CFD Modelling of a Two Stroke Engine to Predict and Reduce Short Circuit Losses

Paper #:
  • 2015-32-0702

  • 2015-11-17
Garg, M., Kumar, D., Syed, M., and Nageswara, S., "CFD Modelling of a Two Stroke Engine to Predict and Reduce Short Circuit Losses," SAE Int. J. Engines 9(1):355-364, 2016.
One of the major reason for lower efficiency and higher unburned hydrocarbon and carbon monoxide emission for two stroke engine is short circuit losses during the scavenging process. An attempt has been made in this study to understand and improve this phenomenon. A three dimensional transient CFD model is developed for a loop scavenged, Schnullar type, 70 cc two stroke engine. Three major processes, namely, blow down (expansion); scavenging and compression have been modelled. The model is validated with PIV measurement done in motoring mode. Model is also validated with experimental data for trapping efficiency with Watson method and for in-cylinder pressure during expansion, blow down and intake events. A good correlation is observed between experimental and simulation results. CFD model is used to quantify various parameters, such as, delivery ratio, trapping efficiency, scavenging efficiency, and amount of fresh mass short circuit at different load and speed points. Around 20 to 25% short circuit losses are observed at all the load cases analyzed. The mechanism for this short circuit loss is understood with the help of a passive scalars based visualization method. Two major root causes identified are, port design and exhaust pressure wave dynamics. Three different designs of ports are analyzed. With the best design, the short circuit loss is reduced by 12%. Same design has been tested on a chassis dynamo-meter, that resulted into a reduction of 10% in HC (hydro-carbon) emission in a mass emission test on a drive cycle and 10% to 25% improvement in WOT (wide open throttle) wheel force in mid and higher speed operating points.
SAE MOBILUS Subscriber? You may already have access.
Members save up to 40% off list price.
HTML for Linking to Page
Page URL

Related Items

Technical Paper / Journal Article