Week Eight

Following the results of the Week 7 testing, the group got to work immediately in Week 8 to complete more testing with the decided improvements.  The thermoelectric cell was applied to the computer with the new thermal paste that was chosen due to its higher percent metal composition.  The application of the Arctic Silver 5 paste, as well as the process of applying the cell and the heat sink to the computer, is shown in detail in the following pictures.  

A small amount of paste is put on the cell.

The paste is then spread to a thin layer on the cell surface.

The processor heat sink of the computer, location for cell.

The thermally pasted side of the cell was then placed on this location.

The silicone based heat sink compound was then spread onto the “cold” side of the cell.

Finally, the heat sink was then placed over the cell.

After this process was completed, the cell and thermal paste were allowed to set for 12 hours, with the hope that the thermal paste would work more efficiently when given time to set rather than immediately testing.

Week 8 testing with multimeter.

Week 8 Test Results:

Using the same testing procedure as in Week 7 with the multimeter, the cell produced a maximum voltage of only 0.3 volts. 

·         The cell voltage improved from 0.2 volts (Week 7) to 0.3 volts, a 50% increase, if at all relevant.

·         The cell voltage was relatively constant at 0.3 volts, whereas Week 7 testing did provided voltages that constantly varied from 0.1 to 0.2 volts.

·         The surface of the computer was still uneven, and more sanding needs to be done to make a flat surface.

While the Week 8 results were improved from the Week 7 results, they were still not as good as expected or desired for the project.  The cells should be produce a greater voltage than they did, considering the heat applied and the presumed effectiveness of the heat sink.

Because of this, the group decided to run heat gun tests on the cells again to see if they work as well as initial heat gun tests.  The heat gun was applied to the cell, however this time with the new heat sink on the other side, producing the most effective heat dissipation and therefore gradient. 

The conclusions from these heat gun tests were that the thermoelectric cells were broken.  The two original cells were tested, the first cell rose to the expected 1 to 1.3 volts, however it rose as quickly as ever, and then it provided an audible “pop,” and then broke, producing no more current.  This was most likely due to the intense heat of the heat gun, and the effectiveness of the heat sink; these were the first heat gun tests with the actual heat sink.  The second cell, the one that has been used on the computers, was also broken, believed to be worn down from weeks of heat gun testing. 

Because of this, the group immediately ordered two new cells, and these cells will be applied directly to the computers, with no heat gun testing.  The heat gun was effective for initial tests of the project to prove concepts, however it is no longer productive to continue heat gun tests.    

Another productive outcome of Week 8 was that the group was finally able to use thermocouples to get accurate temperature readings of the test surfaces. 

Week 8 Thermocouple temperature readings:

·         Computer surface: 56° - 58° C

·         Room temperature: 21° - 22° C

o   A max temperature gradient across the cell (assuming the cold side is at room temperature) would be 34° - 37° C

o   Thermocouple should be used to measure actual temperature gradient, this can only be done underneath heat sink when cell is functioning.

·         Heat gun application site temperature: about 180° C

Several deductions were made from these temperature readings.  First, based on the operating specifications of the thermoelectric cell (see Week 6 page), the thermoelectric cell should be functioning with a voltage output higher than what is being achieved.  Therefore, the cells may have been broken for Week 7 and 8 testing.  Also, the heat gun applies a very intense heat, and can be considered the cause of any broken cells. 

The next step will be applying the new thermoelectric cells to the computer surface once again, this time with the most even surface possible, and with the most efficient application of the thermal paste on the hot and cold sides of the cell.  The group will do this as soon as the thermoelectric cells arrive in Week 9, and the plans and precautions taken should generate better results.