If the current/wattages are correct, nothing. But be aware, then the power factor of the lamp is about 90% due to voltage vs current waveform shape mismatch (current is nearly sinewave, voltage squarewave), so if you measure 80Vrms and 0.5Arms, the lamp wattage is only 36W and not 40W. With voltage measurement you have to take care of the correct shape calibration coefficient (depend, how the meter exactly work), if you are not using true-rms meter. Most non-truerms are using "rectified average" with build in calibration factor of Pi/sqrt(
= 1.11, so to recalculate the voltage reading, you have to multiply the measured value by sqrt(
/Pi = 0.9 in order to get the rms value of the nearly square wave voltage.
CWA would perform very poor - you would need to lower the voltage very extensively (out of ballast's reguation range) to get some current reduction, so this ballast is totally unsuitable for any such variation.
But there are questions about setup stability. Using "higher power rated" ballast with lower mains voltage mean, the OCV would be lower and ballast impedance higher.
Generally the setup would be more sensitive for power dips.
For LPS i would expect harder starting, but if it start, it should work good.
For HPS, there are two effects going opposite to each other:
- The ballast lower impedance would cause the system to be thermally more stable, what should avoid the thermal runaway (root cause of arc voltage rise, so consequently cycling at EOL) with more worn-out lamp, what would mean longer effective lamp life.
- But the lower OCV mean the lamp will extinguish when the arc voltage reach lower value, what mean it would start to cycle earlier.
