Welcome to my REU web page.

Quick Links

•  Dr. Alfonzo Ortega
• Experimental and computational heat transfer group
• REU students and home pages 2004
• Weekly Progress Reports
  

The apparatus and what we are pursuing.

The experiment was originated by a graduate student in 2001 named Lall. Balwant Lall was a Ph.D. student whose theses was to study the behaviors and extremities of the heat as it is dispersed from a discrete heat source in a shallow enclosure. His heat source is a 1" x 1" heater that sits on a board that is 6 times the area of the heater.
He used three different boards for his study.

1.  The first was a board made of Balsa wood. This material would be an insulator and would not change in temperature relative to the heater. The board is 25 mm thick
   
2. The second board was an FR-4 board. The board is 1.57 mm thick. This material is  a fiber impregnated epoxy. It's frequently used for computer boards.
3.  The third board is an FR-4 board, but this one has a layer of copper on one side. The FR-4 layer is 1.57 mm thick and the copper layer is 0.036 mm thick.

The results of this study were seemingly very successful, until recently when a research student named Jeremy Evans wanted to mathematically model the results. The Balsa board and the FR-4 board were modeled with out variation, but the FR-4 plate with the copper layer didn't mesh. The error was consistent, possibly just a mistake with the numbers. Our goal is to verify the original numbers or verify the numerical error. In the case of the FR-4 and copper board we will try to use the same board and components used by Lall in the original study.

* Jeremy Evans numerical study of conjugate natural convection from a discrete, heated plate. IBM- Tucson 2003
In this graph P is the predicted numerical data, and experimental is the data collected and reported in Lall's dissertation.

The apparatus.

We start with a cube made of cold plates. They look just like this:

They are copper plates with channels that water will flow through as to maintain a cool temperature. The outside of these walls are made of plexiglas and the inside is painted with a black spray paint to diffuse radiant exchange. Partial assembled, the cube of cold plates look like this:

 


In addition to the retesting of the FR-4/copper board, I will be designing, fabricating and testing a board made from the material called Renshape. Renshape is an insulator made from a urethane but has the same texture as wood. It's also extremely easy to machine. This is a block of Renshape.

My board (the renshape board) will fit inside the cold plate cube and will also have a 1"x1" heater at the center, but the board will be 1" thick and the heater will sit above a cut out square filled with styrofoam beads. This will help to keep the heat transfer to be limited to above the heat source because stagnant air is a better insulator than the Renshape, thereby giving more accurate results.

In Lall's original paper he has included surface temperature profiles for all three of his boards.
*Balwant S. Lall Ph.D. theses in mechanical engineering
University Of Arizona
*The Balsa Board


*The FR-4 board


* And the FR-4 with copper board

 The first thing that you can see in the relationship of these graphs might be that the board with a copper layer has temperature increase spread over the entire surface, where the other two boards mainly have temperature increasing directly from the heat source. Because copper is a good thermal conductor the heat conducts over the surface of the copper. The other boards though (the FR-4 and Balsa wood) are insulators, so the heat doesn't transfer by conduction to the board as readily. All these graphs measure the surface temperature only, with thermal couples that run through the middle of the board from one side to the other.







Results:

FR-4 with Copper                   Renshape