leakage current is much higher. Silicon Carbide Schottky rectifiers are even better, as they have lower leakage, are faster and can withstand much higher voltages. As for FETs, I am not sure why a metal gate is superior. Most CMOS (complementary metal oxide semiconductor) ICs use polysilicon gates which are faster. Gold doping is also used to increase transistor speeds. Another technology is GeSi.
Here's a post by HARVEST_1934 listing advantages of SiC devices over Si based devices.
Microwave and Power Devices by: harvest_1934_1999 11/21/1999 7:23 pm EST Msg: 12243 of 12246
Cree is the only company to demonstrate a MESFET device with high output from a single chip.
SiC enables the fabrication of microwave devices that can deliver 5X the power of conventional silcon or GaAs Benefits over silcon and GaAs: High thermal conductivity High operating voltage Lower cost Applications:
At 1 and 2 GHz Solid state broadcasting, Wireless base station, Military radio systems
3 to 8 Ghz Military Radar systems S and X band
Silicon begins to break down at over 2 GHz RF transistor markets and SiC advantages More power per transistor Allows for fewer transistors Less complex circuitry Requires less cooling system infrastructure Lower cost per watt of power Transistors are estimated to be 30 to 40% of the cost of a wireless base station, this gives Cree a big cost advantage over competition
This was not all of Harvest's post. The rest is in POWER UTILITY SWITCHES.
>From a Gnordo post:
Potential RF applications, to name a few:
1. Driver stages in FM radio and TV station transmitters (perhaps someday, the final stages).
2. Common two-way VHF/UHF transcievers (fire & police, aircraft, the Maytag man, etc.).
3. Radar transmitters, both fixed, mobile, and airborne.
4. NO, not microwave ovens.
5. Wireless cell phone relay stations (big market).
6. Fixed UHF/microwave data/telecommnications transmitters (like Ma Bell's).
7. Military base and mobile stations and government surveilance stations.
8. Satellite transmitters: TV, communications, military, etc.
9. Outer space transmitters (excellent app. due to temperature capability).
POWER UTILITY SWITCHES
Imagine electrical on - off switches without contacts to fail or to cause resistance to electric current flow. Further, this switch if designed to allow 20 amperes of current to pass would never let more than 20 ampere's flow even if there's a short circuit in the line somewhere. While not a circuit breaker, this development is pretty amazing to me. I've never heard any discussion on this topic but when electric motor's are energized the "the initial current inrush" is typically 6x full load current. This event causes the greatest "wear" on winding insulation. Variable Frequency Drives minimize this, but create voltage spikes of their own that damage windings. But a "limiting" switch would it seems to me automatically protect motor windings from this type of failure. Of course, inadequate torque may be developed in that case so that this technology may not be universally applicable. Additionally, it's been estimated that power companies will save 10 - 15% in "losses" as these switches are nearly perfect conductors. (That's why a power company was willing to pay to get these developed in the first place.)
If memory serves, Hunter stated in the Oct 99 conference call that Cree's power device would begin shipping in Q3 of this FY. Coincidentally, Asea Brown and Boveri (ABB) has announced they will begin selling SiC based power switching devices this winter in Europe. (A link to this site was found by Libra Sing, it's on the Cree club board). I understand that ABB signed some kind of contract with Cree and so I believe ABB is using Cree wafers for this device.
In a later communication with Fran Barsky, Cree's investor relations person, Fran stated the following: "We are anticipating the first introduction of SiC power devices in FY 2001. I can't elaborate on the process since that is proprietary technology."
Here's the remainder of Harvest's previously mentioned post where you've already read the RF and Microwave component:
Power Rectifiers and Switching Devices:
Dissipates switching power up to 90% faster than Si and allows electricity to flow to other electronic components Operates at higher temperatures up to 200-450C and voltages than Si 10X thinner device than Si Energy losses 90% lower than Si Commercial product by FY 2000 $3 million developement agreement with Kansai Electric power co. for power switching diodes
Power system applications:
Automotive Factory robotics lamp ballast Motor control-pumps Traction control HVDC and power transmission station
Device blocking voltage up to 100,000V Device current rating up to 10,000A
DEVELOPMENT
Technology being developed at this time because of research contracts, include various High Powered radar and microwave devices and blue laser diode development, and edge emitting blue laser diode development.
The blue laser diode is being developed using "production techniques" so that when a successful device becomes evident, production won't be far behind.
Aluminum Nitride wafer growth. Eric Hunter developed the process and assigned the patent rights to Cree. So Cree currently is the only company in the world who has demonstrated the ability to grow AlN wafers at a rate fast enough to be commercially viable. Cree has (AlN is the second best semiconductor known to man next to diamond.) Power utility switches developed from AlN I would imagine will far exceed even SiC switches capabilities.
CREE'S PROSPECTS
Cree Research currently has an effective monopoly on SiC production and technology, and their lead is growing. They have over 80 US / worldwide patents protecting their intellectual property.
Over time Cree's management team has displayed a significant amount of focus and have by and large accomplished most every goal they've set for themselves. Hunter to date seems to have focused Cree's energy on developing device's that will bring the most rapid return on investment. Since Cree is now very profitable and has a relatively significant "war chest" with which it can address device development, new devices are likely to be in the pipeline at an increasing rate.
To fully comprehend the intensity that Cree has focused on growth the example of LED development and sales will serve nicely. Cree has only in the last year and a half turned into a production company. In that year and a half Cree has come to provide more Blue, and Green LED'S to the world market than any other company. If this trend continues, it won't be a surprise when Cree comes to dominate the world's Blue, Green Nitride based LED market.
The SiC RF and Microwave device market is currently in it's infancy. Cree appears to have no direct competition for it's anticipated and "just developed" devices. The RF and microwave device market is rapidly expanding of course with as telecommunications explodes into the next millennium. The next few years should see exponential growth for these devices. It's been stated several times on Cree's message board on Yahoo, that LEDS are the lowest profit margin device that Cree will end up manufacturing. All other devices, RF and Microwave will be far more profitable. And Cree it would seem is just waiting for the design in process to yield sales. Additionally, once these type sales have commenced a long term customer will have been created who will order parts over and over again. People familiar with this industry and Cree indicate that this area is where Cree will make it's "BIG MONEY" in the years to come. YUM!
The power device market with the revolutionary characteristics of SiC power switches will be enormous. I previously thought that Cree would have no competition for a few years, but ABB is preparing to market their product. This area has a billion dollar a year market potential, and is of course global in scope. My hope is that Cree's power switches are superior to ABB'S. In any event, the market is certainly more than large enough to support more than one player. At least Cree sells the wafers to ABB.
The development of the Blue Laser diode will be a boon to consumer electronics where data storage is concerned. This will be a billion dollar industry from sales for DVD CD ROMS, laser printers, and any other application where a laser is used to read or write data. Data storage density will increase approximately 4x on the spot and I've read of estimates greatly exceeding even this. However, data storage is a rapidly changing field and it's certain (IMHO) that other storage mechanisms will in time be developed which will make the blue laser diode obsolete for data storage uses. The agreement with MVIS to develop an edge emitting blue laser diode will be I suspect almost a "niche" market for Cree at that point because of the size of the total market. But if Cree does come up with a viable blue laser diode in the next FY as anticipated Cree's market will be likely be greater than the laser diodes it can manufacture.
PATENT PROTECTION
This list of Cree's patents (Just U.S. I believe, they also possess 36 worldwide patents protecting their intellectual property) was provided by r_rh. I don't know if this is a comprehensive list of US patents.
CREE
1 5,972,801 Process for reducing defects in oxide layers on silicon carbide
2 5,969,378 Latch-up free power UMOS-bipolar transistor
3 5,923,946 Recovery of surface-ready silicon carbide substrates
4 5,912,477 High efficiency light emitting diodes
5 5,838,706 Low-strain laser structures with group III nitride active layers
6 5,831,288 Silicon carbide metal-insulator semiconductor field effect transistor
7 5,812,105 Led dot matrix drive method and apparatus
8 5,776,837 Method of obtaining high quality silicon dioxide passivation on silicon carbide and resulting passivated structures
9 5,739,554 Double heterojunction light emitting diode with gallium nitride active layer
10 5,724,062 High resolution, high brightness light emitting diode display and method and producing the same
11 5,719,409 Silicon carbide metal-insulator semiconductor field effect transistor
12 5,718,760 Growth of colorless silicon carbide crystals
13 5,686,737 Self-aligned field-effect transistor for high frequency applications
14 5,679,153 Method for reducing micropipe formation in the epitaxial growth of silicon carbide and resulting silicon carbide structures
15 5,631,190 Method for producing high efficiency light-emitting diodes and resulting diode structures
16 5,629,531 Method of obtaining high quality silicon dioxide passivation on silicon carbide and resulting passivated structures
17 5,612,260 Method of obtaining high quality silicon dioxide passivation on silicon carbide and resulting passivated structures
18 5,604,135 Method of forming green light emitting diode in silicon carbide
19 5,592,501 Low-strain laser structures with group III nitride active layers
20 5,539,217 Silicon carbide thyristor
21 5,523,589 Vertical geometry light emitting diode with group III nitride active layer and extended lifetime
22 5,506,421 Power MOSFET in silicon carbide
23 5,465,249 Nonvolatile random access memory device having transistor and capacitor made in silicon carbide substrate
24 5,459,107 Method of obtaining high quality silicon dioxide passivation on silicon carbide and resulting passivated structures
25 5,416,342 Blue light-emitting diode with high external quantum efficiency
26 5,409,859 Method of forming platinum ohmic contact to p-type silicon carbide
27 5,393,993 Buffer structure between silicon carbide and gallium nitride and resulting semiconductor devices
28 5,381,103 System and method for accelerated degradation testing of semiconductor devices
29 5,359,345 Shuttered and cycled light emitting diode display and method of producing the same
30 5,338,944 Blue light-emitting diode with degenerate junction structure
31 5,270,554 High power high frequency metal-semiconductor field-effect transistor formed in silicon carbide
32 5,264,713 Junction field-effect transistor formed in silicon carbide
33 5,210,051 High efficiency light emitting diodes from bipolar gallium nitride
34 5,200,022 Method of improving mechanically prepared substrate surfaces of alpha silicon carbide for deposition of beta silicon carbide thereon and resulting product
35 5,155,062 Method for silicon carbide chemical vapor deposition using levitated wafer system
36 5,119,540 Apparatus for eliminating residual nitrogen contamination in epitaxial layers of silicon carbide and resulting product
37 5,093,576 High sensitivity ultraviolet radiation detector
38 5,061,972 Fast recovery high temperature rectifying diode formed in silicon carbide
39 5,027,168 Blue light emitting diode formed in silicon carbide
40 5,015,240 Hypodermic needle shield
41 5,008,735 Packaged diode for high temperature operation
42 4,966,862 Method of production of light emitting diodes
43 4,946,547 Method of preparing silicon carbide surfaces for crystal growth
44 4,918,497 Blue light emitting diode formed in silicon carbide ....
CTHR
1 5,955,735 Apparatus and method for positively identifying synthetic silicon carbide gemstones
2 5,882,786 Gemstones formed of silicon carbide with diamond coating
3 5,835,205 Optical testing system for distinguishing a silicon carbide gemstone from a diamond
4 5,762,896 Silicon carbide gemstones
5 5,723,391 Silicon carbide gemstones
EARNINGS GROWTH
Just to give a rough example of where I personally expect Cree's valuation to go in the next couple of years I offer the following observations.
>From Harvest: Cree's SiC output increased 12x (1200%) in the last year. >From the annual report: Cree's valuation increased 338% in the last year. So we have a rough relationship that a 12x increase will result in a 3x increase in stock price. Not necessarily a rigid relationship certainly but I'll go with it for now. (I don't have a better model, I'm just a vibration guy).
Cree has announced and begun to build a new plant. It's my understanding that there will be 3x the growers in the new plant than Cree currently has. So that's a 3x increase from current levels. The new plant is scheduled to be completed within 12-18 months. I've read on the Cree message board that growers will actually be in use prior to actual plant completion. Now, earnings to date have been the result of primarily 2" diameter wafers. The new plant will almost certainly produce 4" diameter wafers. The increase of radius from 2" to 4" increases the area of SiC produced to 12.56 sq in, from 3.14 sq in. You'll note this is a 4x increase in SiC. So we have 3x grower increase and 4x area increase equals a 12x increase in output. (Not to mention the 3" diameter SiC wafers which are just being implemented into the current plant. At this time the effective "yield" of each wafer is currently in "the low double digits" according to Hunter at the last conference call. After rather persistent questioning by Robert Jacobs (Go Robert!) Hunter indicated that a 3x increase of current yields is economically feasible. I don't recall that a time frame was actually stated for the yield improvements. But lets say that yields are doubled in the next year. That would result in a 2x increase of actual useable SiC from each wafer. 12x * 2x = 24x increase in total output. A 24x increase would yield a 6x increase in share valuation. (Let's use $50 to be conservative...6 * 50 = $300. This is when the plant's completed. So..12 - 18 months)
If yields are tripled in the next year. That woul |
