Shop OBEX P1 Docs P2 Docs Learn Events
proximity obstacle detection sensors problems — Parallax Forums

proximity obstacle detection sensors problems

shaharl2shaharl2 Posts: 3
edited 2014-09-18 09:47 in Accessories
Hi,
I am trying to built a presence detection device with a small microcontroller and sensors which is located on the ceiling and could alert on whether an object (non moving) is present within the radius of 1 meter in all directions. It should be very small and cheap (less than 5$) so any solutions like radar or laser scanning is out of the question, so I have left with the ultrasonic and infrared possibilities, but due to budget and size limitations i need to use as few sensors as possible.
I have tried to make some measurements with infrared transmitter and receiver in the same housing but so far I've been able to detect only objects in the range of about 20 cm, and pretty much only if it is perpendicular to the sensor. I have used TSOP1838 as a receiver and TSHF5210 as an emitter and supplied it with different current values but I could not significantly increase the range and the beam detection angle
1) how could I increase the range and angle? I check the IR detector with a normal TV remote control and i see that the range could reach up to 5 meters and the detection angle is very large. is the IR transmitter in the remote control is amplified? and if so, how? with optical lenses or with electronic components?
2) I understood that the carrier frequency of the receiver is 38kHz. Then how come when I operated the emitter with 38kHz there was no detection at all? actually the detection was the strongest at around 50Hz and when I applied it with steady DC there was again no detection. what should I calculate to find the optimal frequency?
3) how can I increase the detection angle and save the use of multiple sensors? I thought of using metal reflectors to bounce the beam in all directions. would that work?
4) I have not tested it yet with ultrasonic sensors because they are a bit more expensive but if the IR fails I will have to try it. could you recommend on a ultrasonic sensor which is cheap and small and can do the job?

Thank you very much in advance,
Shahar

Comments

  • Duane DegnDuane Degn Posts: 10,588
    edited 2014-09-18 06:34
    shaharl2 wrote: »
    could you recommend on a ultrasonic sensor which is cheap and small and can do the job?

    If you purchase several at once, you can get ultrasonic sensors for about $1. This thread lists several sources.

    When using IR to detect an object the color of the object is important. If the object reflects IR well it will be easier to detect.

    How large and what shape is the object you wish to detect? Which sensor will detect an object the best, depends on the object's shape and its surface qualities.
  • shaharl2shaharl2 Posts: 3
    edited 2014-09-18 07:01
    Thanks Duane for your answer. The purpose of this is to serve as an add-on to a smoke detector and alert when large objects as a wardrobe for example is located near the detector and could undesirably affect the detectors sensitivity. Small objects would not simnifically affect the detectors sensitivity and therefore, the objects to be detected are rather large, but should be detected at any direction under the detector at about 1 meter radius.
  • Tracy AllenTracy Allen Posts: 6,658
    edited 2014-09-18 09:02
    An IR receiver like the TSOP1138 will respond best when the the 38kHz comes in short bursts. The data sheet is quite specific about the timing. Bursts should be 6 to 70 cycles (158 to 1842 µs) spaced with gaps of at least 10 cycles (263 µs). It won't receive a continuous transmission. Maybe that explains your observation #2?
  • shaharl2shaharl2 Posts: 3
    edited 2014-09-18 09:26
    Thanks Tracy. It certainly explains it. I missed the part that the 38kHz should come in shorts bursts.
    what I don't get is how a normal tv remote control, which for my understanding does not use some high power or high brightness IR led emitter, but rather a normal 960nm one, is bale to transmit to a range of about 10 meters and with such high angle. why can't i get more than 40 cm detection? what am I missing here?
  • Tracy AllenTracy Allen Posts: 6,658
    edited 2014-09-18 09:47
    Well, do give it a try with bursts. That is how TV remotes work. The Propeller is quite good at generating and detecting signals. What microcontroller are you using?

    I have better luck with my TV remote when I point it at the painted wall opposite the TV than I do with pointing it right at the TV. That is 5 or 6 meters. I think you shouldn't have trouble with a 1 meter range.

    The LED you specified has a relatively narrow beam, 10°. You might want one a bit wider, or a diffuser.
Sign In or Register to comment.