Kevin VanDelden '17 & Leonardo Rubin Comin, mechanical engineering

The purpose of our research was to measure the fast changes in pressure that occur in a diesel cycle piston-cylinder system and to correlate the pressure to the crank-angle of the shaft itself. This would enable the creation of pressure-volume (P-V) diagrams to help understand the combustion process in small diesel engines and contribute to modeling the particle emission data collected. The engine being tested was a Yanmar L100EE, single-cylinder diesel engine. To measure the pressure inside of the cylinder head, we chose to install a PCB piezoelectric pressure transducer. We decided on a BEI HS35 optical encoder to measure the crankshaft position.

 After researching pressure measurement in engines, we found that there were a handful of key factors to consider. The effect of thermal shock causes the diaphragm of the pressure transducer to deform due to the different expansion rates of the cylinder head and pressure transducer itself. This deformation can cause inaccuracies in the data. We did research to find thermal images of a typical diesel engine head, which allowed us to identify the regions of the head where there would be less thermal shock. The coolest region was near the intake valve of the head. We also found that the area immediately surrounding the fuel injector would see the hottest temperatures during combustion. These facts, coupled with machining restraints of the top-side of the head, narrowed the placement of the pressure transducer down to one region near the intake valve. Because the transducer was very deep inside of the head, we had to manufacture our own installation tool. This tool allowed the device to be screwed into the head at a specified torque while being connected to its cable. The installation of the pressure transducer was overall very successful. Because we isolated the cables of the transducer from the engine, the vibrations of the engine did not interfere with the data, and we did not experience much experimental noise.

We determined that for ease of installation and service, the optical encoder would be attached to the flywheel end of the engine. This involved removing the starting cord and designing and manufacturing a mount to securely fasten the encoder to the engine. The final design included a hollow shaft that could fit over the fastening nut of the flywheel and a new flywheel cover. The cover included holes to allow for cooling air-flow and would enable access to fasten the O-ring that attached the encoder to the shaft. Unfortunately this method of fastening was not as successful as hoped, in our data we found that there was instances of slipping of the encoder that occurred cyclically. We were able to accurately investigate single cycles of the engine and create PV diagrams but we were unable to compile data from multiple cycles, which was the original design.

Advisor: Professor Peter Stryker