Report from the CoolingZone Summit, August 2005: http://www.coolingzone.com/summit2005

The focus on high heat flux was timely, as the summit was filled to capacity. What struck me the most was that although high heat fluxes and the potential need for liquid cooling is on everyone's mind, there are only very few suppliers that have products ready for market. Information about component performance is still hard to get in many cases, and the range of market-ready solutions still seems limited. It's a bit of a chicken-and-egg problem, where suppliers won't ramp up production (making their product affordable) until they see the demand for it -- and the demand won't climb if products are still so expensive. The solution to this? Suppliers and customers working together at all levels: purchasing, manufacturing, marketing, AND engineering. Roadmap sharing seems a good place to start.

Another point that struck me was that although liquid cooling for electronics, and especially for computing equipment, has been around for a long time, the demand for it used to be limited to the high end (thinking about IBM mainframes here, which were liquid cooled until the early to mid-1990's -- and were million-dollar products). Now, it seems that the potential need for it exists at many more levels of product, and thus at many more -- mainly lower -- price points. That probably implies a much larger installed base, which carries with it a need for greater reliability at lower cost than the historical implementation range. This isn't to say that the problem is unsolvable, or even inevitable -- just that there are plenty of opportunities for good engineering design!

CoolingZone Summit Liquid Cooling Short Course Feedback

During the short course, I asked the attendees to list their top concerns about implementing liquid cooling. See these on page 2 .

If you were not able to attend, the course will be scheduled again in the fall at a Bay Area location. Contact me for details.

Spreadsheet Tricks for Thermal Engineers : Data Analysis

At the Semi-Therm conference, I did a tutorial on my favorite tricks in Excel -- which, according to Tony Kordyban, is "the toolbox you already own" for thermal calculations. The tutorial covered tricks for calculations and modeling, and others for data analysis. There weren't any handouts, but here is a quick summary of the highlights that I use for data analysis:

• To be handy for data analysis, your data needs to be set up in a way that Excel sees as a database: labels one to a column, data in rows below. There should be no blank spaces and no blank lines below the columns, no merged cells anywhere, and preferably several (three seems to work well) blank lines above the columns. You can have other descriptions above the three blank lines.
• For a quick overview of the data, you can use Data Autofilter. This is described in a previous newsletter.
• For more detailed and statistical descriptions of your data, there are several built-in database functions. More about these on page 2...