Interesting IR sensor test results

[Mr. Freeman]

My team just got our board today (we were on break and unable to get into the school to retrieve it). I decided to take it home, get it working, and preform some unofficial tests. I searched before posting this and didn't find anything similar, sorry if someone has already posted this.

I found that the board will reliably receive and indicate signals from much over 30 feet in the presence of a lot of ambient light. I took two mag-lights and pointed them straight at the IR sensor on the board and the board still reliably received and indicated signals from the TV remote I was using. (30 feet was actually 30 paces, and I only stopped there because I ran out of room)

I also found that the board could receive signals from 360 degrees. With my hand covering the LEDs on the remote the board reliably received and indicated commands from 2 feet away.

Now, the really interesting thing I discovered is that it takes only 1 other signal to jam the board. I made sure the board could reliably receive and indicate the signal from 2 feet away, then pressed a button on another remote while also pressing a button on the remote the board was trained with. With all of the remotes I tested (5), it took only one of them at any one time to completely jam the receiver.

I suspect things will get interesting with 6 teams all sending signals to their robots at once. I guess you probably shouldn't program your robot to drive very fast during autonomous or else you might end up slamming into a wall before you can get your bot to acknowledge your turn command. I hope no one will abuse this and just jam the entire field for everyone with their remote.

[Kevin Sevcik]

While you're in the testing mood, would you try a fluorescent light on the IR sensor and see how much that might interfere? I suspect it might be worse and would probably better simulate the metal-halide lamps that some of the gyms and arenas are lit with.

[XXShadowXX]

fluorescent light testing i think is kind of pointless, fluorescent shine primarly in the ultra-violet frequencies (but have coating so they radiate in normal light) therefore i think their disruption would be minimum.

[MadEyeMechie]

I was helping Mr Freeman do the experiment and all the houselights are compact fluorescents which we turned on for the test. So, we already tried it. We are still pretty surprised that it still sees the signals when fired through your hand. Very very sensitive little receiver (enough though we know that some IR penetrates the body).

[ubermeister]

I personally would not put any reliance on the IR system at all. Unless many optics and precise targeting are used, it will be impossible to send a clean signal with 5 others being broadcasted. Plus, let's not forget about the IR lap detectors...

[XXShadowXX]

gyms and arenas can be lite with (depending on age) very harsh lights;

that are not fluorescent, if these have are made like standard bulbs, with filliment (excuse spelling) then they would throw alot of infared

or carbon arc lamps (these were phased out) but some places still use them, these are lite with a arc of electrons jumping between the carbon poles, this would be an infared nightmare (radates all the way across the spectrum from radio to light, but mainly concetrated in the infared to light wave lenght i think).

best test i think would be take a million candle power flash light, and 5 other remotes and have them all send interference, if you can get a single throught that then your cooking with gas!

[Gdeaver]

That's what we suspected. Well, There is always a custom fequency ultrasonic reciever and transmitter.

[SpartenSoldier]

When the remotes were jamming each other did the frequencies of the remote matter or did any frequency jam any other frequency?

[Tottanka]

I believe tht FIRST wants us to find a solution for the Jamming. It was highly encoureaged to make a tube that will get signals only from a certain location during the kick off, including Dave's cercastic remarks on that. It is n idea i personnaly support (its a challange for us!). Any i am also sure that some neat solutions will be found.


A sugestion:

Plan on usng the IR only when really cose to the Robocoaches, and have a pipe from the Ir direct to them. There's a lot of planning here and a lot of sensors usage.

[Mr. Freeman]

When the remotes were jamming each other did the frequencies of the remote matter or did any frequency jam any other frequency?

I didn't change the frequencies on the 5 remotes I picked up. One was for something made by JVC, one was for a receiver/tuner, one was for a different tuner, one for a DVD player, and I forgot what the other one was.
So I guess any frequency will jam any other frequency.

[Bongle]

I didn't change the frequencies on the 5 remotes I picked up. One was for something made by JVC, one was for a receiver/tuner, one was for a different tuner, one for a DVD player, and I forgot what the other one was.
So I guess any frequency will jam any other frequency.

This would make sense. IIRC from 2004 when we had those IR receivers in the KoP, it only outputted a 1 when there was IR, and a 0 when there wasn't. So it would see no distinction between different frequencies, and therefore the IR board wouldn't be able to either.

Are you sure that flesh is transparent to IR light? I'd think a more probable explanation is that it reflected off your hand to surrounding walls, and then got picked up by the sensor. You could test this with an IR night-vision video camera and see if the room gets lit up when you fire the tv remote.

[mneary]

Most IR remotes use one of three frequencies; 36, 38, and 40 KHz. But their light outputs are all the same IR; the sensor will see them all.

IR remotes and their receivers use different codes, not different frequencies. If you're sending one code while someone else is sending another, the chances of either one getting through without errors is very small.

Try this at home: Hold the button on one remote (VCR?) and then try to use another (TV?). Very seldom will it work. There will be up to six robocoaches operating signaling devices during the Hybrid period.

[abeD]

I know for one of my college classes we had to make an IR receiver transmitter for a robot we were using and we could control the frequency at which we broadcasted the signal and would just make a good enough filter so that the phototransistors would only see close to this frequency. If you are allowed to do something like this you would get no interference from anyone else (as long as there not on the same frequency but you can ask them and can change your's accordingly if your setup allows it)

I haven't read the specifics of what your allowed to use but this consists of a simple IR LED's and some phototransistors which I don't think are against the rules.

I am also not on a team anymore so I haven't seen the receiver board or any spec sheets for it.

[JHSmentor]

I know for one of my college classes we had to make an IR receiver transmitter for a robot we were using and we could control the frequency at which we broadcasted the signal and would just make a good enough filter so that the phototransistors would only see close to this frequency. If you are allowed to do something like this you would get no interference from anyone else (as long as there not on the same frequency but you can ask them and can change your's accordingly if your setup allows it)

I haven't read the specifics of what your allowed to use but this consists of a simple IR LED's and some phototransistors which I don't think are against the rules.

I am also not on a team anymore so I haven't seen the receiver board or any spec sheets for it.


Do you think this would something you could/would build around the exisiting IR board and receiver that where given to us? kind of like a filtering shield that only lets in the frequency that you will be transmitting?

[k4r3n2]

Did you press the same button on the remote that you were jamming it with (i.e. play and play)? We haven't started playing with our IR board, but I was curious as to whether the selection of the buttons would affect the effectiveness of the board, i.e. if we picked obscure buttons, our signal would work better. Thanks for posting your testing results! I find it very interesting that it works for 360 degrees, most receivers apparently aren't good for over a 100 degree span.

[abeD]

Do you think this would something you could/would build around the exisiting IR board and receiver that where given to us? kind of like a filtering shield that only lets in the frequency that you will be transmitting?


Yea that's what I was trying to say although i have no idea if that is possible with the given board. Someone else that's actually working with this thing will have to say.

[Lions for First]

i dont know all that much about IR but what kind of "filter" would you use that blocks all other frequencies but one?

[abeD]

You don't block all frequencies but one (if you could do that simply then you would be making alot of money) but with some combination of low pass and high pass filters you can get a nice bandpass of filters centered around where you are transmitting. Do a wikipedia search on filters and try to get through it. All of the stuff I've learned about is college level engineering so it might not be the easiest thing to understand.

There are also several types of active filters as well that have a sharper "edge" to them on their bode plot. (a magnitude vs. frequency plot).

[JHSmentor]

You don't block all frequencies but one (if you could do that simply then you would be making alot of money) but with some combination of low pass and high pass filters you can get a nice bandpass of filters centered around where you are transmitting. Do a wikipedia search on filters and try to get through it. All of the stuff I've learned about is college level engineering so it might not be the easiest thing to understand.

There are also several types of active filters as well that have a sharper "edge" to them on their bode plot. (a magnitude vs. frequency plot).

I like the idea of it but don't know how either. I read some patents and some college papers and it's a bit confusing (particularly because I could not view any of the images).