Wednesday, November 12, 2003
Today was the day of dry-erase boards and math. We temporarily abandoned construction for the more mind-bending physical and mathematical properties that are driving our ram pump. Essentially, we have no idea what is drawing our ball away from the gate. Monday we thought it might be due to Helmholtz's Law but we pretty much scratched that idea. Now our unsupported theory is totally based on pressure waves and areas of high and low pressure. Karst waded through some gross calculus and pulled through with potentially our first equation. Working towards eventually getting some sort of working model. Maybe one day. Hopefully before January.
Brian responded to our "thermodynamic emergency" and gave us some stuff to think about. Still unknown: why the ball rolls back even when it is horizontal.
Monday, November 10, 2003
Today during project the group finished waterproofing the resevior and dump tank. The group also deliberated how to make the pump fit the backing, and the merits of copper tubing versus flexible tubing. Simulink models were started today.
Tuesday, November 4, 2003
All the frame pieces were finished today. The frame was then assembled and made to be free standing. We also discovered that the Great Pyramid of Giza may have been an ancient form of the ram pump.
Monday, November 3, 2003
We gave our Presentation on the ram pump. The presentation outlined our project goals, and addressed the tests that we performed. It also described the direction that the project is heading.
Sunday, November 2, 2003
We started to construct our free-standing and self-circulating prototype so that we can run quantitative tests. We also created a PowerPoint presentation for our preliminary class presentation on Monday.
Friday, October 31, 2003
Today we took another trip to Home Depot in order to get the supplies to make our system re-circulate. We bought an electric pump and the reservoirs necessary as well as 1/2 inch copper tubing to replace the flexible tubing coming from the top reservoir. We believe that the copper tubing will not jolt as much as the flexible tubing, so it should allow the pressure wave to work more effectively.
Wednesday, October 29, 2003 – Experiment 6
After several casualties involving our water supply and spilling all over the place (thank you Facilities and Nathan for dutifully cleaning up our mess), we managed to make some progress on testing the parameters of our system. A computer mouse ball turned out to be a perfect fit; however, the mass of the ball is still questionable. With a usable ball we were somewhat able to test our prediction that a larger gate for the wastewater would create a more efficient system. The pump performed its best yet, pumping water a whole 3-6 inches up the out take tube. We were almost able to fill up 2/3 of a Coke bottle from one trashcan of water. Sweet. We also discovered that the angle of the waste gate definitely makes a difference in the strength of the pressure wave. The next step is to build a frame and a way to re-circulate water for the system so that we can begin quantitative experiments. Below are pictures of our setup for Experiment 6.
Monday, October 27, 2003 – Experiment 5
We discovered several things in our experiments today. After making a quick run to the toy store, we tested the pump with a bouncy ball. Unfortunately, the ball floated in water and refused to work unless we angled the gate upwards. Then, however, air bubbles filled the system and basically killed the pressure in the tubing. Thus, our conclusion is that the ball must be denser than water and air bubbles=bad news.
In trying to figure out how to optimize the ram pump, we have come across the paradox that a larger gate/hole size actually produces less wastewater and more usable water per trial. We are not sure why this might be true. After much discussion, we decided it was something weird that we haven’t learned yet. More tests are necessary.
Sunday, October 26, 2003 – Experiments 3 and 4
Our biggest concern from Experiment 2 was that the viscosity of the water in the narrow output tube was actually limiting the amount of water that we could get out. We were worried, however, that an increase in tube diameter would nullify anything gained by decreasing the friction caused by viscosity. So, we installed a new 3/8'' diameter tube and set the pump to work. Water came out of the output tubing much faster than it did with the 1/4'' tubing. So, cool, the viscosity was partially to blame.
In the next experiment, we changed the diameter of the waste tubing (the out take tube) from 3/4'' inches to 1 1/2''. This in itself isn't anything that should change the way the pump works. Instead, it's the size of the hole in the cap of the gate that determines the flow rate. To accommodate the new, larger, cooler tubing, we chose a new ball (a piece of a Drimmel tool sawed off to fit our needs). We ran the experiment. Unfortunately the bastard didn't work. The ball valve works only once (i.e. it never resets itself). So, it shoots water up the out take tubing and then just sits there. So, we went to the white boards and tried to figure out what's wrong. We came up with the following hypotheses:
- The larger hole in the gate screws up the diffraction in the pressure wave.
- The leakage around our imperfect sphere is preventing a pressure wave from even occurring.
- The thicker tubing somehow absorbs the pressure wave. Experiment 5 will hopefully discover the cause of this problem.
From Experiment 4, we have some concrete things that we need to figure out. We changed the size of the waste gate to exactly 3/4'' inside diameter. We had another go at the experiment. With the new, clear tubing we can see exactly what is going on inside the valve. Namely, we can see when the ball is doing its job (sealing off the hole and then moving off) or when it is completely not working and just sitting there instead. Oddly, it works sometimes and sometimes it doesn't. We saw that whether the pump works or not is basically dependent on which side of the ball comes in contact with the seal. So, we concluded that imperfections in the ball (it is, after all, a drimmel bit wrapped in teflon tape) are letting too much water past the waste valve to create a pressure wave.
Saturday, October 25, 2003
The group went on a massive shopping trip to Home Depot. We bought materials to test having a larger waste tube. The results should be interesting.
Friday, October 24, 2003 – Experiment 2
After installing the new self-timing mechanism, our ram pump runs completely under its own power. The self-timing mechanism is simply the ball being pushed by the water to close off the gate and then rolling back, allowing water to flow through the gate. After priming the pump, a gentle flick to the intake tube starts a process that is only limited by the the amount of water one trashcan can hold. From this experiment, we found the following problems:
The clear tubing for the intake tube is not the best possible material. It shakes as the pressure waves travel through it, absorbing energy that could otherwise be pumping water to our destination.
The diameter of the out take tubing is too narrow. The viscosity of the water is visibly limiting the flow of water up the tube. A larger diameter tube will (hopefully) increase the flow rate. However, this larger diameter will also make it harder to push the water up. More data is needed to decide the best dimensions.
Wednesday, October 22, 2003 – Experiment 1
 |
Having gone to Home Depot earlier in the day, we used our newly acquired parts to build our first ram pump prototype. Using two trashcans, a good amount of water, and the stairs in an undisclosed building on an undisclosed college campus, we ran our first experiment. Running the waste valve by hand (slamming our hand down over the hole, causing a massive spurt of water to cover us from head to waist), we succeeded in pumping water to twice the drop height. The picture at the left is of the waste output, which contains the ball in the white PVC tube. |
Monday, October 20, 2003
We prioritized our project goals in the following manner:
- Figure out the self-timing mechanism using a ball
- Decide on final dimensions
- Construct a housing structure
- Start Simulink
- Begin relevant math and calculations
We also concluded that we would need to buy a clear one-way valve, reservoir tanks, pump, clear PVC, and a pressure bulb.
Friday, October 17, 2003
We began discussing the design of the mini ram pump and created a timeline
of what needs to be done for the project.