Analysis of Dimethyl Ether (DME) injection spray characteristics in comparison to diesel and theoretically predicted spray characteristics

摘要

The objectives of this research are to compare the spray characteristics of diesel andudDimethyl ether (DME), in order to investigate parameters which inuduence the sprayudcharacteristics such as penetration, cone angle, and spray velocity. A standardudcommon rail system was used to inject high pressure fuel into a constant volumeudpressure chamber. Tests were conducted at 300 bar to 500 bar injection pressure, andudatmospheric pressure to 17.7 bar chamber pressure using carbon dioxide as anudambient gas at room temperature. A Z-shaped schlieren system with a high powerudLED was used to form images for the high speed camera operating at 20 000fps. Anudexperiment controller was constructed to simultaneously control the triggering of: theudinjector, the high speed camera, and the recording of pressure data. The diesel andudDME spray penetration and tip velocity was found to increase with an increase inudinjection pressure, and the opposite result was seen with an increase in chamberudpressure. DME consistently penetrated less than diesel throughout the testsudconducted, the di erence in penetration was larger at the lowest chamber pressures.udThe di erence between DME and Diesel decreased at higher chamber pressures to 6 -ud9%. The cone angle for DME was consistently larger than that of diesel for all theudtests conducted. Both fuels showed a relatively linear increase in cone angle with anudincrease in chamber pressure, with the exception of the tests at atmosphericudconditions. The DME had a tendency of udash boiling at these conditions resulting inuda large cone angle. Experimental results were compared to theoretical predictions ofudpenetration and cone angle. The experimental penetration for diesel and DME wereudreasonably predicted by some of the theoretical models at the higher chamberudpressures. No correlation was found in terms of injection pressure for the diesel coneudangle, whereas the DME showed a slight increase with an increase in injectionudpressure. While analysing the results a hesitation in the intial 0.2ms was observed,udthis was found to be due to the placement of the single-hole in the nozzle and theudresultant pressure distribution around the injector needle. This hesitation wasudfactored out when comparing penetration results to theoretical predictions.