0

For my application, I need to detect two light levels: pitch black (close to 0 lux) and a low ambient light level of about 0.1 lux (or higher). It seemed like the BH1680FVC ambient light sensor was a good option because it can detect the illuminance from 0lx to 50000lx with a photodiode and built-in amplifier (https://www.farnell.com/datasheets/2265813.pdf). However, the problem is that I need a faster response time- when the light level drops from a high level to 0 lux, I need to measure 0V after about 0.5ms. The fall time was very slow when I tested it and it took about 10ms to drop down to 0V. The response time isn't even on the datasheet.

Now I'm looking at ambient PIN photodiodes that have fast response times. These two parts seem like the main available candidates:

  1. SFH 2270R (https://www.digikey.com/en/products/detail/ams-osram-usa-inc/SFH-2270R/4879477)
  2. TEMD5510FX01 (https://www.vishay.com/docs/81293/temd5510.pdf)

The problem is that the SFH 2270R datasheet says "Recommended operating range for illuminance" is "5 ... 100000 lx". The TEMD5510FX01 datasheet has a plot of illuminance (lx) vs photocurrent and it starts at 10 lx. Does this mean these parts won't work for my application? I'm not sure how to find out (without testing) if the dark current would be able to be distinguished from the low current at 0.1 lux.

donut
  • 165
  • 2
  • 15
  • I worked for OSRAM for a time as a contractor. The OSRAM device looks *lots* better in terms of its response over wavelength relative to human perception than the Vishay device. However, it is mostly useless for your range. So it doesn't matter. (You could get one and put a lot of work into characterizing it at the range you want. But it would be a LOT of work and even then likely not usable in the end.) – jonk Aug 29 '22 at 06:48
  • I sort of remember that photodiodes actually respond fairly quickly, but I have not worked with them extensively. Are you sure you can't just use an ordinary silicon photodiode with low dark current? Seems like the BH1680FVC was not optimized for speed. – user57037 Aug 29 '22 at 06:55
  • 1
    Agreed. I think the OP is probably looking at a nanoamps. (Depends on the optical acceptance angle and the ambient situation. But we are talking roughly about a threshold of 1/10th of the light one gets from the moon, at night. So very very small numbers even for a largish 1 cm ^ 2 photodiode.) – jonk Aug 29 '22 at 06:58
  • The fall time was very slow when I tested it and it took about 10ms to drop down to 0V. <-- what was the light source that you de-illuminated <-- how quick would it dim? – Andy aka Aug 29 '22 at 10:56
  • Your worries about distinguishing dark current from light current are valid. You might consider using an optical chopper, and possibly some cooling below room temperature. – glen_geek Aug 29 '22 at 12:41
  • The optical source that was de-illuminated was an LED pulse. – donut Aug 29 '22 at 14:53
  • After measuring; shorting sensor could be useful ... or perhaps with also a "negative" supply voltage ... – Antonio51 Aug 29 '22 at 16:57
  • How would shorting the sensor or a negative supply voltage be useful? – donut Aug 29 '22 at 18:31
  • 1
    A Digikey search of photodiodes focused on ambient wavelengths only yielded PIN type photodiodes. – donut Aug 29 '22 at 19:26
  • @donut Faster internal charges balancing? Not for BH1680FVC, but for other "simple" devices. – Antonio51 Sep 01 '22 at 11:34

0 Answers0