FreeTrack Forum
FreeTrack Forum > FreeTrack : English Forum > Support : Tracking System > IR LED Alternate
freelancer27 | #31 08/01/2008 - 23h38 |
Class : Apprenti Off line |
Hey all I'm looking to put my freetrack hardware together this weekend and have a couple questions.
I have 3 IR leds (The serial number that was recomended.) and will be using a 9V battery. When I do the calculations I get a result that says to use a 75 ohm resistor. I noticed the above posts that are using 100 ohm resistors and wondering if something is wrong with the data I entered. Per the LED spec sheet I have: Forward current IF 100mA Forward voltage VF IF = 100 mA, tp = 20 ms 1.5 (< 1.8) IF = 1 A, tp = 100 μs 3.0 (< 3.8) I assume the voltage I enter should be the 1.5VF not the 3V/1A. Any reason you guys would come up with 100ohm (r) while I get 75ohm (r)? Thanks |
Deimos | #32 09/01/2008 - 00h25 |
Class : Beta Tester Off line |
You're right about using forward voltage 1.5V, but the forward current If=100mA is a absolute maximum rating, meaning that for using it won't damage the LED, but for normal use values lower than that should be used.
On the practical side, the higher the current, the shorter the battery life. So you need to find some compromise. Personally, i wouldn't recommend using the 100mA unless your camera has a strong IR filter, and you don't want to remove it, but in that case it would be better to use some regulated power supply, or at least a rechargeable accumulator not a battery. If you have the IR filter removed, you can safely use lower forward current values - i would recommend something between 30 and 60mA, depending on your camera and filter. Using 100R resistor would give 60mA forward current for LEDs, making them much brighter, but it's still rather far from the absolute maximum rating, so the LEDs won't drain the battery and heat up as much. The other forward voltage - @1A is just for reference - to show how much forward voltage changes with current. Trying to drive the LED with 1A for longer than a fraction of second would literally fry it. By the way, what are you using for your visible light filter, Radwolf? If it's a magnetic disk, then you might want to try using two layers of exposed photographic film instead. The magnetic disk (at least the one i tried) seemed to filter out some IR light along with the visible light, so IR LEDs seen trough it seemed much darker than seen trough the exposed film. Maybe this way you can stay with 200R resistors, and longer battery life |
freelancer27 | #33 09/01/2008 - 01h33 |
Class : Apprenti Off line |
Thanks for the reply. As far as the visible light filter... I haven't gotten that far yet but I was planning on using the suggested Photo negetive as I have some lying around the house.
So If I understand you correctly the 100ma listed on the data sheet will drain battery power very quickly, but if I remove the IR filter from my web cam (Microsoft VX-3000) then I can use a 100ohm resistor which will reduce the current to 60ma. A question here, if I'm using IR diodes and the camera has an IR filter, then wouldn't I have to take off the filter in any case? Where can I learn how to take off this filter? Thanks again for your help. |
Redwulf | #34 09/01/2008 - 02h53 |
Class : Apprenti Off line |
Hey Deimos,
I'm using the photo negative paper. Egore, you were partially right on the obstruction. Can't believe I didn't think of it! It doesn't solve the whole problem. I think I still need to boost the output. I've moved the lower LEDs out and they are able to track now. The only thing I'm running into now is some high jitter. It's not constant it comes and goes. I have to boost the gain and exposure a bit too much which seems to be the cause. If I back the exposure and gain up the jitter is real low but it won't track real well on the extreme angles. I think if I brighten the LEDs now I will be able to get the settings to a good spot with no jitter and I should be set. Thanks, sean |
Deimos | #35 09/01/2008 - 18h52 |
Class : Beta Tester Off line |
Well, the IR filter in webcams isn't perfect, so it cuts off only part of infrared spectrum - a little part of infrared light can always get trough. That's why you can see IR LEDs like ones used in TV remotes in freetrack as faint blue-is lights. The visible light filter (like the photo film) on the other hand cuts out virtually all visible light. So when you have a camera with the IR filter and additionally apply a visible light filter to it, it'll see a little of the IR light, and no visible light at all. So if the IR light source is strong enough (like a led driven by 100mA), it will show up clearly on images captured by the camera. But if you remove the IR filter and apply the visible light filter, you'll have a much more efficient situation, because the camera will now see most of the IR light (since the filter cutting it off is gone), and still no visible light at all. And that means that much weaker IR light source will be seen as clearly as the very strong sources with the IR filter present. Let me give you some examples: You probably already seen an IR TV remote trough a camera - as i said before, it's just a faint light. Well, after the IR filter is gone, you can literally use a remote as a flashlight. It will actually illuminate surfaces it's pointed at (it can be best seen at distances up to 0,5 meter). Another example - at first i was using an unmodified webcam with a strong IR filter and a visible light filter applied. So i needed to boost the LEDs output really high - they've been running a little above 100mA, and were draining battery really fast, not to mention heating up But they were perfectly visible even at large angles. When i tried to use the same setup after removing the IR filter, it was unusable - the LEDs seen trough the camera were so bright, that they were blinding it, and lens flares were so large, that freetrack couldn't locate the points themselves, even with minimal exposure and gain, and threshold set to max. So, to make long story short - applying the visible light filter cuts off all background lighting that could interfere with tracking process. Removing the IR filter increases efficiency and allows to use weaker IR light sources.
http://www.free-track.net/english/hardware/webcam_filter_removal.php?PHPSESSID=551e199641421ad0f9361a971663bccd And please, search the forums a bit, most of your questions are already answered there... |
egore | #36 10/01/2008 - 17h03 |
Class : Apprenti Off line |
Hey Sean - Glad to hear you can finally "see the light"!
Regarding this comment and the jitter issue: "If I back the exposure and gain up the jitter is real low but it won't track real well on the extreme angles." I would suggest one last time before de-soldering: put the exposure and gain where you think it needs to be (your above quote), but then go back to the main screen and adjust the threshold slider. Likely you will be able to find a good balance. It seems odd that you are having such issue with basically the same setup as me. Try the above and get back to me. Good luck! |
Redwulf | #37 12/01/2008 - 03h50 |
Class : Apprenti Off line |
Thanks Egore,
I had tried that. Same thing. How much system do you have? It could be a system issue. I know that my webcam is running more sytem then most. i dont' think we have the same webcam either. Which are you using again? It doesnt take much to file the LEDs down. I'm still gonna do that. I'm in the middle of painting an R/C airplane and a bunch of other things. I also just got an X52 stick and have been a bit distracted from the Freetrack. But it's real close to working excellent. The hat switch is not very comfortable on this new joystick. So getting the freetrack working is going to be high on the list. Sean |
FreeTrack Forum > FreeTrack : English Forum > Support : Tracking System > IR LED Alternate
> Stats
2 user(s) connected during the last 10 minutes (0 member(s) and 2 guest(s)).
Powered by Connectix Boards 0.8.4 © 2005-2024 (8 queries, 0.019 sec)