Detect minute ambient light changes?

Thread Starter

Ryan1327

Joined Nov 25, 2017
2
What would be the best sensor for a project circuit to detect tiny changes of the ambient light level in the night sky? (Sensitive enough to maybe... detect the nav lights of an airplane passing high overhead) and could take a reading twice a second? Photodiode? Thank you.
 

joeyd999

Joined Jun 6, 2011
5,234
What would be the best sensor for a project circuit to detect tiny changes of the ambient light level in the night sky? (Sensitive enough to maybe... detect the nav lights of an airplane passing high overhead) and could take a reading twice a second? Photodiode? Thank you.
You can use a sensitive photodiode (or a photomultiplier with sufficient ambient darkness).

In the case of a photodiode, I'd integrate the current over time to increase sensitivity and reduce noise.
 

philba

Joined Aug 17, 2017
959
Noise at those low levels is going to be a killer. Noise in this case is the random rate of arrival of photons. In photography, it's call shot noise. I suspect something like a phototube or photomultiplier is needed, depending on sensitivity.
 

joeyd999

Joined Jun 6, 2011
5,234
Noise at those low levels is going to be a killer. Noise in this case is the random rate of arrival of photons. In photography, it's call shot noise. I suspect something like a phototube or photomultiplier is needed, depending on sensitivity.
Yes. But, on average, there are more photons arriving when nav lights are passing by. This is why I suggested integrating the photodiode current.

Can it be done twice per second? No clue -- never tried.
 

Thread Starter

Ryan1327

Joined Nov 25, 2017
2
Thanks all. Phototubes look out of my league, and price range. Also, i dont need an accurate reading of light quanity, just yes/no to a change.
 

crutschow

Joined Mar 14, 2008
34,281
Hmmm. Our eyes can do it fine.
Certainly it can be readily done with an imaging system since it can ignore the bright moon and look for the small light from the plane.
I worked on a CCD system for plane navigation that could detect stars in full sunlight to determine the plane location from the star patterns
(obviously that system had an impressive dynamic range).

But I think using a single detector to detect the difference in total light from the sky between the full moon, and the full moon plus the lights from a high flying plane will be a real challenge (disguised as an impossible task :rolleyes:).
 

philba

Joined Aug 17, 2017
959
Yes, but are the photons from the nav light going to stand out from the noise floor when the whole sky is covered? Part of the reason humans can see the plane going by is we can point the fovea in the right direction. Maybe if you could have multiple detectors that each only cover a small part of the sky.
 

joeyd999

Joined Jun 6, 2011
5,234
Certainly it can be readily done with an imaging system since it can ignore the bright moon and look for the small light from the plane.
I worked on a CCD system for plane navigation that could detect stars in full sunlight to determine the plane location from the star patterns
(obviously that system had an impressive dynamic range).

But I think using a single detector to detect the difference in total light from the sky between the full moon, and the full moon plus the lights from a high flying plane will be a real challenge (disguised as an impossible task :rolleyes:).
I liked your post because you may be right (aside from the "impossible" comment). In my younger days, I would have considered this a challenge. I rarely lost those.
 

DNA Robotics

Joined Jun 13, 2014
647
Certainly it can be readily done with an imaging system since it can ignore the bright moon and look for the small light from the plane.
That is the way I would go.
For Microsoft Visual Studio there is sample code "DibLook". It lets you display video and gives you access to screen memory. You could scan the pixels for red and green.

DIBLOOK: Illustrates the Use of DIBs and Color Palettes
The DIBLOOK sample illustrates the use of device-independent bitmaps (DIBs) and the closely related use of color palettes.
 

nsaspook

Joined Aug 27, 2009
13,081
Certainly it can be readily done with an imaging system since it can ignore the bright moon and look for the small light from the plane.
I worked on a CCD system for plane navigation that could detect stars in full sunlight to determine the plane location from the star patterns
(obviously that system had an impressive dynamic range).

But I think using a single detector to detect the difference in total light from the sky between the full moon, and the full moon plus the lights from a high flying plane will be a real challenge (disguised as an impossible task :rolleyes:).
Not impossible, just tricky.
https://forum.allaboutcircuits.com/threads/super-moon-shine.100322/page-2#post-904573
 

nsaspook

Joined Aug 27, 2009
13,081
But would that circuit pick up planes "high overhead" as the TS wants?
Very possible under the right conditions. It all depends on the possible S/N ratio and the collected light signature of a "high overhead" plane of the solar panel. I didn't design the system for plane detection (I changed the DSP system to reduce the spiking) or max sensitivity. (the adc 128 V/V PGA was bypassed with unity gain)

Slightly Raw and processed signals:
https://forum.allaboutcircuits.com/threads/50nv-resolution.130468/#post-1073894

From the driver:
static const uint8_t ads1220_r0 = ADS1220_MUX_0_1 | ADS1220_GAIN_1 | ADS1220_PGA_BYPASS;

* comedi/drivers/supermoon.c
*
* A special version for the TI ADS1220 SD ADC converter chip (and MCP3911 later) for low voltage sensing and
* solar panel panel light detection. +- 2.048, 1.024 and 0.512 voltage ranges @ 20 bits of usable resolution
* ADC is in single-shot conversion mode @20SPS, PGA disabled and gain from 1, 2 and 4 in differential
* signal detection mode, 50/60Hz rejection enabled. 500kHz SPI clock with direct RPi2 connection
* Analog +- 2.5VDC from Zener regulators for the bipolar input stage with external 2.5VDC Zener input
* signal protection.
 
Last edited:
Top