Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Hewlett-Packard (HPQ) -- Ignore unavailable to you. Want to Upgrade?

To: PCSS who wrote (1630)	8/8/2002 4:22:27 PM
From: PCSS	Read Replies (1) \| Respond to of 4345

..............PART 2 of 3 HP engineers still face a number of tricky engineering hurdles. Inkjet printers deposit ink on paper. What HP is proposing is to reuse the fluid, which will likely mean building a tiny reservoir into the processor die itself and then engineering a circulation system of some sort. A far more serious problem is one of fluid thermodynamics, far away from HP's traditional expertise in electrical engineering. Inkjet printers use the finest mist possible to increase resolution. But if the fluid is too fine, the particles won't have enough momentum to actually overcome the rising heated air from the chip, said Ratnesh Sharma, also a member of HP's technical staff Worse still, heavier droplets will actually pool on the chip's surface. While the drops will initially cool the chip, the pool actually serves as an insulator, again allowing the heat levels to rise. Researchers said that HP will likely use some sort of inductive fluid, not water, which won't short out any electrical connections. In addition, if the thermal density pattern on a chip varies, then the cooling must change as well. HP's Sharma demonstrated a system by which different micronozzles in the cartridge pulsed a fine mist over different areas of the chip. The individual droplets are picoliters in size. "It's like taking a shower in the morning and saying let me get more water on my nose and ears instead of a constant stream of water," Patel said. Finding a chip to apply the technology to is another problem. While HP co-designed the Itanium processor and its own PA-RISC architecture, the company is phasing out its proprietary lines and buying Intel's chips instead. HP might not be able to convince Intel to ship it bare processor dice, allowing it to add its own cooling package, executives admitted. Instead, the technology might be more practical in laser diodes, used in optical communications, they said. Sensing the patterns of heat is difficult as well. HP's software engineers can only predict so much. Processors today are typically monitored by a single thermal diode embedded in the chip. HP's system could require several, although they would also need their own I/O pins in the chip's package. "What we need is real data," Bash said. However, design engineers haven't traditionally made thermal issues a priority. In an HP chip, "we had to fight to just get one (diode)," he said. "They screamed."