Reply to thread

A pyranometer is an instrument used to measure solar radiation flux density, specifically the total amount of solar energy received per unit area on a surface. If using a thermal pile measurement, it compares the electrical voltage of a black absorptive surface to one of Bragg gradient selective wavelength. It uses a clear view of the sky with a large area wind-shielded Si-cell.

Since the sky view spectrum tilts up towards the blue spectrum near 450 nm and the SI-cell tilts up towards IR at 1000 nm or 1um peak, the product of the flux and sensor is a relatively constant flat response.  We know from scattering parameters in RF like in optical wavelengths the spectral transmission loss will result in a reflection and comparing clear sky with 9.6% overcast would look like fog if you flying thru it.

Given that the size distribution of rain to fog is huge which affects the spectrum of reflection, I decided to use two LEDs of opposite ends of the visible spectrum, UV and IR to be able to detect fog when there is no rain and visa versa.  Some field tests for emitters with narrow beam lens and Si, GaN, AlGaAs or GaInAs detectors with narrow beam lenses could capture the best reflection of  an unobstructed view for 10 m or so to capture the reflected light. The 5mm LEDs with narrow 10 deg lens make excellent photo detectors as long as spectrum is shorter than the emitter wavelength. This would be shaped like the Si spectra but shifted to the left of the emission spectra meaning it only responds to higher energy photo wavelengths up to a certain limit.

  Given these are point source and detectors the RF Friis Loss rules for path loss support that the unobstructed view required to detect fog is about the range of your typical IR remote control unit.  Given the great sensitivity improvement in SNR of using a carrier modulation like TV remotes, that would eliminate false triggers from PWM-LED headlights in the same optical spectra.  Typically the turn on/off times for 5mm LEDs are in the 1 to few microsecond durations, 100 kHz modulation may be a reasonable limit far away from AUDI's ( the 1st OEM) or Tesla's or Cadillac's or whatever's LED headlight harmonics. Unless you wanted to trigger/calibrate your FOG sensor with a TV remote control on a rarely used carrier f, you may want to avoid those channels.

This is just the general design thinking I have used in the past to create new instruments and might be a good start for "fog detector".  Is there a demand for this?  How many do you need?  (0?)

Afterward's now I asked Grok to comment on [USER=594824]@prairiedog[/USER]'s interesting chart.

-------------------

1. Spectral Density Comparison: Clear Sky vs. Cloudy Sky

• Clear Sky (Blue Line): The spectral density under a clear sky shows a broad distribution of solar radiation across the wavelength range, peaking around 500–600 nm (visible light) and gradually decreasing toward the infrared region (beyond 1000 nm). This is expected because, under clear skies, direct sunlight dominates, delivering a full spectrum of solar radiation, including ultraviolet (UV), visible, and near-infrared (IR) wavelengths.
• Cloudy Sky (Red Line): The spectral density under a cloudy sky is significantly different. It’s much lower overall, indicating that clouds reduce the total amount of solar radiation reaching the surface. The reduction is not uniform across wavelengths:
  • In the UV and visible range (300–700 nm), the cloudy sky spectrum still shows some peaks, but they are much smaller than the clear sky peaks. This suggests that clouds scatter shorter wavelengths (like blue light) more effectively, which is why the sky often appears gray or white under overcast conditions—scattered light still makes it through, but at a reduced intensity.
  • In the near-infrared range (700–2500 nm), the cloudy sky spectrum drops dramatically, almost to zero beyond 1200 nm. Clouds are highly effective at absorbing or reflecting infrared radiation, so very little IR reaches the surface under overcast conditions.
• 2. Implications for Silicon-Cell Pyranometer Measurements
  • The black curve represents the spectral response of the silicon-cell pyranometer, which shows how sensitive the device is to different wavelengths. Silicon-cell pyranometers are most responsive between 300 nm and 1100 nm, with a peak around 900–1000 nm, but their sensitivity drops sharply beyond 1100 nm.
  • Under a clear sky, the pyranometer captures a significant portion of the solar spectrum because its sensitivity range (300–1100 nm) aligns well with the broad distribution of solar radiation, including visible and near-infrared light.
  • Under a cloudy sky, the solar spectrum shifts heavily toward shorter wavelengths (UV and visible), with very little infrared radiation. Since the silicon-cell pyranometer is less sensitive to UV and has a peak sensitivity in the near-infrared (where cloudy skies provide almost no radiation), it underestimates the total solar irradiance in overcast conditions. This mismatch between the pyranometer’s spectral response and the cloudy sky spectrum leads to the 9.6% error mentioned in the graph.
• 3. What This Tells Us About Cloudy Sky Spectral Density
  • Reduced Overall Irradiance: Clouds significantly reduce the total solar radiation reaching the surface, as seen by the much lower amplitude of the red (cloudy) curve compared to the blue (clear) curve.
  • Wavelength-Dependent Attenuation: Clouds don’t block all wavelengths equally. They allow more UV and visible light to pass through (via scattering) while heavily attenuating infrared radiation. This results in a spectrum that’s skewed toward shorter wavelengths under cloudy conditions.
  • Impact on Measurement Accuracy: For instruments like silicon-cell pyranometers, which are optimized for a broader spectrum (including near-infrared), the cloudy sky spectrum poses a challenge. The device overweights the infrared region (where there’s little radiation under clouds) and underweights the UV/visible region (where most of the cloudy sky radiation is concentrated), leading to measurement errors.
• 4. Broader Context
  • This graph highlights a key limitation of silicon-cell pyranometers: their spectral response doesn’t perfectly match the solar spectrum under all conditions. In contrast, thermopile pyranometers (which I mentioned in my earlier response) have a flatter spectral response across 300–2800 nm and are less prone to errors under cloudy skies because they can better capture the full range of radiation, even when the spectrum shifts.
  • The 9.6% error for overcast conditions is significant in applications like solar energy, where accurate measurements are crucial for estimating power output. For example, if a solar panel system is being monitored with a silicon-cell pyranometer on a cloudy day, the device might underestimate the available solar energy, leading to incorrect performance assessments.
• Summary
  The cloudy sky spectral density is much lower than that of a clear sky, with a pronounced reduction in infrared radiation (beyond 700 nm) and a relative emphasis on UV and visible light (300–700 nm). This shift causes silicon-cell pyranometers to underestimate solar irradiance under overcast conditions, resulting in a 9.6% error. The graph underscores the importance of choosing the right pyranometer for the expected sky conditions—silicon-cell models are less accurate for cloudy skies, while thermopile pyranometers would provide more reliable measurements across varying weather conditions.