College project: weird requirements 'Help!' ;)

College project: weird requirements ‘Help!’ :wink:

Firstly, I ‘think’ this is in the right place. I am asking for advice about building a light with certain requirements.

1.) I will use an infrared LED (Quoted Oslon Black Series 850nm 90deg SFH4715S) this is good for about 3 watts in and 1 watt out (at Tj 25C Vf2.75 typical and current of 1amp) with a 1*1mm die and a 90deg (45deg half) angle output (sounds pretty damn similar to the old cree X-RE, though not the older .9*.9mm die, still sounds quite focusable though)

2.) I want to use a quality host for this, some sort of P60 that screws in and feels ‘nice’ in the hand (good threads, decent thicknes, good thermal management)

3.) I would like to use optionally single 18650 and/or dual cr123 batteries; if I must choose then I prefer pair of CR123a primaries.

4.) I would like to drive it pretty hard, but I do want to operate at a Tj of 100-125 (or so, something like that if possible) in a Temp ambient of something like 10-30 deg C or so (50 - 86deg F). I can go a bit higher or lower but not a lot.

5.) I would like to operate this LED with a Tj of about 100 deg celcius but not more than 125 deg celcius. The LED itself has a thermal path of “11 deg kelvin per watt” (11 deg C per watt), I know this is not up to modern standards but that is ok.

6.) In order to quickly achieve the Tj of 100-125deg C I prefer a turbo mode that can thermally regulate down to a less overheating level after the temp warms up to desired operating level. (I do know that the higher temp means less efficiency and shorter life of LED, that’s ok).

7.) In order to maximize output I would like to use a direct path mcpcb. The pattern on the bottom of this led is similar to that of Cree XB~~* (XB~~ H/E/G/D), but NOT nearly close enough to ‘normal’ cree leds (XM-L, XP-G, etc). If this requires a high priced ’special custom 1 off direct thermal path MCPCB then I am interested in that; I have heard of such mentioned on the forum in various posts but never seen the link to who does this. I imagine that the price for this is much lower than paying for a copper pill and such like.

8.) I want to use the ahorton aspheric lens with this setup, so I can focus this to a nice long range and narrow beam (I have been off the forums for a good few years, I assume ahorton is still the man for the aspheric/plano-convex focusing optic?, and how the heck do you get the dang spacer it’s been 8 years since I last looked and they are still out of stock lol).

7.) I would like an extended bezel that goes well past the end of the ahorton (at least 2 CM or so would be nice) to allow for threading and insertion of 25mm highpass and lowpass filter lens kit we have at school. The kit is not threaded, simply plain lenses but I could put in a thin metal threaded ring fore and aft of the 25mm lens and put some kind of packing around the edge of the lens to prevent rattle. This does not have to be a ‘quick’ change out but it should be doable. I am certainly open to suggestions in this area as well

Ok, additional notes:

a.) this is for a college project, so it does not have to be cheap for mass production or anything but should not cost more than $250 bucks or so if at all possible. *note at end

b.) Current, Voltage and Modes: I primarily need a ‘high’ mode of at least 1 amp (typical Vf for 2.75V). A turbo mode to heat up and get that initial burst of ‘oh wow even my phone can see that!’. The high mode (whatever it is) must be sustainable for at least 25-45 min. I would like to be able to change out the emitter without throwing away too much other stuff as I understand I will be killing emitters. I would like to save the driver when I do this and if the p-60 pill is not really cheap then I want to be able to save that even if it take a bit of clean up work.

c.) The Pill may be copper/aluminium but I suspect that the brain peeps here may tell me that at that operating current I can get away with aluminium (replaceable and threaded to body like in some lights). If unthreaded brass is enough then I would like to use a higher current and a better pill; at the very least I will be packing in copper/aluminium foil between the pill and the housing but I’d rather use a better setup than this, this is not my cheapie light.

d.) A lot of the things I’m trying to do are tedious and possibly even wasteful in some ways, while the traditional characteristics (output vs cost and such like) could be achieved more cheaply (I ‘think’?). Regardless of that these are the characteristics I am trying to use for my project. On the other hand I (‘think’?) I have suggested and achievable price point for the total package I want to obtain.

e.) Advice: I am willing (hoping) to take lots of advice on this and productive discussion is welcome as well. I know that I am asking for a very specific set of uncommon characteristics to be desired but this is what I am trying to accomplish for this project.

f.) I appreciate any feedback and advice anyone can give me, especially if you know your thermals realities. Just think about the old xr-e, think that had 9degC/watt… nope, data sheet says it was 8degC/Watt, should give some bit of a guide to performance. the temp limit on this led is similar to the cree XR-E (145deg C for this, 150deg C for the 7090 XR-E). I don’t need magically exact levels of temp control although if the Driver can handle some of this function that would be great.

g.) I do not know how to program a micro controller and I will not learn this in any kind of time frame for this project

h.) note cost I wish to be realistic, it will cost what it will cost so if it’s cheaper and almost as good TELL ME, if this can only be done for more TELL ME : ) I am not trying to make a batch of these, just one and a replaceable emitter (even if it means higher initial cost, I don’t want to spend a fortune on parts when I replace my fried emitters, especially if I guess wrong and cook the emitter right off).

In advance; Thanks for any helpful input and suggestions

I don’t know about all of your specifics, but you will have to electrically isolate the PCB from the flashlight on a DTP Oslon Black IR because of the way it is designed. It can easily be built on any of the aspheric lights that are out, the 1503, 1504, 1505, but they all use 18650’s. The Oslon Black IR’s can be easily driven at 1.4A. Good luck with your project.

If you want to use the Ahorton lens take a look at the cqng zoom host. Link to forum thread: CNQG host with Ahorton lens

Thanks, I hadn’t thought of that. I still think that except for IR cree is the king in general (been thinking that since 2003 when I got in and they were neck and neck ‘almost’ with pre-phillips luxeon). I wonder if there is a better more powerful IR out there, if so I can ask if I can use it :slight_smile:

Was also thinking that if I make the pill swapable I can add a couple of white leds (they will be fun to filter due to the half blue half yellow output). The heat can also demonstrate phospher performance degradation

Oh yeah, that’s a good thread. Appears that all I need is to find a bezel ‘ring’ that is actually a nice long hood made of metal.

Anyone know where I can get a pretty standard thread patter bezel “Ring” that is really long, if only the weaponized version is available perhaps I can file it down if the crenulations are not too bad.

I don’t need to use Ahorton lens, any good one is plenty fine but I would like to be able to use the lab 25mm filter lenses if possible and I would rather not end up with a giant 70mm super heavy plano-convex