6

I will preface this by saying that I am a hobbyist not an engineer, which is probably why I even have to ask this question.

Like the title says I want to know what type of sensor I can use to get the height of an object from the snowpack covering the ground, not the ground itself.

It will be attached to an object in motion.

Some sensors I have looked up state that they go through snow in their descriptions, but I am unsure if they simply mean falling snow in the air or if the meant packed snow on the ground.

So far the options seem to be laser, radar, infrared, ultrasonic, and possibly optical.

An accuracy of inches would be preferable but feet would be ok. This rules out barometric pressure.

Any help would be greatly appreciated.

Anindo Ghosh
  • 50,188
  • 8
  • 103
  • 200
cmgriffing
  • 125
  • 1
  • 7
  • 1
    What is the range of heights you expect to measure? Is the device going to be suspended pointing downward from your object in motion? – Anindo Ghosh Nov 01 '12 at 05:18
  • I was hoping the range would be 0 to 50ft or more. However, I have read that some sensors have a minimum proximity above zero. Yes the sensor would be attached to the object pointing down. – cmgriffing Nov 01 '12 at 05:47
  • You can wait a few more years and not worry about snow anymore :v – Bryan Boettcher Nov 01 '12 at 21:41

2 Answers2

8

Some of the options mentioned in your question would work for the purpose, with varying precision and suitability:

  • Ultrasonic: Probably your best bet: Narrow aperture piezoelectric ultrasonic sensors (single-unit send/receive) are reliable in ranges from a few inches to a hundred feet or more. Cost goes up disproportionately with range, though. Precision is fair-to-high, and it works with most surfaces, including reflective and transparent ones i.e. packed snow or sheet ice. Scan frequency is fairly high.
    • Senix TSPC-21S series has a range precisely matching your requirement, 50 feet range / fractional inch interpolated precision, but costs around $500-800. Some Chinese clones do exist, at around a tenth the price or less, but no idea on precision or reliability. We had ordered one, it arrived badly damaged, seller was unresponsive. Maybe give it a try if you don't need it waterproof.

Senix narrow-aperture ultrasonic distance sensor

  • Laser: Pulsed time-of-flight principle, similar to ultrasonic sensors. Scanner type, since reflector type will not work with snow. Range up to 40 feet, precision extremely high.
    • SensoPart FT 90 ILA might suit the requirement if 30 feet range is acceptable. It claims 2 mm precision, which is meaningless for uneven snow surfaces.

SensoPart scanning laser distance sensor

The following would not be recommended:

  • Optical: These too use lasers, but operate on a stereoscopic triangulation principle. Range is limited, typically to 1 meter. Precision is very high, scan frequency is high.
  • Radar: Insufficient precision and insufficient minimum range. Integrated RADAR devices are usually for 10 feet minimum to several miles maximum range, weight is massive as range increases, and precision is in tens of feet.
  • Infrared: Insufficient maximum range for your purposes. Conventional IR reflectometry works up to a couple of feet at the most, and is more popular in sub-foot ranges. However, the Laser device listed above is also infrared, in that it uses an infrared scanning laser.

I hope this helps

Anindo Ghosh
  • 50,188
  • 8
  • 103
  • 200
  • Thank you, after some reading I am under the impression that Laser doesn't work for water(assuming snow too) because the laser doesn't properly reflect back. Please correct me if this is incorrect. – cmgriffing Nov 01 '12 at 08:52
  • Laser ranging fails with water because the transparency of the water causes the scanning beam to pass through the water surface rather than be reflected sufficiently for detection. This is typically not true for snow, and on the fence for sheet ice, though the wavelength of the laser would be a factor there. The only ways to confirm this hypothesis are to ask the manufacturer to test it, or by actually experimenting with the specific laser you select. – Anindo Ghosh Nov 01 '12 at 09:14
  • With any height sensor, one needs a refector to focus the beam of any source so the refection comes from a small area. Imagine the refections from all the distances above ground with a wide angle and then you will have to use a parabolic reflector with reflection absorbing material to avoid false readings beyond a certain beamwidth. This would apply to any source be it ultrasonic, RF or light. – Tony Stewart EE75 Nov 01 '12 at 17:00
1

Perhaps the most reliable is a human readible still image sensor (eg a webcam) of the ground with IR light or daylight and a yardstick, given that there were no other requirements for power or interface, but that may be too obvious.

Reflection coefficient of the snow, "whether" it is airborne, wet, dry, loose or packed will vary significantly with wavelength from audio to VHF to IR. Perhaps you could elaborate on;

  • the conditions when false readings are acceptible
  • how frequent this must be monitored or response time
  • powerand interface requirements
  • distance from sensor to user interface
  • preferred user interface
  • number of sensors
  • proximity and type of interference or moving objects nearby or typical setting
  • budget for project

Long ago when I was testing VHF antenna designs with a simple directional coupler and "hot carrier diode" detector and DVM to measuring relative motion or position of reflection to people or wall reflections using the reflected (diode detected) voltage of standing waves of the antenna signal from a directional coupler ( or cable splitter). They use these now for alarm motion sensors more than anything else, but it might detect a change in height above ground to the strongest layer of moist surface for reflecting RF. I could easily detect slight motion of someone standing 10 feet away but would require restrictions & calibration.

This is hypothetical and I have not tried it on snow, but know that moisture reflects RF but probably not accurate enough..

Tony Stewart EE75
  • 1
  • 3
  • 54
  • 182