Day 5 – July 14 – Et tu, IMU?? (or: More Heartbreak…)

Posted by admin-robotlord on July 14, 2011

Well, it’s day 5 and it’s been a lot of highs and lows, mostly with us ending on a low note.

We arrived at the TRANSDEC bright and early and eagerly unloaded our equipment and set to work. We were trying to remain optimistic that today was the day we were going to get through that gate! Alas, SubmURSA had other ideas. A preliminary test of the plumber’s putty in the hotel pool the night before indicated that it should work in practice as long as we placed it on correctly. However, as we were sealing the lid for a test in the testing pool, we heard a sickening ‘pop’ as one of the draw latches tore it’s mount from the pressure hull. Undeterred, we substituted a C-clamp as a temporary fix and got it in the testing pool. This is where we discovered our second problem.

After adjusting the pump nozzles to push SubmURSA underwater, and statically balancing the hull (with more C-clamps, those things are handy…) we tried to have it submerge itself using the pumps. The pumps worked well enough, but the problem was that once we had it underwater and shut off the pumps, it was unable to return to the surface on it’s own. It was completely neutrally buoyant! We spent some time adding weights and floats in various permutations, discovering it was a very fine line between negative and positive buoyancy. As we were attempting to achieve the perfect balance, another of the draw latches broke loose (2500 psi shear strength my a**…)

Then (oh, joy) we discovered that the depth sensor and IMU were no longer working, in spite of them working fine the previous night. So now, SubmURSA had no idea how deep it was, or which way it was pointed. We would have to do our run blind as we had no usable external sensors. For regular readers, you would know that we have two USB cameras on the front of the vehicle that would allow us to see the gate, but it was decided that making the system more complicated at this point in the competition would not be in our best interests.

So we spent a few hours tweaking the pumps to get SubmURSA down to a fixed depth and traveling in a straight line. Needless to say, SubmURSA wasn’t co-operating very well. It was getting down to the last minute before our qualifying run. It was tense. I was in the testing pool with SubmURSA and Veselin was on the side furiously making adjustments to SubmURSA’s autonomous program. We could get SubmURSA to dive, but no matter how much we increased the left side thrust, it wouldn’t track straight. When we checked the pumps on the left side, I discovered that none of the left-side horizontal pumps were working anymore. Maybe it was a dead battery, maybe one of the motor controllers, but we didn’t have time to pop the lid to check. I made a valiant last-ditch effort to move the hose from the right front thrust nozzle, to the left front thrust nozzle (the beauty of our modular design) but it was too late…we missed our chance. However we did learn a few things, particularly that if we wanted to have SubmURSA surface on it’s own, we needed more buoyancy, which meant more downward thrust than the pumps could provide. This lead to the decision to use the thrusters for diving, entailing major modifications to the platform. But first, food.

So, back at the hotel, we re-epoxied the draw latch mounts on and headed out for dinner. After getting turned around a couple times, we finally made it there (in spite of 3 backseat drivers yelling different directions to Meaghan the driver). We ate and then returned to the hotel for a long night of work, with the hope that we could do a quick water test before bed.

Once there myself, Allen, Aassem, and Meaghan went to work to figure out how to add thrusters to the side of SubmURSA, as two thrusters would only require one motorcontroller versus two for the pumps. Chris went to work adding a start switch to the mainboard, and Lee and Kris went to bed. We figured out how to get the thrusters on (yay zipties!), but we did have to modify the thruster cable to get the required length. So, we set that up to dry overnight, Chris finished the start switch, and we decided to call it a night, scrubbing a pool run for the night, and hoping for a better day tomorrow. We WILL make it through the gate!

~ Mike


Leave a Reply

Your email address will not be published. Required fields are marked *

linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram