Infra red light => help needed to build one (or two)

I have a friend who is a helicopter pilot. Works for army. They’ve been equipped with IR googles recently also the aircraft gained huge IR reflector but recently he asked me to prepare an IR flashlight so he could light up surroundings just before picking up the good place for landing. He aims to light up to 150 meters so probably I’d use C8 size body.

But… I’m totally green if it comes to IR flashlights. Not only I don’t have a night vision googles but also I’m suffering from lack of any experience. Gulp!

I’ve browsed BLF a little but found just 3 old threads not saying much to me so I dared to start a new thread. I hope this is not a rocket science and using parts from KD or Ali I could successfully build one.

I found at KD the following LEDs from Osram:
SFH 4716AS IR850NM
SFH 4715AS IR850NM
SFH 4725AS IR940NM

And one from Luminus:
SST-10-IR IR850NM which I discard due to 5V requirement.

My questions are:

  1. What are the differences in Osram LEDs? Besides tiny details they all appear the same to me.
  2. Which driver would suit them best?
  3. How far can I overpower them? Nominal is 1,5A but I hope to push them harder.
  4. How much heat is generated comparing to the white LEDs? Is it possible to use a driver with temperature management?
  5. Are there any pitfalls that I should be aware of?
  6. What’s the difference in wavelength? Is 940nm any better than 850nm? Please mind I don’t know the googles specification. Those army toys are top secret all the way through…

Any hints and tips are most welcome. Count on you guys!

For amateur useage the choice in IR lights is rather limited. AFAIK only 850nm and 940nm leds are on the open market. 850nm wavelength is closest to visual light, barely visual as faint dark red by human eyes. In principle 940nm is not visible by human eyes. So I would ask your friend the chopper pilot what his preferred wavelenght is.
Military systems often operate with a mix of methods, wavelenghts (and conversions). Like a night vision amplifier, which uses the last remaining bit of visible light.

Yeah, it’s quite obvious 850nm is closer to what a human eye sees. I’ve certainly asked him about wavelength he prefers, but he doesn’t know. I’ve also asked for the specification of goggles, but he was only capable of sending a picture which says nothing. So probably 850nm would be safer to start with since it has a bigger chance of suiting his needs. Hopefully.

I would think for potential military use you would want to stay as far away from visible wavelengths as possible.

For SST-10-IR 5V mentioned is reverse voltage, max voltage that can be applied in reverse direction without damaging LED
From spec it has Vf about 1.5V

If you want SST-10 look at SST-10-IRD (e.g., SST-10-IRD-850nm), this is dual junction version with Vf ~3V. This is close to white LED and will work better with drivers made for white LEDs, it is also brighter with the same current (dual juction = 2 LEDs in series - that’s why Vf is twice of single junction led)

850nm would be better I think, it will appear brighter than 940nm, 940nm is used when it is important that light source is visible as little as possible.
I guess 810nm mentioned above will appear even brighter than 850nm but I wonder if at that point beam is still completely invisible for naked eye (with 850nm you can see light source but beam is invisible)

If that’s not enough power you can look at CBM-90-IRD. You will need high current driver and way to fit emitter that is pretty big and oddly shaped.

I can’t help on the emitter but I’m curious. As for the host, and considering he’ll be trapped in an airborne cage, I’d just make sure that you pick something that has dual springs and maybe consider going double-spring on one or both ends just to give the best impact protection for the cell in case it’s ever dropped or knocked hard. Consider potting or silicone foam thermal blocks, too, maybe. If the light ever decided to heat up and vent it could be more easily tossed out from a helicopter, I suppose, but best to avoid that scenario entirely. Maybe an anti-roll ring on the bezel if you can cobble something together (Maglite used to have some larger ones for the C and D cell lights, if those are still around).

OSRAM LEDs you linked are dual junction as well (if not stated in spec you can tell by Vf)
You can use them as replacement for white LED, I mean you can pick white LED host and replace LED keeping original driver as long as output current is within spec of IR LED).
That would be my first approach, if this is not enough power then we can start complicating things :slight_smile:

Another option, you can look for P60 IR drop-in.
They are not so popular anymore but you can still get them
Host for example

I’ll be really interested to see the outcome, and hope it’s successful!

That said, I don’t know if you’ll generate enough light in a small handheld light.

If the goggles mean the C8+ looks like white light equivalent of 5000lm…. My Q8 does similar output, and having been in a helicopter (during daylight) landing in the bush, there’s no way I’d be comfortable if the pilot could only see the small patch of ground that’s lit up.

Maybe the Heli has extra radar bits to help or maybe they’re aiming for huge landing spot?

Given they’ve acquired the goggles, how come no accompanying light?

Thanks guys, really appreciate your input. As for now, I lean towards osram 850nm in C8 with 5 or 6 AMCs.

Gr8 info, well appreciated. I wish Neven’s store was still opened. :cry:

I can’t enlarge your images but if the LEDs are the same except for the dome then pretty much the same amount of light will fall within a 40 degree cone(?) for the 5AS as will fall in 75 degree cone for the 6AS.
(I use the term ‘cone’ as i don’t know the technical term for the shape of the light distribution, but it looks like a cone :D)

In other words, more light is projected forward from the 5 compared to the 6 so the 5 will work better in a zoomie as more light will fall on the lens and less will be wasted to the sides…
More light is projected to the side from the 6 than the 5 so the 6 will be more throwy in a reflector as more light will hit the reflector (which is to the sides of the LED) and be reflected forwards.

A little update. I’ve talked to the friend again and it turned out it’s very important to be discreet. So following your advice I’ve ordered 940nm Osram, hoping it’ll be impossible to see it with bare eye.

I’ll pull up the datasheets later for a peek but they sure look to be pretty power hungry as current goes up, like UV emitters. I would assume that “half angle” here just means double the number for your actual beam angle of the bare emitters. The 16 is quite a floody little dude…maybe intended for more of an open air beacon, or to cover a larger area when used for gesture recognition, etc. Sounds like the 15 is what you want here for a flashlight (or, say, a tv remote…).

I’ve built this flashlight but I can’t find what’s wrong.
I’ve burnt 4 drivers already and I’m totally pissed. These leds seem to be awkward.
I used 6x AMC Convoy biscotti drivers, all gone burnt as soon as I installed the Osram IR. When finished building I installed first white led to set the desired group of modes. But as soon as I install the IR 940nm led, the driver spring gets compressed (due to high resistance i believe) and smoke appears in drivers cavity. Last time even the red driver’s wire got unsoldered.
I have two of these 940nm leds. When I test them with my multimeter I receive weird reading. I get 1998 in both cases, while the white leds give always the value 1.
Any ideas what’s going on?

I feel really bad because tomorrow my friend goes to the mission on Eastern border and he was hoping to take this flashlight.

Total disaster :person_facepalming:

Guys I need your help more than ever.
Could it be these leds require special pcb? I remember in led4power shop there was a special insulated pcb for IR diodes.

sorry your DIY did not work (yet).

I hope you figure out how to fix it

meanwhile, maybe buy the light suggested in post #9?:

Well… I thought this would be easier.
I even reflowed the leds to L4P insulated pcb. It’s not working either, but at least it doesn’t ruin drivers any more. And multimeter shows value 1 after reflow so I suspect KD did bad soldering in both pieces :person_facepalming:

Anyway this is a major failure. I’ll try to play with it some more later, when my anger goes away…

I think a C8 with 3-5w ir will absolutely not do it. i have night vision scope, and i build 10w 850nm ir light, but it would not be anywhere close to enough to spot a landing site from 150 meters, i’ve flown in choppers before, you need to see vegetation, landscape\slope, what can be picked up by rotorwash……when you look for a landing site. i think he needs at least 50W of floody light, to see a big picture and make a fast decision. As far as discreetness, a helicopter hovering at 150m will be way easier to spot than a tiny red light that is barely visible from point blank, let alone 150meters up in the air. not to mention the other guys on the ground will see your light with their nvg too. If i wanted to be discreet i’d use thermal sight. it needs no light,

Sure, they are already equipped with huge IR reflector mounted outside the cabin. But he says it doesn’t help much at closer distance. And it drains the battery quick.

Of course, It does not help up close because it has a narrow beam, so does C8. your best bet would prbly be getting a soda can light with multiple leds and replace them with ir leds, the more the better. to get a flood down 150m you need a lot of watts, i do not know any formula how to calculate wattage for given distance\area, but judging by my 10w light, you’ll need 3-5 time more. imo. oh and mine is 850nm, if you use 940 you’ll need more power, prbly at least 2x as much compared to 850. it would prbly easier to build one from a scratch, than find a host that will fit your needs, how much time do you have, and how good are you with tools?