You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We speculate that this is due to dark current being slightly higher than the actual current from the measurement photodiode, due to noise, physical variation across the die, or similar. We think that the Si1133 measures the current from the measurement diode, then measures and subtracts the current from the dark-current photodiode, and doesn't prevent the result from going negative. The "24 bit" measurement mode is in fact a signed measurement (the saturation value is 0x7FFFFF, according to the datasheet), and the dark current is pretty small; we can treat any value in 0x800000 .. 0xFFFFFF as being negative; since light can't be negative, we can substitute zero.
At darker (low-light) region, Si1133 returns channel data as FFFFxx on read. Instead it should return zero.
The text was updated successfully, but these errors were encountered: