On this page we provide our results and the final conclusion, we like to say at this point once again, that the page should not be used as a research page in the sense of just copying the data and the result but further more as a helpful link to your own research project.
The procedure appears deceptively easy. Sealing the engines and regulating the heat source was difficult. The 5.5 cm and 6.5 cm engines were the only ones that data could be derived from. The epoxy of the other engines, despite its heat rating, was consumed in the heat and fire and the engines lost their integrity. We buffed off the old epoxy with a Dremel© tool and reapplied; however, the engine must have been wet because it did not adhere to it very well. We applied a third coat after heating the three engines with a blowtorch but to no avail. Also, the alcohol flame underneath the engines blew around the engine and was not centered underneath it. So, beyond the first two engines there was no data. The graph appears below.
Above trendlines have been added to show the average slope of the line. The coefficient before x is the slope and the degree of proportionality. It seems to show that that decreasing the length of the exhaust tubes allowed for a slower engine performance. The distance was held constant during each run. Although speed cannot be calculated over such a distance since it is hard to know the point it has stopped accelerating, the time it takes can be a reliable comparison. The fluctuations in the data are also problematic. Each point is the average of four runs so at first glance it appears to have a tolerance to experimental errors; however, it may be a combination of factors. The flame itself did not remain underneath the engine to provide a constant heat source, and the boat often turned between its two guides to slow it down or bind it up. We tried to throw out these obviously faulty runs but the friction was still a factor. The picture below is of the guidance system used with the boat.
Since everything was not working, we decided to go with another engine design. We retested the engine with the pipe caps and found that on closer inspection it did work, but only with a hotter, more concentrated heat source. For the heat source the blowtorch was used on a common heat setting. The flame could be focused on one part of the engine and was not affected by air currents. The sizes went from 4.5 to 6.5 centimeters in ½ cm increments. The first round of tests with no trimming of the exhaust tube went without a hitch until the solder started to melt on the 6.0cm engine and the 6.5cm engine. It seems that they are simply too long so they must get to such a high temperature to operate. The hotter the copper the higher pressure the steam makes, but in a larger engine this higher pressure is needed to compensate for the larger size. The engine does pulse but it does only that and no forward movement can be detected. It seems to be sucking in its exhaust. This is probably due to the flame not being hot enough.
|
4.5cm |
5cm |
5.5cm |
6cm |
6.5cm |
|
3.7625 |
3.7225 |
3.32 |
|
4.36 |
The table above shows the average time needed to complete each course. The 6cm engine lost its seal and ended up shooting its pipe cap about 30cm on one of the engine pulses. Very cool, but not good for data collection; however, the trend does seem to show there is a preferred engine size around 5.5cm. After these tests the blowtorch began to have problems with ice from the gases decompression, and the effect of the exhaust tube length could not be tested.