This week I spent helping one of the engineers gather data for a prototype they’ve been working on. The device is a level sense: it measures how high the water level in a tube is to know how much water is in the container. My job has been to run experiments at different pressure levels to see how the water level is affected by air pressure and water pressure. Some of the experiments were very hands-on in that I took a data point after every change in pressure while others had steady pressure and I took a data point every 10 minutes. This was a little more challenging in trying to multitask: as soon as I would get settled and into a groove with my second task, the timer would go off and I’d have to collect another data point then back to my task then timer goes and so on. But I have been told that the data is good and that the graphs I made were very helpful in the device presentation.

The test set up. Water is pumped into the tubes. The air pressure is controlled by the regulator attached to the tank.

A look at the data screen. The probes inside the tubes send data to the computer which is then graphed.









We added water to the ADCP wells to check for leaks. We had a couple of small leaks so I tightened down some of the nuts around the window and managed to get the leak to stop. We then left it overnight to make sure no more leakes appeared. In the morning we detected no new leaks so we decided we could install the transducers. Once they were bolted in place we could then add our pressurised water to the system. There is space between the transducer window glass and the transducer itself; this space is filled with pressurized water for two reasons. The first the transducer can be damaged if it were to send its signal in the air so it needs water to send its signal. The second reason we used pressurized water is to counteract against the force of the ocean acting against the glass at depth. In order to pressurise the water we have a hose that goes from the transducer well up to a head tank that is filled with water. The elevation of the water creates .433 psi of pressure per foot of water. Our tanks are about 13 feet above the window. This creates about 5.6 pounds of pressure per square inch pushing down on the inside of the window. The pressurised water was added and the fore window started to leak. This was really bad because at this point there is not much we can do to stop the leak. We thought about it over night and we came to the conclusion we would have to pull the whole thing apart again and start from scratch. This was tough because not only was it a week’s worth of work getting it done but we would now be delayed getting the ship put back into the water. So we were going to have to work as quickly as we could and figure out a quick way to test the new window without having to put the whole transducer back in test. It is a trick too because between each step there are different cure times for the adhesives we use which can be over 24 hours. We pulled out the transducer and window as quick as we could and cleaned it all up which was not not easy because it was all secured with a permanent adhesive called 5200. Once everything was cleaned up and a new gasket was created we set to work installing the window. We took special care to make sure everything was perfect. We had inspected the window that had leaked when we had pulled it out to see where it might have failed. It looked like part of the problem was the gasket itself might have been too wide; it was hanging over the edges and this could have kept the window from getting a good seal. For our new gasket we made it a lot more narrow. As we were working on the window we had the shipyard machinist fabricate a metal blank with two air fittings so we could add air pressure as our window test. Once everything was finished we installed the blank and added pressurized air to the window. The tecs wanted to test the window with 10psi. Because that’s what they thought the pressure of the tanks were. We had a small leak so I took off the blank, tightened some of the nuts and reinstalled the blank for another test. Before we did the next test I had been thinking about the pressure we were using and I felt it seemed a little high. So I looked up the formulas myself and re-did the math and I got the 5.6 psi myself which is the calculation I mentioned earlier. So not only was the pressure a lot less than we were testing at but no one took into account the fact that the ship was out of the water so it didn’t have any of the counter pressure of the ocean pushing back. The window normal sits at a depth of 5ft so that’s about 2.2 pounds of pressure pushing back up against the window. while the window is in the water the pressure pushing out against the glass is about 3.4 psi. So I brought this to the attention of the tecs saying that we were putting too much pressure on the window causing it to fail. So instead of testing it at 10psi we tested it at around 4psi. The window passed our last test. We then reinstalled the transducer. And decided not to add the pressurized water till we went back into the water. 