Veiwing angle of Ultrasonic range sensor

oracacle

Senior Member
I know it not a viewing angle per se, but is it quite narrow, or wider.
I have seen some advertised as about 15 degrees (HC-SR04)

I am thinking about using it as a proximity sensor, I may need to monitor a wider angle, so need to figure out if I need more than 1 or servo mounting.
 

AllyCat

Senior Member
Hi,

As usual, we (or at least I) need more information:

What "proximity" range and accuracy do you want? What acceptance angle, and spread across one or two axes (horizontal angle or solid angle)? Is there a power restriction and does it need to run 24/7 (a servo or multiple sensors will need more power and might wear out)? Presumably you don't need to identify the actual direction of any "disturbance"?

It might be possible to "diffuse" the beam from a normal sensor, perhaps using a "diffraction grating" principle. But maybe difficult to make a "lens" work accurately at short range (there is the spacing between the TX/RX to consider) and conversely large distances may not work well if the transmitted energy is spread over a much larger area/volume.

"Large Area" ultrasonic (e.g. room or car interior) security detectors normally use(d) a dfifferent principle: Continuously transmitting and detecting for changes in the "standing wave" pattern. Probably (now) better done with a PIR.

Cheers, Alan.
 

westaust55

Moderator
As Allycat indicates, we need more details on which Ultra Sonic range finder you are looking at.

If it is the SRF-05 or SRF-08 as sold by Rev Ed through the PICAXEs tore then the respective datasheets are available here where the signal polar diagram for range and beam distribution is shown (near the bottom of the data):
http://www.robot-electronics.co.uk/htm/srf05tech.htm
http://www.robot-electronics.co.uk/htm/srf08tech.html
Note that these may not apply to SRF-05 clones made in China using different ultrasonic sensors.
If you have one of the others also produced by Robot Electronics then have a look at their full range:
http://www.robot-electronics.co.uk/products/sensors/ultrasonics.html

In terms of beam width, consider that many robot constructors mount either multiple range finders or mount one with a servomotor to swivel the US sensor through a wider range of angles.
 

oracacle

Senior Member
I'm just looking at way of detecting approaching things. I may still use a PIR, but using ultra sound mean that I could look for a significant and defined change so that a bush blowing in the wind doesn't trigger anything.

More just musing about future project ideas more than anything else at this point.
 

techElder

Well-known member
oracacle, I used to define an ultrasonic transducer beam with a flat, concave or convex (depending on strength of reflection) reflector and a synced oscilloscope.

I would watch the oscope RF reflection as I move the reflector into the edges of the beam. I watched for a set (usually specified) change in amplitude that I could call the "edge."

This method falls apart if you try it within what is called the "near field" of an ultrasonic transducer where the sound mixes quite significantly, so you might be able to define the near field also.
 

Pongo

Senior Member
Microwave/doppler radar is another technology to consider. The sensors are available for a few $, look for "HB100" on ebay for example.
 

rossko57

Senior Member
Note that "blowing in the wind" does not seem like a good mix with ultrasonics, with or without swaying bushes.

PIRs aren't very good at detecting approaches either, better at crossings. Swaying bushes need not be an issue though - remember its looking for temperature changes against the background.

Break-beam technology may be another reliable option.
 

eggdweather

Senior Member
I've used the HB100 units, but the Doppler shift output is very low and needs some signal amplification to use with a Picaxe. What I do is look for a sudden change in ADC input, range is about 5Metres max. I have found the beam with of the Ultrasound units (SR04) to be about 30-degs and much to my surprise someone has actually measured them see the polar plot here: http://www.google.co.uk/url?sa=i&rct=j&q=&esrc=s&source=images&cd=&ved=0ahUKEwiip77k0I_OAhULGBQKHasPBdcQjBwIBA&url=http://forum.arduino.cc/index.php?action=dlattach;topic=231775.0;attach=76469&psig=AFQjCNHUVhW5nR6A86sSRAlgsH0UQwF_6A&ust=1469571015353099&cad=rjt
 

Pongo

Senior Member
PIRs aren't very good at detecting approaches either, better at crossings. Swaying bushes need not be an issue though - remember its looking for temperature changes against the background.
Another downvote for PIR's. My outdoor lighting PIR sensor is pretty much useless below 5C when it's windy due to false triggering, bushes are the only potentially non-static things anywhere close to the field of view, and when it's close to 35C it has a hard time too, even detecting vehicles.

Break-beam technology may be another reliable option.
Probably the most reliable IMHO.
 

erco

Senior Member
Last edited:

oracacle

Senior Member
hmm Doppler radar, that's something that had not crossed my mind. this could in theory allow me to look for a speed instead of a movement.
Beam break sensors are in final stages of development now but could be quite trigger to get working of a larger area

As a side I found this while doing some reading of the radar, it maybe be useful for some, and gives me a starting point if I decide to have a play
http://www.limpkin.fr/public/HB100/HB100_Microwave_Sensor_Application_Note.pdf
 

inglewoodpete

Senior Member
Swaying bushes need not be an issue though - remember its looking for temperature changes against the background.
I have an out-of-the-box sensor light at the front of my house. While it only happens on windy nights in summer, false triggering does occur. Admittedly, this is in Australia when a hot wind is coming from inland and my front garden is busily transpiring to keep cool!
 

oracacle

Senior Member
there isn't any reason why false triggers need to be listened too if using picaxe to control the outcome.
initial activation wakes the system, if another activation occurs within a given time action is taken, else back to sleep it goes.
 

eggdweather

Senior Member
I agree, it is relatively easy to implement some form of low pass (to characterise it) filter in the software, so that it requires more than a single or series of detection threshold detections to then trigger, even a simple averaging of counts and trigger when it reaches a point would do that.

The Doppler modules are indeed useful for measuring speed and to a certain degree size of object, greater speed of approach = higher frequency and size of object (reflection) = greater amplitude. As in most things you have to compromise and decide if speed or detection of objects is the goal as the amplification required will determine what is achieved in my experience. Low level signals detecting speed need about x 5 to x10 amplification, whereas large objects would result in a chopped/square wave output as the signal was over amplified.

Premium house burglar alarm systems use a similar if not identical module to the HB100, they may be worth a look for the housing, lens and ability to withstand the elements, then add a picaxe signal processing to do with what you need.

Worth a look: https://www.rapidonline.com/pdf/50-0331.pdf
 
Top