I wanted to start posting little tidbits about development we’re doing on Trident on the forums, and I realized that a quick set of tests I did recently to see what size of payloads we can fly with would be a great start.
Alright- I had some fun with this.
At first, I attached just an empty GoPro case (mostly filled with water to make it neutrally buoyant - I’ll use an actual GoPro with supplemental flotation later) to the bottom of the ROV as shown. To my delight, Trident handled it very well, and I could hardly notice a difference in performance. This was all at night which makes it harder to notice subtle details, but I’d say the only noticeable difference is that the pitch was a bit off, but that was easily correctable by adding a slight amount of vertical stick- almost none at all.
Since that went so well, I figured “ah heck- let’s put something bigger on” so I hot glued on a plastic box I found (which would also be flooded) and took that for a test drive. Okay- well, this configuration didn’t work so well. If I were flying very slowly I could just about make it across the length of the 20’ pool without hitting the bottom or the surface, but for the most part it pitched up (yes, up) like crazy once any speed was added.
At this point I was on a roll, so I rummaged around for other stuff around the lab to glue on the ROV and found a Pelican 1010 case. I attached this case roughly where the GoPro was and let it flood. Despite the noticeably larger size, this test also went very well. I was certainly able to notice that the ROV was more pitch sensitive (I couldn’t decide if it wanted to pitch up or down) but I was able to control it pretty well on the stick. At low speeds I didn’t even need to use the stick very much and the whole thing flew pretty gracefully in straight lines. When I turned, I noticed a lot more roll than before. Anything but slow turns will be a bit sloppy, but still very doable.
Finally, I decided to try an idea I had been thinking about for a while which was to mount the big box (which has the dimensions of 185 x 120 x 45mm) to the back of Trident, behind the fins. I used some scrap acrylic and a bunch of duct tape to do this. This is the test that really got me excited- Trident flew beautifully! As I suppose one might expect, it turned a little slower than usual but certainly it was fast enough - maybe it would take 1.5 seconds to turn 90 degrees. Where it really shined was the tracking during forward flight. I guess this is also intuitive, but having the drag behind the vehicle and aligned with its CG along that axis caused it to fly like an arrow. I could point it to a part of the pool (both in heading and pitch), and by only pushing on the forward joystick it would go straight there. I envision that for larger payloads such as side scan electronics, fancy scientific sensors, acoustic positioning transducers, etc, there could be a bracket that screws into the bottom and runs straight back for things like this to mount on. I don’t see any reason payloads couldn’t even be comically large compared to the vehicle- it’s got plenty of power and the performance falls off pretty gracefully. I half considered taking an empty Trident shell and and taping it onto the operational prototype backward just to see what would happen, but then I realized I was very hungry for dinner and that the FroYo place up the street was about to close.
I plan to do more precise testing once we have dynamically accurate prototypes ready (this prototype has a different CG and slightly different geometry than what we’ll ultimately have) but I am feeling pretty good about attaching stuff to Trident and and doing transects. Especially if we can add some hydrodynamic cowling, it seems that even fairly large items can get strapped onto the bottom without major issues.