The Graben Pressure:. Tank Test

The purpose of this test was to make sure that the old tank could still be pressurized and to find out how long it would take to pressurize the tank using solid CO2. After the tank was cleaned, 343g of dry ice was placed into the tank and the top was screwed on. The tank was set outside and data was recorded using a digital timer and pressure gauge. The results are graphed below. From this test we learned that the tank could indeed be pressurized without failing. Although we could pressurize the tank using solid CO2, we found that it took too long to bring the tank up to operating pressure. From the data we calculated that once the pressure started to increase, it did at approximately 16 psi per minute. So, now if we want to continue to use dry ice to pressurize the system, we need to increase the rate at which the pressure builds up.

Second Pressure Test:

For the second test, we modified the top of the pressure tank to attach the new regulator. The purpose of this test was to measure the rate of pressure build up using a larger amount of dry ice, to check the flow through the new regulator, and to see if we can reach the critical point of CO2 at about 1073 psi. Using about 40 oz. of dry ice, we found that the pressure builds fairly uniformly at about 5 to 8 psi/minute. As in the first test, there seems to be a stop over at 70 psi for awhile which corresponds to the triple point of CO2, then the pressure climbs steadily after that. At around 500 psi, the tank started to warm up and the ice on the outside of the tank melted. The pressure topped out at 820 psi. The tank looked good, no swelling or any funny looking stuff going on. The 1/8" NPT release valve was opened to atmospheric pressure when the reservoir pressure reached 820 psi and we got the following results which are graphed below. Immediately the pressure dropped from 190 psi to 160 psi. It took 6 seconds to drop from 160 psi to 150 psi and took 12 more seconds to drop from 150 psi to 100 psi. It drops steadily for 12 more seconds to about 50 psi and holds for 2 or 3 seconds, then drops toward 0 psi. Total time for all pressure to release is about 50 seconds with 18 seconds at or above 100 psi. From this test we learned that the tank can be pressurized nearly to the critical point of CO2 and that we should have no problem maintaining the system pressure during the burn and beyond.

Flow results of the second pressure test.

Phase diagram for CO2

Phase diagram for CO2

Results from the first round of pressure tests.

Third Pressure Test:

A third round of pressure tests were conducted to figure out roughly how much CO2 we needed to maintain the system pressure during operation. We also wanted to find a faster way to charge the pressure tank, waiting an hour and a half to two hours just seemed to be too long. So, after running numerous tests we found that by adding a small amount of tap water to the dry ice in the pressure tank we could almost double the rate at which the tank charges. We tried different amounts of dry ice and water, but found the best repeatable results were obtained using 2 teaspoons of water and between 400 - 500 grams of dry ice. This combination effectively charged the tank to 500 psi in about 40 minutes. With the pressure tank attached to the t-stoff and gasoline tanks, we were able to conduct flow rate and system pressure test when we discharged the pressure tank. This allowed us to empirically determine the system pressure required for the desired flow rate as well as the minimum amount of dry ice needed to maintain the system pressure. Below are images of the test stand and the data collected for pressure tank charging times.