Our Aquarium Computer

On April 8^th 2008, the A+ class built the computer inside the 5 gallon fish tank. We filled the tank up with 3½ gallons of mineral oil at 3:30PM. The temperature of the oil was at room temperature.

The computer specs are:

Processor:

                              Pentium III – 450MHz (Slot 1 - Core speed 448.1 MHz - Multiplier of x 4.5 - Bus Speed 100 MHz)

16 KB of L1 cache with 512 KB of L2 cache

Motherboard info:

Manufacturer: Gateway

Model: Tabor MP440BX

Chipset: Intel - i440BX/ZX Rev. C1

Southbridge: Intel - 82371EB/MB (PIIX4E/M)

BIOS: v4M4PB0X1.15A.0029.P17 (08/04/1999)

RAM:                   

640 MB of SDRAM

100 MHz Frequency (CL - 3 clocks, RAS - 2 clocks)

                    250W power supply

                    20G hard drive

                    Also included are:

5 gallons of mineral oil at $12 per gallon

                              1 - 5 gallon fish tank for $40

We will be keeping daily updates for the progress of the computer. At the end of the quarter, students will write a report of the project as their final exam. All of the temperatures are in Celsius, so here is a chart that converts it to Fahrenheit.

Day 1: Tuesday April 8, 2008

Students installed the components of the computer into the aquarium to bring the computer into functional order. Once the computer was able to boot up without any warnings or errors, we began the installation of Windows XP. While the installation was going, we went and purchased the mineral oil. Returned back to the classroom and poured the mineral oil into the tank. We filled the tank up with 3½ gallons of mineral oil at 3:30PM. The temperature of the oil was at room temperature.

We tested electrical pulse to see if there was any electricity in the oil using a volt meter. The result was zero.

Temp after running for 30 minutes was 22.2°C (72°F)

Click on picture to view them.

Day 2: Wednesday April 9, 2008

 Temperature at 9:00 AM – 102° F (39°C)

Sometime around 9:45 AM the system locked up so we had to power off and reboot

Added separate fan to circulate the oil at the surface

Submerged hard drive into the oil and blew the power supply fuse

Added another PS  (200W). We removed the power supply from its case so it does not have any fans running on the power supply itself.

Running CPU burn in to keep the system load at 100%. Temp running at 41°C (105.8°F) after 1½ hours.

Day 3: Thursday April 10, 2008

After running the circulation fan overnight, the oil temp was brought down to 37°C (98.6°F). This shows that by circulating the oil even a little, helps cool down the temp.

At 9:00AM we turned on the burn-in process to see how hot we can bring the running temp under full system load by the end of the day. If this brings the temp up to 42-45°C (107.6-113°F) by the end of the day, we would need to find a way to cool the system down while running on a full load.

After running at 100% for an hour the temp went up to 41°C (105.8°F)

Hour 2 brought the temperature up to 42°C

We inserted a new fan to try bringing the surface temp down. This brought it down to 40°. This drop could have been because we had the top off of the aquarium. Running the CPU at full load still to see how the new fan works.

The temperature as of 1:00 PM is 41°C.

At 2:00 PM the temp was at 42°C.

Around 2:30, the system was powered down to do some hardware mounting to the back of the tank. After mounting the hardware and rebooting, the system was cooled down to 40°C again. We are going to let it run over night without any system load to see where it cools back down to.

Day 4: Friday April 11, 2008

We were surprised to see the temp down to a cool 33°C this morning.

Here are some new pictures of the system.

As of 10:00 AM the temperature was up to 35°C.

At 11:58 AM the computer blue screened with a message stating that there was a hardware failure. This may have been caused by the RAM burn-in test. Rebooted the system. System came up without errors.

Day 7: Monday April 14, 2008

This mornings temperature was 34°C. The system was still running as it is supposed to.

We have started the updates installation to install updates for Internet Explorer and a few other updates.

Running the CPU burn-in all day has brought the temperature up to 41°C.

At the end of the day, 3:00 PM, the temp was at 42°C.

Day 8: Tuesday April 15, 2008

After running the burnin process for 24 hours, the highest temp was 44°C.

We did not do much to the system today.

Day 9: Wednesday April 15, 2008

The temp this morning was back down to 38°C but the room temp is at 84°F this morning.

We are going to install a game on it today and play it during the A+ classes to see how it reacts. We will write in which game they decided to install later this morning.

We have missed a few days of updates here.

We have done some modifications.

Day 15: Monday April 21, 2008

We added in some rock and plants to give the feel of an aquarium.

Here are some pictures.

Day 17: Wednesday April 23, 2008

After running the burn-in test for 48hours, the system did not exceed 43°C. There are two fans circulating the liquid at the surface and blowing bubbles down into the bottom of the tank. We have concluded that the bubbles are bringing some of the cooler air from the surface down into the oil and heating up and bring the warmer air to the surface.

Here are some clearer pictures.

Day 31: Wednesday May 7, 2008

After running the system for a month and a few mishaps with power supplies, the system is still running strong. We have not done much to it since the last update.