2
\$\begingroup\$

I am working on what I hope is a simple project as I am dabbling in this field. I am trying to create a target (in the shape of a box) roughly 1m by 0.5 meters into which soccer balls will be shot (diameter of between 10 and 25 cm). I want to be able to detect any ball which passes through the opening of the goal. I have looked at PIR sensors but am not sure if they will be able to detect a fast moving object (potentially up to about 3 cm per sec). Another concern I have is that it may be possible for a ball to remain in the sensor(s) field if it hits the edge of the goal and does not pass cleanly through and exit the sensor field. If this happens I would want to make sure that the sensor(s) could pick up the motion of another ball which passes through the goal on either side or above the stationary ball.

The idea is to connect the sensors so that each "goal" is detected and that any miss (default if no goal is scored within a set time frame) is also detected.

I have read some similar questions regarding moving bullets and even soccer goals but these issues were not addressed. Also, I would like this to be outdoor usable, meaning weather resistant. I have looked at some weather proof ultrasonic range finders, although with any sensor I am concerned with dirt and other materials from obstructing the sensor(s).

Any help would be greatly appreciated.

\$\endgroup\$
  • \$\begingroup\$ have you considered laser sensors? \$\endgroup\$ – triplebig Feb 9 '14 at 22:42
  • 3
    \$\begingroup\$ 3cm/second? If this is for outdoor practice (say with kids for example) and using real soccer balls and real shots, it could easily be more like 700 to 1000 cm per second (maybe 20mph). \$\endgroup\$ – mikeY Feb 9 '14 at 22:43
  • \$\begingroup\$ Triplebig: Yes, my math was way way off. I was actually trying to calculate based on 100 km per hour speed, although that is well over the likely speed. I believe that would translate to about 2,777 cm per second. \$\endgroup\$ – user36925 Feb 9 '14 at 23:32
  • \$\begingroup\$ Did you find a solution? \$\endgroup\$ – Crashalot Oct 26 '16 at 17:58
1
\$\begingroup\$

How about modulated IR LEDs and remote control IR receivers? Mount IR LEDs in black plastic tubes along one short side and the receivers on the other side, also in dark tubes. The LED tubes should be pointed straight down the detector tubes. Mount enough of them with even spacing so that any ball passing through will pass a shadow over at least one sensor, preferably at least 2 sensors. For a 25 cm ball in a 50 cm opening, 2 sets should be enough, spaced 12cm from each side. If the ball goes through the hole, it is guaranteed to break at least 1 beam.

Then, modulat the LEDs at the frequency the detector is looking for. I think it's going to be around 30 or 40 kHz. You can use a 555 timer with a decent transistor to drive the LEDs. Then all you need to do is AND or OR the outputs of the detectors together to get your 'score' signal, depending on the output logic polarity of the detectors.

The reason I suggest IR remote control detectors is becuase they have two very useful filters to get rid of interference. First, they are generally encapsulated in plastic that blocks everything but the IR wavelength of the LEDs, eliminating a lot of background light. Second, they have a narrow bandpass filter and peak detector that is only sensitive to a narrow band of modulation frequencies, so even if there is a lot of e.g. sunlight shining on the sensor, it should still be able to extract the signal from the emitters.

Edit: Here are all of the IR remote receiver sensors from Digikey. I highly recommend using a bunch of these instead of bare photodiodes or phototransistors because of their built-in bandpass filters:

http://www.digikey.com/product-search/en/sensors-transducers/optical-sensors-photo-detectors-remote-receiver/1967024

\$\endgroup\$
  • \$\begingroup\$ I am not really familiar with IR LEDs but it seems that most of the suggestions that I have received or researched suggest the use of some light grid rather than ultrasonic or PIR. When you refer to black plastic tubes are you suggesting they run down the side (height) of the goal with opposing tubes on the other side containing the receivers? Again, not being familiar with the IR LEDs, are they sensitive to alignment with the receivers? I need to make sure it is robust since balls will likely strike the frame of the goal frequently which could shift the alignment. Thanks for the great input! \$\endgroup\$ – user36925 Feb 9 '14 at 23:39
  • \$\begingroup\$ The idea would be to use the tubes to block out as much light as possible so the detector can only see light coming from the emitter on the far side. If alignment is an issue, then you can use larger diameter or shorter tubes. Also, make sure that the IR LEDs that you use emit a wavelength that is compatible with the receiver as you can get IR LEDs at several different wavelengths. \$\endgroup\$ – alex.forencich Feb 10 '14 at 0:04
  • \$\begingroup\$ Thanks, interesting stuff. I wonder whether the commercially available IR LED strips 850 nm 5050 LED (waterproof available) could be used for this application. They could be recessed or shielded from the front and back to prevent sunlight from affecting them. I do not see any IR led receiver strips but assume that a series of receivers of a compatible wavelength could be used on the other side. \$\endgroup\$ – user36925 Feb 10 '14 at 0:11
  • \$\begingroup\$ An LED strip is probably not a very good idea. You want to have one sensor per detector, and you really only want each detector receiving light from one emitter. Otherwise you could get blind spots where a passing ball does not block light completely for at least one detector. \$\endgroup\$ – alex.forencich Feb 10 '14 at 0:28
  • \$\begingroup\$ Thanks Alex. I am trying to source the components that you reference, the IR Led transmitter and receiver as well as what type of transistor I would need to use. Frankly I am way out of my league here but am trying to muddle through it. Any help on specific components would be appreciated. Also as do I need to be concerned with the possible speed of the balls passing through the curtain? How would I make the curtain sensitive enough to be able to detect a ball that may be travelling 100 km per hour? \$\endgroup\$ – user36925 Feb 10 '14 at 23:00
0
\$\begingroup\$

If the back of the goal box was open and allowed the ball to pass thru, it could be caught by a net and this means there is no doubt about if the ball has crossed the line. Because a net is present, there would be no chance of the ball having enough momentum to deflect out.

Once the ball is netted it could be detected by a sensitive load cell.

Maybe I'm missing something here?

\$\endgroup\$
  • \$\begingroup\$ The goal box will be open on both ends, in fact I would like to make the structure as narrow as possible; like shooting through a frame. Balls will be caught by a net, however this will be a training device so the net will capture multiple balls. Also some will be shot harder than others and will strike balls already in the net, maybe even causing them to bounce around etc. For this reason I did not think that having a sensor behind the frame of the goal would work. \$\endgroup\$ – user36925 Feb 9 '14 at 23:51

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.