How to see infrared

I initially assumed he had higher sensitivity to light after so much time spent inside the hospital, or at least that's how i interpreted his eye movements when i first met him.
This higher light sensitivity brought me to think that the only way to have an accurate reading of his eye will be thanks to infrared rather than visible light.
That's why i modified a couple of cheap webcams by removing the IR cut-off filter, thinking it would be sufficient.

Of course it wasn't sufficient, i then rabbit holed that we also need an infrared source of light, so back to the drawing board.

The webcams I use have limited space internally, and as I don't have a multimeter at hand I can't currently tell if all the LEDs actually light up by default.
For now the idea is to externalize the LEDs to some other PCB, but if I ever find out the SMDs are actually working I have these in mind that should fit.
In the meantime I ordered some other LEDs and currently playing with an optimal way of irradiating the eye without exceeding safety levels, according to IEC 62471 this means a Risk Group 0, or RG0.

Of course in the datasheet for the HIR333_A infrared (the LEDs i ordered) there is fuckall information about the risk groups of IEC 62471, as it's the first time i ever read the seemingly random character combination "IEC 62471" and you might be wondering as well why this matters.

The standard is for evaluating the photobiological safety of stuff emtiting light, including LEDs. It assesses hazards like retinal and skin damage from optical radiation and classifies them in 4 risk groups, from safest to most dangerous.
For the sake of our (human) use case we try to stick to RG0 (Risk Group 0) or RG1 at most, I may arrive to tolerate RG1 for testing purposes but i'll strictly enforce RG0 for when it comes to production (him testing it).

The only reason for this research is that while our eyes can tell us when too much light is hurting us (ever tried to watch the sun with your naked eyes?), they have no idea what's going on with infrared as they can't see it.

The vast majority of the infrared sensors sold around is in the IR-A band (780-1400nm) which means that we're still in the "hot" zone for the skin.
IR-A irradiates the hypodermis, so you might think we probably want an IR-B (1,4 μm to 3,0 μm) or IR-C (3 μm to 1 mm) to only affect the outer layer of the skin.
While this might be true for the skin, the eye is completely different.

IR-B is even said to be beneficial to the skin according to this old paper, found reading this other paper. I like papers.
Here another one about how infrared can be used to target tumors along with a 3D-printed graphene scaffold. ✨Magic✨.

If you live in the western world you might have seen a movie in which people go to places to get a fake tan, and in those little "light tanks" they wear glasses.
Those are ultraviolet (UV) cabins, which is on the other (out)side of the visible spectrum compared to IR (see picture above).
Regardless, playing with infrared we should employ a similar protection level and a certain amount of fear, which in this case is sane and justified.

The vitreous humor makes around ~80% of the eye, and the main issue with it is that is mostly made of water, which does absorb infrared as well, along with the virtual totality of the eye.

Overall, the eye is an amazing, completely exposed organ, and is very sensitive.
This sensitivity complicates our work, as infrared light can heat the eyes causing cataracts and retinal burns, which are both bad things in case you were wondering.
To limit heat deposition in the eye we probably want to stay as close as possible inside one of the valleys in the link above (here again), possibly even far-red.

According to the above picture, IR-C stops at the cornea, while B may reach the lens and C bypasses it completely, hitting the cornea.
Image above taken from this paper on photobiological safety, according to which light on the 380 — 1,400 nm may bring retinal thermal injuries.
But again, IANAD.

Due to the eye's DOF (yaw and pitch) we can't just limit the IR LED's FOV to avoid the pupil/cornea area, so we need a safe LED or a way to block most of the harmful light from the eye.

Thanks to OSRAM we don't have to do that, as according to the above document we can use Hyper-red LEDs, as they typically fall into Risk Group 0.
So yeah, far-red or hyper-red it is.

To answer the title's premise: use a telephone camera.

I'll keep you updated with my findings 🦝