At work I have to get things perfectly dust free and I am struggling with that a bit.
I use a microscope that has strong downlight and one light that shines from the back.
Since I found out that those HDD repair people use green light, I have put a filter in front of the back light and it is definitely better. I don’t use the downlight for dust detection.
I also use a small 365nm UV flashlight which detects sometimes something that I didn’t see with the green light.
I also tried with a warm white light and with that I see a good amount of dust that I don’t see with cold white light.
But I can’t always check everything with 3 different light colors, it just takes too much time.
Is there a light or combination of light colors that really bring out dust particles?
It’s not necessarily about colours or wavelengths unless you work with specifically coloured dust. The way to really highlight it is to illuminate it at a shallow angle, like in wallgrazing, that way the particles cast longest possible shadows making them easiest to see.
As a corollary, the light itself should be on the throwy side, so that the light sent out is sufficiently parallel and does not illuminate a dust particle from multiple angles, which obscures its shadow. A compact thrower of the S2+ size works well; too big, and the light won’t collimate correctly for short distances.
I’d say it’s less a matter of throw/ flood and more a matter of the LES being as small as possible (being a point source) with the ideal being a bare single LED. Throwier lights tend to have larger reflectors with large ‘exit pupils’ (not sure if this term applies exactly here) and especially up close they don’t behave like point sources at all.
This is curious observation. Since a cold white light (high cct, like 6500K) will light up dust particles, smoke, water droplets, snow or mist in the air. To the point that it’s hard to see far with a thrower, as the beam is more visible than what ever it is that you try to point at. Not sure but I think it’s called backscatter.
Warm light (low cct, like 3000K) will penetrate the atmosphere much better which makes them more useful in bad weather. But in your case I would guess cold light would work better. But I barely have a clue what I’m talking about.
Also, might be good idea or bad, but I immediately thought of a cheap laser level or other type of laser line applied from the side to cast shadows. Like, from a regular hardware store. They should be pretty safe in normal use but not sure if reflections or such in combination with microscope lenses would be eyeball friendly…
I just did a quick and dirty 5 minute experiment (s).
I knew going in that the angle is important. Lay your light down on the floor or on the countertop or on the wall to look across the floor or across the wall and you are going to see things that you will never see looking straight on to the wall from 3 ft away.
But what if you can’t get the light on the same plane as the surface?
Then it seems like there are many more variables. The color and texture of the surface matters. The other ambient light color and source direction matters.
In some instances a spot works better than a flood and vice versa. And sometimes warm or neutral or cold or UV works better.
When I have more time I’m going to experiment a little more.
But I don’t think there is a definitive answer.
Even with the light on the same plane as the surface there are variables.
A point source is ideal. The reason I recommended a compact thrower is that most flooders/mules have a beam that diverges so quickly that one can only use it to spot dust at a very short distance relative to the head diameter of the light, which means the dust is being illuminated from many different directions (the entire head). Perhaps an aspheric like Convoy Z1 would be a good choice–from the perspective of the target of the light, it’s essentially a point source with variable size.
I use this technique you suggested for picking out tiny SMD components I drop on the floor.
So I already have a shallow angle and somewhat pointy source. The beam can be adjusted, but it’s as small as it can be.
And you can see that even a 14500 flashlight would be way too big.
I use one of these little lamps with a UV ( CUN66A1G) LED in it:
The warm white light that I have tested is one of the side lights of this lamp, 2 SMD LEDs ~1cm apart, so not super pointy but very close and of course super adjustable to the most shallow angle possible.
I have ordered a 219 in 1800K, a 519A in 2700K, a LUMINUS SST-20 in 660nm (red) and I am about to order a blue and green XP-E2.
I will put these LEDs into the aforementioned lamps and test them out one by one.
The green laser has a lot of the right optical characteristics for this, though one issue I imagine is that as a coherent source, it has lots of splotchy interference patterns that may obscure the dust. I had imagined the slit (and the laser itself) to be oriented horizontally to maximize the coverage, and at the same time, the reduced vertical height would make picking out dust easier. I imagine that the room would be devoid of ambient light when you actually do the inspection.
What about a grounded clean box or room of sorts? Have air constantly circulating only in the clean area, and put a static filter to catch any particulate.
I would love to have a air filter box like for those HDD repairs, but that’s not possible here.
It’s a normal large room, normally lit, ~60 People in it, no cleanroom, no pressurized room. Just sticky mats outside the door and extra shoes and a lab coat, that’s it.
The air that gets pumped into the room (we can’t open any windows) is filtered, though.
So, all I can do is make the work easier to do, and that’s why I initially put a green light filter in front of the lamp (which by the way works awesome!).
I am curious how much a pure green LED will bring me, or red, or blue, warmwhite…
I will report
Pretty cool! I think with a laser you really need to go parallel with the surface being examined to visualise particles properly. Also, a more efficient way of obtaining a laser lightsheet is by placing a cylindrical lens (like a glass stir rod) in the way of a collimated, pointer-like beam instead of a slit.
EDIT just realised it’s a filtered wideband light, not laser, nevermind. It’s still something of a lightsheet so going parallel would be the way to make it work its best.
Also, beware strong green lasers around reflective surfaces, most green lasers are DPSS- they contain a powerful infrared laser diode and a strange quantumy frequency-doubling crystal that converts some of the IR light into green light. The process does not consume all of the IR light and proper green laser modules have IR-cut filters that only let the green light through but cheaper ones don’t and the danger is in that IR light diffracts differently from green so any optics might place the potentially dangerous invisible hotspot away from the green one.
I can’t use a laser in there, the watches and parts I look at under the microscope have polished surfaces and I don’t want a laser hitting my eye.
Also I can’t really modify the microscope. I could ask if I could take the LED module home with me and plop in a green LED, but that’s about it.
So I think with a flashlight and a green LED I will be better off - it’s also more versatile
