| Main > Lightguns |
| GUN4IR - The Ultimate 4 Points Lightgun System |
| << < (204/227) > >> |
| BadMouth:
I haven't followed closely, but noticed recent posts about using a wii camera. I have two awesome looking third party Walther P99 replica Wii guns from way back when we were using glovepie scripts. :oldman To convert them to GUN4IR using the cameras already in them, would I be able to just follow the instructions or is there going to be a lot more to it. I'm not looking to go down a rabbit hole......but I'm accustomed to ending up going down a rabbit hole. EDIT: I got 8 9 numbered pins (with two sets combined) where the camera board is soldered to the next board. A quick search for Wii camera shows 8 pins. There are lots of tiny surface mount components on the camera board though. |
| BadMouth:
It appears there is a piece of software called Lichtknarre that uses 4 IR LED tracking with unmodified Wii remotes. Looks kinda janky, but I am going to try it. Hopefully I have some bare IR LEDs somewhere. Might try it with the dolphin bar first just to see how the software functions. |
| JayBee:
--- Quote from: RandyT on October 06, 2022, 08:44:31 am ---I understand all of that and suspected as much about the MCU levels. But there is a real-world situation which could end in a dead or damaged Wii camera every time. What if a user has an Arduino lying around (who doesn't) which has been used previously for tinkering and has some other code present which has one of those pins set up as an output already and set to high level, or as an input with internal pullups enabled (pretty much common for a switch)? If they build the circuit first and then apply power to program the arduino, 5v will be connected to the data pins of the Wii camera. So with that particular circuit, it is absolutely imperative that the board is programmed and verified to be working properly prior to attaching it to the rest of the circuit to avoid possible damage. That's why I stated that it will work but was a little spooky. --- End quote --- That is why I always recommend flashing the arduino first and/or adding a level shifter to the mix. And as I state in my guides, those DIY schematics aren't for beginners. You must know what you are getting into. There is absolutely no way of making a DIY circuit foolproof. But on the few thousand guns made to this day, none got their cams fried for that reason. --- Quote from: RandyT on October 07, 2022, 11:19:25 am ---There's also something else not right about that circuit for the Wii camera. The pinout for the camera calls out 25mhz on the CLK line, but the schematic shows a 24mhz oscillator. Is the camera being underclocked intentionally, or is one of these in error? Starting to think that this is the reason 1942 is not having success, as he is using a 24mhz oscillator, and every reference I have seen to this part calls out a 25mhz clock. --- End quote --- There is nothing wrong there, the cam itself supports between 23 and 25MHz. I have been using cams with 24MHz for years (as it was the easiest oscillator to get), but 25Mhz works as well. --- Quote from: Zebidee on October 08, 2022, 08:51:30 pm --- I know I'm coming in sideways and haven't read all the previous 25 pages of the thread, so I apologise if I misunderstand the context, but... It is generally better to run your LEDs in parallel rather than in series. That means each LED has its own resistor and 5v power (you daisy chain the 5v, much like we commonly do for grounds). If you have Vcc=5v, forward voltage over LED = 1.2v, want 10mA current, then 380R (for each LED) is the "right" resistor. This is the safe minimum, you can use higher value resistors safely (the LEDs will be less "bright"). As each resistor/LED pair is only passing 10mA current, you can use any old 1/4W or even 1/8W resistors, don't need to be high wattage rating. If wanted, you MAY add another resistor (one, in series) on the 5v supply, to all LEDs, to limit the current a bit more. Again, everything is going to be <30mA so 1/4W rated resistor is fine. You CAN wire all the LEDs in series behind a single resistor. However, in practice this often means the last LED is struggling to get enough current, especially as even the best 5v supply will have ripples. Makes the quality of your power supply more important. Also means a single fault anywhere prevents all from working, and makes it more difficult to troubleshoot. Love this project, been following for a while, really happy to see more great lightguns happening :D --- End quote --- Putting LEDs in parallel is generally a bad idea for few reasons; - The more power you need to drop/limit with a resistor, the more power it will need to dissipate, increasing its heat and minimum rating by a lot. The rating isn't only related to the current going through it, but also the ratio between input voltate and led forward voltage. To give a small exemple, the leds we are currently using/recommending ideally need 80mA, at around 1.5v. That's only 37mW per resistor when they are by 3 in series, but 261mW if you use them one by one. - The power dropping on last led is also happening in parallel, it's even worse if the wires/resistors/leds have some variations, or if you are using the power supply max current. We use them by 3 because even if there are some variation it's not so much noticeable. - More components and more soldering also means more possibilities of failure, especially for DIY stuff :lol --- Quote from: RandyT on October 09, 2022, 08:21:58 am --- --- Quote from: pacooka on October 08, 2022, 08:06:47 am ---For the IR LED, I would like to use the black 940nm IR LED SIR5028B instead of the transparent ones because I heard they perform better. By looking at the calculator, I see that for VCC 1.2V I should use 47 Ohm (for transparent LEDs - 27 Ohm); can someone please advise if this is correct, or point me to the right black LEDs & resistors? --- End quote --- The current to radiant intensity curve on this part looks to be very linear, and I assume that you want the maximum amount of brightness given that you seem to want to use this part for it's performance characteristics. This particular LED shows to have a continuous current rating of 20ma, so driving it at half that (your stated 10ma per part), while possibly increasing the longevity considerably, is likely to be giving up a good chunk of that performance. If it were me, and based on the data for the part, I would run them "hotter" at 16-18ma which should give them a higher radiant intensity, while providing a bit of a safe zone. Given that it's only 3 LEDs, I'd have no qualms about running them in series with one appropriately sized resistor so long as that worked properly for the application at hand. However, you do need to be aware that when one LED burns out AND dies in the shorted state OR is no longer consuming it's share of the total current, the other LEDs in the series will also be damaged. I consider this arrangement a possible problem only in the event of a premature failure, as IR LEDs tend to wear out at a fairly even rate across the board and usually quicker than other types. So just like the brake pads on either end of a vehicle, if you replace one side due to wear, you might as well replace the other because they are probably due. As for the proper wavelength, there is anecdotal evidence out there for 940nm being optimal, but to really know for sure one would need to look at not only the specs for the sensor, but also the specs for the filter material in front of it (if it has one). Ideally, the filter should be tuned to pass the wavelengths at the highest levels, which match those most favorable to the sensor. In other words, if the sensor has the highest sensitivity at 850nm, but the filter blocks 15% of light at that frequency, it will be exactly the same as feeding it unobstructed light at a frequency to which it is 15% less sensitive. So gains will be limited if the optical components are not properly matched to begin with. The issue is that these components seem to be hacks built on hacks which have little if any official specifications available, meaning that more hacking (and not just with the LEDs) might be necessary to really use them to their fullest potential. --- End quote --- Failing LEDs usually aren't passing juice anymore, so if one fails it's very common that the others don't have any issue. While IR might have shorter life than others, it is still several years of continuous use for the good ones, I wouldn't worry too much about it. That being said, the LEDs mentionned there don't seem great. The only reason we recommend the other black led model is because of the model specs themselves, and the esthetic. No, the sensor is 100% sure 940nm, and needs a proper additional 940nm filter to work. I have reverse engineered that sensor, studied and used it for years, what I am doing with it is anything but hacky. Heck I am using those sensor better than they have been used in the wiimote themselves. DIY doesn't automatically means it comes from amateur work ;) --- Quote from: Totophe on October 12, 2022, 05:37:00 pm ---Bonjour à tous Je suis en train de fabriquer ma borne et ce projet du gun 4ir m'intéresse beaucoup. Ma question est si je donne la contribution paypal il y a aussi la liste de pièces à commander ? En attente de votre retour bonne soirée. --- End quote --- Oui, une liste globale les pièces nécessaires. --- Quote from: BadMouth on October 13, 2022, 01:38:20 pm ---I haven't followed closely, but noticed recent posts about using a wii camera. I have two awesome looking third party Walther P99 replica Wii guns from way back when we were using glovepie scripts. :oldman To convert them to GUN4IR using the cameras already in them, would I be able to just follow the instructions or is there going to be a lot more to it. I'm not looking to go down a rabbit hole......but I'm accustomed to ending up going down a rabbit hole. EDIT: I got 8 9 numbered pins (with two sets combined) where the camera board is soldered to the next board. A quick search for Wii camera shows 8 pins. There are lots of tiny surface mount components on the camera board though. --- End quote --- Those are most likely different sensors and probably won't work I'm afraid :-\ |
| RandyT:
--- Quote from: JayBee on October 13, 2022, 03:58:49 pm ---There is nothing wrong there, the cam itself supports between 23 and 25MHz. I have been using cams with 24MHz for years (as it was the easiest oscillator to get), but 25Mhz works as well. --- End quote --- It's still odd that the main guy who seems to have started all of this hacking is calling out a 25mhz oscillator. I'm assuming that the value was taken from the clock source on the unit he was hacking, but one would have to assume others were doing the same, so something seems to have changed somewhere along the way. Still, the processor inside the camera has to be affected one way or the other, unless of course it is self-clocking and separate from the camera. --- Quote ---Failing LEDs usually aren't passing juice anymore, so if one fails it's very common that the others don't have any issue. While IR might have shorter life than others, it is still several years of continuous use for the good ones, I wouldn't worry too much about it. --- End quote --- For the most part, I agree. LEDs are cheap enough now that it really doesn't matter. However, I have seen cases where a faulty or damaged LED in series has brought down the others. As they age or become damaged, the luminous intensity decreases. The question is "what happens to the power?" when this occurs. If it's turned to heat instead of light, then no problem. But if it's just not consumed and allows the unused power to reach the others in the series, then it will degrade those as well. I looked for this information, but sadly couldn't find much. --- Quote ---No, the sensor is 100% sure 940nm, and needs a proper additional 940nm filter to work. I have reverse engineered that sensor, studied and used it for years, what I am doing with it is anything but hacky. Heck I am using those sensor better than they have been used in the wiimote themselves. DIY doesn't automatically means it comes from amateur work ;) --- End quote --- Here's where I think we don't agree fully ;) Anything to do with these particular "blob sensing" cameras is a straight-up hack, unless it's coming from Nintendo. The manufacturer made these exclusively for them, and the datasheet is a closely guarded secret, to the point that the legal departments seem to be activated if there's a leak and one finds it's way to the internet. The fact that you can't even get one of the parts without tearing apart an original WiiMote, unless you buy one in a different package pre-extracted by someone else, really says it all. Now, that in no way is meant to detract from the amazing reverse-engineering work done by those who use them in their projects, just "calling a spade a spade". Now to why 940nm isn't necessarily ideal. First of all, if you look at similar public facing components from that same manufacturer, it seems that they call out ~850nm. This could be different for the Nintendo part, so that doesn't necessarily mean anything. According to the internet, however, the camera is sensitive to a much wider frequency band, limited by a 940nm internal filter. Why 940nm? Well, again guessing, these "blob sensing" camera systems are hyper-sensitive to stray light, so pushing them farther outside of the visible spectrum makes sense. On the other side of the coin, you also don't want the sensor to become oversaturated when it is close to the emitters, which again, according to the internet, they are when the 940nm LEDs are used in that proximity. So it's a balancing game, and it makes a certain amount of sense that the manufacturer would call out something "middle of the road" and then rely on the software controllable "sensitivity" and firmware to monitor and adjust dynamically, based on what the camera is (or isn't) seeing. It's probably not necessary to do this and still have it work (possibly perfectly with a specific application), but again, without that 20+ page (based on internet posts) document from the manufacturer, only they and Nintendo knows if it is implemented as well as it could be. And only they know what else that sensor can do. |
| JayBee:
--- Quote from: RandyT on October 14, 2022, 04:32:00 am ---It's still odd that the main guy who seems to have started all of this hacking is calling out a 25mhz oscillator. I'm assuming that the value was taken from the clock source on the unit he was hacking, but one would have to assume others were doing the same, so something seems to have changed somewhere along the way. Still, the processor inside the camera has to be affected one way or the other, unless of course it is self-clocking and separate from the camera. --- End quote --- Yes the 25MHz was only mentionned because that's the oscillator used with it on original hardware, based on the limited knowledge he had at that time. I know for a fact the cam circuit can work from 23 to 25MHz. It does impact its working frequency, but not the way you might think, and doesn't have any negative impact on performance or longevity for our specific use. You seem to think I am relying on other people hacky work, or incomplete info found online to use that sensor? You don't know the half of it :lol --- Quote ---For the most part, I agree. LEDs are cheap enough now that it really doesn't matter. However, I have seen cases where a faulty or damaged LED in series has brought down the others. As they age or become damaged, the luminous intensity decreases. The question is "what happens to the power?" when this occurs. If it's turned to heat instead of light, then no problem. But if it's just not consumed and allows the unused power to reach the others in the series, then it will degrade those as well. I looked for this information, but sadly couldn't find much. --- End quote --- LEDs that fry usually stops conducting current properly. The power not used anymore by the led doesn't need to go somewhere, it's just not used. If one LED in a series fails the other components only get less power (or no power at all if it's completely dead), which won't harm them in any way. The dying LED can very rarely, in extreme cases burn up, crack or even explode, but even then it's rarely dangerous for the neighbor components. Here is a fun list of things that can happen; https://en.wikipedia.org/wiki/List_of_LED_failure_modes But I am sure you already know that, it's basic electronic stuff ;D --- Quote ---Anything to do with these particular "blob sensing" cameras is a straight-up hack, unless it's coming from Nintendo. The manufacturer made these exclusively for them, and the datasheet is a closely guarded secret, to the point that the legal departments seem to be activated if there's a leak and one finds it's way to the internet. --- End quote --- Those sensors are only a variation of an old model that company used to make. Nintendo was just using them, as a client, and not to their full potential, far from it. The reason we are using those is that they are by far the most common and easy/cheap to get. For a full commercial and non DIY product, I wouldn't use those at all obviously. Docs and informations about this sensor and others indeed aren't available publicly online, but it doesn't mean we can't get very reliable info on them through reverse engineering, research and official docs from other models. If I was limiting myself to simple hacks, barely working libraries or incomplete leaked info found online, my gun wouldn't perform anywhere near as it does now. If I wanted a sluggish and unreliable light gun system I would have stayed with a wiimote+software solution :dunno --- Quote ---The fact that you can't even get one of the parts without tearing apart an original WiiMote, unless you buy one in a different package pre-extracted by someone else, really says it all. --- End quote --- They just aren't produced anymore, and haven't been for many years. --- Quote ---Now, that in no way is meant to detract from the amazing reverse-engineering work done by those who use them in their projects, just "calling a spade a spade". --- End quote --- Thanks. And sure, let's call it a hack :P --- Quote ---Now to why 940nm isn't necessarily ideal. --- End quote --- It doesn't matter, that specific sensor (and not the several other models from the same company) is essentially designed to work with 940nm, filter or not. It was made that way not for technical reasons, but because they didn't want the users to see the leds with their naked eyes, as 850nm can be seen, but still wanted the sensor to perform optimally with them (which a native 850nm sensor wouldn't do). Of course the sensor can see the 850nm LEDs as well, because it was hard to make a sensor that only sees 940nm range in a reliable manner. That's the reason why they always need an extra IR pass filter (that black plastic in front of the sensor on the wiimote isn't just for show), as opposed to more recent 850nm sensors. --- Quote ---On the other side of the coin, you also don't want the sensor to become oversaturated when it is close to the emitters, which again, according to the internet, they are when the 940nm LEDs are used in that proximity. --- End quote --- That or the fact light blobs have a limited size in the sensor memory banks and becomes garbage passed that size. There are also several other issues like internal or external reflections and such ;) |
| Navigation |
| Message Index |
| Next page |
| Previous page |