RCWL-0516 Microwave Radar outdoors

Hi

I have a RCWL-0516 Microwave Radar Sensor connected to arduino so i can detect someone approaching. Inside my house it works ok, detecting me walking up to 5m, but when i go outdoors to open space, sometimes it only detects me when i'm some centimeters near.

I also made a simpler test, without arduino, powered by a 6V battery and a led on it's output, the same result, inside works ok, outside no.

Has anyone could get this sensor to work outside?

Thanks

Move it away from metal objects.

That particular sensor does not seem to be a “Doppler Radar” in the conventional sense, but instead appears to rely on interference between transmitted waves and waves reflected by surrounding objects (not just the object moving). The lack of such reflections might explain your observations.

Roger Clark describes his very thoughtful attempts at reverse engineering the sensor here Investigating a RCWL 9196 / RCWL-0516 “Radar” motion detector module – Roger Clark

You might try one of the other offerings, which may work on a different principle.

wvmarle:
Move it away from metal objects.

I believe there is more metal inside my house than outdoors, and it works ok inside...

jremington:
That particular sensor does not seem to be a "Doppler Radar" in the conventional sense, but instead appears to rely on interference between transmitted waves and waves reflected by surrounding objects (not just the object moving). The lack of such reflections might explain your observations.

Roger Clark describes his very thoughtful attempts at reverse engineering the sensor here Investigating a RCWL 9196 / RCWL-0516 “Radar” motion detector module – Roger Clark

You might try one of the other offerings, which may work on a different principle.

I read that article before, but no relevant information about outdoor usage.

Other modules like HB 100 are not omnidirectional and consumes a lot (30mA vs 3mA from RCWL-0516).
Do you know other modules? My goal is to detect movement (someone aproaching @ 5m) and low consumption to be battery powered.

geologic:
I believe there is more metal inside my house than outdoors, and it works ok inside...

It's not about the quantity, it is about metal objects close to the sensor. That includes your Arduino and wiring.

Ok, but the amount of metal things close to sensor is the same, inside or outside. I have everything inside a plastic box, i just put it inside or outside to do the tests, so there is nothing i can "move it away".

If the sensor worked ok outside but not inside, the metal interference could make sense, not the opposite.

By reading this PDF it seems to not be Doppler at all, but a single transistor oscillator that has circuitry on both sides of the circuit board and detects a large frequency shift when an object is in its vicinity.

And it detects from both sides of the circuit board.

Paul

Paul_KD7HB:
By reading this PDF it seems to not be Doppler at all, but a single transistor oscillator that has circuitry on both sides of the circuit board and detects a large frequency shift when an object is in its vicinity.

And it detects from both sides of the circuit board.

Paul

I never mentioned that is a doppler, and i know it is omnidirectional.
What i don't know is why it works very well inside my house and not outside, at open field

Why it does not work in the open is tentatively explained in reply #2.

appears to rely on interference between transmitted waves and waves reflected by surrounding objects (not just the object moving). The lack of such reflections might explain your observations.

geologic:
I never mentioned that is a doppler, and i know it is omnidirectional.
What i don't know is why it works very well inside my house and not outside, at open field

I know you didn't, but almost all the Google hits sell it as a Doppler radar detector.

You now have the device inside a plastic box, but I don't see you telling us it was inside the plastic box when you tested it inside your house. The plastic will certainly affect the frequency of the oscillator.

Paul

In my experience it only react to moving obejcts, not at all to stationery objects, and indeed it does so from quite a distance (I got about 4-5m). So it definitely needs CHANGES in reflections it receives.

So still a mystery how this thing even works. Makes it even more of a mystery on how it's possible to sell them this cheap :slight_smile:

@Paul - the plastic box and what's inside is the same indoors and outdoors, so the frequency disturbance should be the same. And i already tested without the box, same problems.

@jremington - after lots of different tests, i must admit that the problem seems to be what you described, lack of reflection in the open space.

@wvmarle - yes, too good to be true. It's small, it's cheap, does not consume a lot, but it is good only for indoors.

Thank you all for the help, if someone knows a low power method to detect omnidirectional movement around 7m, let me know.

PIR detectors are low power and work very well. Have you excluded them for some reason? It would probably take 3 or 4 to cover 360 degrees.

jremington:
PIR detectors are low power and work very well. Have you excluded them for some reason? It would probably take 3 or 4 to cover 360 degrees.

I tried PIRs, digital version from Panasonic, they worked great inside, but when going outdoors i had different problems:

  • false detections: i think because of tree / leaves movement or temperature / sun reflection;
  • range / blind spots: depending the angle of aproaching, sometimes it just detect me 1m or 2m from the sensor, so i will need more sensors to cover 360º, not a very elegant solution

These kind of sensors are, I believe, reliant on reflected energy coming back from the room, and people
act to absorb such reflections and change the level/phase of the reflected energy. Outdoors you'd have
to wear metal objects/clothing to act as a reflector rather than absorber, to cause changes that go back
to the module. They are just a microwave oscillator with a very sensitive circuit amplifying the changes
in oscillator power consumption with a bandwidth of a few Hz. Reflected RF impinging on the module will
cause minute changes in antenna impedance that act to affect the oscillator to perhaps a few parts per million
or so - amplify that enormously at a narrow range of frequencies of interest and you get great sensitivity.

Note that radars detect metal, not people (who are usually very efficient absorbers of microwave energy).

Hence the need for people to move in front of metal to be easily sensed with any kind of radar technology.

Passive IR suitably tuned and aimed is much more likely to be effective (ie aimed down from above, tuned
to ignore passing clouds).

That's a nice explanation, and it makes sense.