The electronic ballasts as evaluated were not designated with a particular ANSI lamp/ballast designation and did not have UL/CSA labeling. All were clearly marked for supply voltage ratings (120 or 240 volts) and did have power cord hookups for the lamp socket. All lamps tested did start on the various electronic ballasts, but some exhibited a swirling effect of the arc known as acoustic resonance. Because the electronic ballasts operate at very high frequency they can cause unstable conditions inside the lamp that can actually make the electric arc swirl or dance during starting. This may be seen as light coming from the fixture moving or flickering, but if the lamp were to be viewed with welders glass (very dark tinted glass) the electrically charged vapor in the lamp would been seen moving around. Acoustic resonance is hard on lamps and may cause premature failures due to the excessive vibrations. Electronic ballast users may need to replace lamps more frequently over time as this condition will eventually weaken all brands of lamps — some sooner than others depending on the construction of the lamp and the actual level of acoustic resonance generated by the ballast. Ballast designers know about acoustic resonance and can design to eliminate it for a specific lamp; but it may be hard to eliminate in all brands of lamps without careful compatibility evaluations.
Ballast efficiencies for tested electronic ballasts were very close to that of traditional metal ballasts. They consumed about 50 to 70 watts each and ran the lamps at roughly their rated 600 watts. This makes the total power consumed by the electronic 600 watt lamp and ballast system to be about 640-670 watts.
Light outputs from lamps tested on both traditional and electronic 600 watt ballasts tended to be about the same. There was no dramatic increase or decrease in light output from any of the ballasts choices; however, all lamps tested on electronic ballasts had advanced sodium loss and reduced lumens.