Water Pump
Overview
For the Mechanical Synthesis class I took in the spring of 2021, I worked in a group with five other students to design and manufacture a water pump. We were tasked with moving water up at least 1.5m and at a rate of at least 1 liter/min while also staying within the provided size and financial constraints. As the team leader and design coordinator, it was my responsibility to track the design process and revise all of the parts in CAD before they were sent to be machined, make sure the team met deadlines, schedule weekly meetings, and communicate with course staff.
Design Process
Following the brainstorming session, we were placed into groups of six and shared our ideas in order to come up with a consensus. The other two pairs both came up with a scotch yoke system that operated two pistons on opposites sides of the drive shaft. We decided on the scotch yoke design with two pistons due to the fact that buying three large syringes would strain our budget and the size of slider crank needed to pump 1 liter/min would put us outside of the size constraint. Our first assignment as a team was to create a team charter to determine the duties and responsibilities of each member. Due to my leadership experience and CAD abilities, I was elected as the team leader and design coordinator. Once the team charter was finalized and signed, we submitted it and I got to work on some preliminary CADs for our design to present to the teaching staff for feedback.
Our presentation to the teaching staff turned out to be very beneficial as they pointed out that our pistons were currently not mounted to the backplate in any way and that our current CAD would impede us from generating part drawings as they were all one piece. This necessitated a complete redo of the CAD in order to accommodate the mounting and the part drawings, but as the team leader I was able to create a schedule where every member could contribute to the CAD and revise the part drawings before the deadline for machining parts. Below is a render of the final CAD along with a few of the main part drawings we used for machining.
As you can see we reduced the size of the backplate to save on material costs and allow the mounting of the pistons to the sides of the backplate. We also changed from a circular rotational piece to a smaller rectangular one as it was easier to machine and allowed for clearance for the four mounting holes. Lastly, I cleaned up the CAD by adding hex nuts and the appropriate threaded rods.
Manufacturing
Our manufacturing process began with us creating a bill of materials and ordering all of the supplies needed. We then split up the part drawings amongst the team members that had shop training. As I was studying remotely, I was responsible for signing team members up for machining slots and generating the shop drawings for the appropriate part along with adding dimensions and notes to facilitate the machining. For more detailed information on the machining and assembly of the water pump, you can download the full report under “Water Pump Report” at the bottom of the page. Below is a video of the assembled water pump being tested.
Testing and Conclusions
Our water pump was tested by attaching the faceplate to the backplate and connecting the intake tube to the left piston and the outtake tube to the right one. Unfortunately, our pump was unable to be operated during the testing period, so we do not have actual data of how our pump performed. Our issue was a mounting problem. We did not realize that the screw heads from the screws used to mount the backplate to the face plate would interfere with the rotation of our scotch yolk center. The clearance issue can be clearly seen below in Figure 6.
If the pump had been attempted to be operated, it would have violently collided with the screw heads and likely caused a catastrophic failure. If the group had tested the pump’s motion with these screws, the problem would have been easily identified. The solution would be to counter bore the thru holes on the backplate so that the tops of the screw heads would be flush with the face of the backplate. The group also should have read through the specifications for the project more thoroughly to ensure that something like this wouldn’t happen. In hindsight, it makes complete sense that if thru holes were needed, there would be a screw head that we would need to account for.