Experiments were conducted with a commercially available six-cylinder water-cooled turbocharged direct-injection diesel engine. The cylinder head was modified to permit access to the combustion chamber with an engine videoscope. The engine was operated with base diesel fuel (BP-15) and other blends, base diesel with 20 wt% biodiesel (B-20) and with 20wt% diglyme (O-20). A neat biodiesel (B-100) and a 95 wt% blend of diglyme with base diesel fuel (O-95) were also considered. These fuels were used for observing the effect of the fuel properties on injection timing, heat release, flame structure, and luminosity. All the tests were performed with the engine operated at light load (61 N m, 10 per cent of the rated load) and 1800 r/min. Visualization showed that the start of injection occurred 0.4° earlier with B-100 than with BP-15. B-100 showed the earliest start of injection among the fuels. An earlier start of injection was also observed with B-20 and O-20 blends compared with BP-15 fuel. Combustion analysis showed a lower premixed combustion heat release rate with the diglyme blends compared with the B-20, B-100, and BP-15. The highest premixed burn peak and the lowest premixed burn peak were observed with BP-15 and O-95 fuels respectively. It is difficult to distinguish between the spray flames of BP-15, B-20, B-100, and O-20. However, with much higher oxygen content in the O-95 fuel the natural luminosity of the flame was too faint for detection with the camera. The combination of combustion analysis and in-cylinder visualization employed in this study provides a unique opportunity to understand how oxygenates behave in a commercial engine.