IR LED Light Curtain

Discussion in 'The Projects Forum' started by WolfmanFL, Feb 20, 2014.

  1. WolfmanFL

    Thread Starter New Member

    Feb 19, 2014
    9
    0
    Hi, I am working on a side project with my son who is an avid soccer player. One component of the project is to create a soccer goal approximately 1 meter wide and 0.5 meters high that can be used indoors or outdoors (in full sun light) and which will have the ability to sense a ball being shot into it, even if shot at a high speed (up to 75 km) with a ball being between 20 and 25 cm in diameter.

    Based on the research I have done on several forums and elsewhere it seems that a series of IR LED emitters and receptors located on opposite side walls of the goal (which will in fact be a plywood box, open in front and rear) spaced about 10 cm apart (meaning between 4-5 pairs) would provide me with the "light curtain" that I need to capture any ball that traverses the box.

    I understand that in order to achieve the response time that I need and to avoid false triggers through sun light etc. I need to consider the pulse frequency and sequencing of the emitters. I am not sure if I can achieve the response time by sequencing and wonder if i can use a PASM object to manage more than one emitter at a time. Is there any reason I cannot run all emitters the whole time and avoid sequencing? I plan on having the emitters and receptors recessed in the box to protect them from sunlight etc. and using some optics to narrow and focus the beams.

    Now, I have no experience in electronics whatsoever and while I understand the basic components I have no idea how to put this all together. I have been looking for someone who is in my area (Tampa, FL) to help me with this project as it will require some field testing, so if there is anyone you know in my area that may be interested in working on something like this (for fair compensation) that would be appreciated as well.

    Thanks, in advance, to all.
     
  2. inwo

    Well-Known Member

    Nov 7, 2013
    2,433
    315
  3. R!f@@

    AAC Fanatic!

    Apr 2, 2009
    8,750
    759
    U can run all receivers at ones. But need a whole lotta circuits some what.
    A μC is fast enough to sequence and detect IR interruption though.

    You can avoid sunlight by apply Filters at the receiver side and use IR emitters in series too.
     
  4. WolfmanFL

    Thread Starter New Member

    Feb 19, 2014
    9
    0
    Invo: I have looked at the commercially available curtains but see a couple of problems; the first is their size; they tend to be pretty long and would have to be located on the top and bottom of the goal and which is problematic since the bottom bar would be in the way of the goal and also likely to become occluded by dirt, grass etc. Also I need this to be able to run off batteries and from what I can tell the industrial curtains would be hard to run off batteries.
     
  5. wayneh

    Expert

    Sep 9, 2010
    12,100
    3,034
    How do you differentiate the ball from other objects, such as the goalie or a butterfly, that might break the plane of the goal?
     
  6. R!f@@

    AAC Fanatic!

    Apr 2, 2009
    8,750
    759
    Good point.
     
  7. MaxHeadRoom

    Expert

    Jul 18, 2013
    10,515
    2,369
    At 1m x .5m I wouldn't think the goalie is standing between the posts, unless they are very young?.:)

    There could be some other technology where some non-powered item is placed on the ball itself, and sensed?
    Max.
     
    Last edited: Feb 20, 2014
  8. WolfmanFL

    Thread Starter New Member

    Feb 19, 2014
    9
    0
    MaxHeadRoom is correct, there would be little "head room" for a goalie. :) No goalie, just a training device to test accuracy of pass/shot. I do not want to use any special balls (RFID) reflective etc. as this would complicate use. Balls would likely be standard size 4 (youth) or size 5 (adult) balls but like to keep the spacing narrow enough to allow for mini balls that are good for training in small spaces. I am not too worried about the errant butterfly or bumblebee, although if mosquitoes (or the like) trigger the device then I may need to space the beams closer and require that at least 2 beams are interrupted to trip the device to reduce those false reads.
     
  9. MaxHeadRoom

    Expert

    Jul 18, 2013
    10,515
    2,369
    I would have though that they could be either side rather than top & bottom?
    Most IR safety curtains 'sweep' through a fast light sequence, if one were designed on this principle, a certain size window could be programmed in to avoid small objects being detected, but it certainly would be a case for a small μprocessor.
    Possibly an off the shelf motion detector could be used?
    Max.
     
  10. inwo

    Well-Known Member

    Nov 7, 2013
    2,433
    315
    I've used 4" curtains. Size shouldn't be an issue.

    Most of the curtains I've used are 24vdc. Although they are quite power hungry powering all the leds, so would and led source competing with day light.

    Closely spaced beams and only sensing things that break several adjacent beams may prevent some false trips.

    The optics are always the hard part for me when doing optic projects.
     
  11. MaxHeadRoom

    Expert

    Jul 18, 2013
    10,515
    2,369
    On the subject of motion detectors, I have never experimented with the commercial ones, but if one were placed on one side the sense angle could possibly be narrowed by a suitable small shield or slot.
    The power demand of the unit itself should not be that great.
    An advantage also if it were to have a sensitivity adjustment this would further make it useable.
    Max.
     
  12. WolfmanFL

    Thread Starter New Member

    Feb 19, 2014
    9
    0
    I have thought about a commercial PIR sensor but am concerned that it would be hard to deal with false triggers from balls striking the outside or frame of the goal, which is far more likely than a butterfly entering the goal box. Keep in mind the goal is only 1m wide and 0.5 meters tall.
     
  13. WolfmanFL

    Thread Starter New Member

    Feb 19, 2014
    9
    0
    But even with a 4' long sensor beam it is wider than the width of the goal and, again, I do not want to locate the beams on the top and bottom as they would likely be covered in debris. I would need to find a sensor bar that is about 18" long so that it can fit along the inside of the goal vertically. Also the response rate that I have seen on most industrial units is too slow to capture a ball moving at 75+km.
     
  14. tracecom

    AAC Fanatic!

    Apr 16, 2010
    3,869
    1,393
  15. R!f@@

    AAC Fanatic!

    Apr 2, 2009
    8,750
    759
    I think you would need two bars in parallel or something.
     
  16. MaxHeadRoom

    Expert

    Jul 18, 2013
    10,515
    2,369
    75Km/Hr ?
    That would equal 1250mm/sec.
    Considering the ball dia, of say 10" (25cm) if the beam width were half of this, the beam would be covered for ~200ms. which is fairly long for electronic detection purposes.
    Max.
     
    Last edited: Feb 20, 2014
  17. inwo

    Well-Known Member

    Nov 7, 2013
    2,433
    315
    4 inch. I was using for minimum example. The link was a 20" one.

    So finding the right size shouldn't be an issue.

    On another thought. An industrial ultrasonic sensor can be programed to respond between two set limits. Say 1 meter and 2 meter.

    It would not be accurate on the other axis, but two at 90°, may work. If they can be programmed not to interact. I'd have to look it up.

    Or one may work if it was placed, such that the sensitivity area was the right distance.

    Another mode is "loss of echo", which is like a thru beam sensor.
     
  18. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,523
    1,247
    Emitters: There is nothing magical about infrared emitters, and they are a pain to aim. If you want to use narrow beam emitters, use visible ones. They look cool, and you can aim them by just looking. They can all be in series to reduce power consumption and drive complexity. If you toggle them fast, something like 50 KHz or 100 KHz, the detector becomes easier.

    Detectors: Rule #1 - there is no such thing as a perfect detector; no matter what you use, you might get false positives. If this is outside and a small dog runs through it, I can't think of anything for a reasonable cost that will differentiate that from a ball. But a little care can reduce the odds of butterflies setting it off. A 100 KHz square wave is easy to detect and hard to fool. For about $10 in parts you can sum the total energy from 8 detectors (remember, the outputs are in phase from the series-driven emitters), bandwidth filter it to clean it up, and synchronous detect it against a threshold value to see the obstruction.

    ak
     
  19. WolfmanFL

    Thread Starter New Member

    Feb 19, 2014
    9
    0
    Sorry, my bad. My main reason for not wanting to use the industrial safety curtains is cost. The unit you sent a link to is reasonable but if I want to build a few of these; I currently intend to build at least 2, I need to be able to source the same components and the new ones are very expensive. I find the idea of ultrasonic sensors interesting since waterproof and reliable units are available but I had not thought about trying to have 2 work together.

    Another thought that I had, keeping in mind that i am trying to keep this as simple as possible and reliable under the likely usage, is to have some sort of pressure plate sensors perhaps hanging from the inside top of the goal box in a series of strips (sort of like the clingy things in a automated car wash) that would be flexible enough to allow a ball to pass through them and through the goal box. I have not really found any pressure pad "strips" that would work as the ones i have found have a focused sensing area which would likely not work.
     
  20. WolfmanFL

    Thread Starter New Member

    Feb 19, 2014
    9
    0
    I do like the idea of lasers in lieu of IR but am really not familiar with the advantages and disadvantages of each. Longevity (including power consumption as this will be battery operated) and reliability would be on the top of my considerations for either. I would want to stay away from mirrors due to the likelihood of the mirrors of shift due to impacts from balls or getting dirty or fogged up etc...

    One thought for reliability would be to have 3 lasers but have one running through the middle horizontally and the 2 others running diagonally from corner to corner, each laser staggered from front to rear. This would allow the unit to maintain a high degree of practical functionality even if one laser or receptor goes down as the resulting blind spot would be relatively small.
     
Loading...