Light focusing tube

of 120 degrees down to a small spot

ns?

object, so it is robust against production changes

Yeah I don't know of a lens that does that. If not a reflecting light tube , then maybe a glass or plastic light pipe. (big fiber optic type thing.) Can you butt the led to light pipe to detector, or does there need to be a gap?

George H.

ote:

2, but a chip of about 0.1mm2. So I would be happy to have the spot of just less than 5mm2. I am ballparking here, I would like the spot to be bigger than the receiving LED, so that I can have production equipment that does n ot need to be top notch to "cover" the receiving LED in photons

are common to illuminate front panels from LEDs on the board behind. If th e path is straight the light pipe just becomes a plastic cylinder. They ma ke them in many shapes and sizes with various attachments.

ound no numbers in the datasheets I looked into

They work with total internal reflection, so I think very little. Look at one in operation. You don't see much light lost extraneously. A light pip e does not act as a collimator. They essentially make the light appear as if it is coming from a closer point. But light at too wide an angle is out side the angle of total reflection so is not conveyed.

--

  Rick C. 

  + Get 1,000 miles of free Supercharging 
  + Tesla referral code - https://ts.la/richard11209

of 120 degrees down to a small spot

ns?

object, so it is robust against production changes

It is unclear what "object" is here. If the variable distance object is th e receiver, you could use two lenses. The first focuses the light to a sma ll diameter point and the second lens makes the light into a parallel beam from that point. Then the beam will have low divergence over a range of di stance.

If the "object" is the light source and the lens is on the receiver, you ar e screwed I think. You can't use a lens.

--

  Rick C. 

  -- Get 1,000 miles of free Supercharging 
  -- Tesla referral code - https://ts.la/richard11209

of 120 degrees down to a small spot

ns?

ar

back and forth between the inner sides of the tube, and when it escapes the end of the tube it still has 120 degrees?

Yes, exactly. A light pipe will be the same, but 120 degree light most lik ely won't be reflected to the destination end at all. lol

--

  Rick C. 

  -+ Get 1,000 miles of free Supercharging 
  -+ Tesla referral code - https://ts.la/richard11209

Its a classic approach when using photomultipliers in huge tanks of liquid neutrino detectors to catch as many photons as possible with the smallest number of tubes. Also for non-focussing solar concentrators.

The analytic solution for a device that causes every ray path that enters the front aperture to hit the target (eventually) is a pair of parabolas placed so that they each have their focus at the base of the other.

Is a reasonable introduction.

--
Regards, 
Martin Brown

Fiberoptic light pipe: The 60 degree entry angle worries me. I guess try it and see if the pipe grabs all the light, or just get a wide diameter fat pipe. There are adapaters to install a lens on the end. Some interesting stuff:

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

Many years ago, I wanted to build a laser photoplotter, for making PCB artwork. After fooling around with LEDS and conical concentrator schemes for a while, I got the first red lasers when they came onto the surplus market (kind of dates the project). I used a 3 mm sphere lens almost touching the front of the laser to collimated the laser as well as I could. Then, I put a pinhole in front of the sphere lens, to get rid of light off the edge. Then, I had a 13 mm F.L. microfiche objective lens to focus onto the film. My rig actually has a double-meniscus lens in the middle, to make the spot smaller. I'm not sure that is actually needed, but it all works. It can produce a 670 nm red spot so much smaller than the .001" stepover that I have to slightly defocus the thing so the raster lines merge together.

Jon

pipe, found no numbers in the datasheets I looked into

But that is only one source of loss; there is also absorption. You won't be able to see that. Presumably a pipe made for the purpose will be a good quality optical material...

Mike.

You could get some Lucite/Plexiglas rod, and heat the middle of it until it softens, then pull the ends. This will neck down the middle. Then, cut at the desired small-end diameter and polish. But, some physics papers claim that you cannot concentrate light this way. I'm not convinced.

Jon

You probably need a flat-face LED and a lens. I don't think a tube would work.

How about a laser? Pointer type red lasers are absurdly cheap.

If there is a tapper in the tube and it becomes a cone, then for short wavelengths, (compared to the tube size) you can use ray tracing, with angle of incidence = reflection, when you try and focus this way you have to 'give up' some light.

(for a reflecting cone it's a battle between the angle, number of reflections, and the loss per reflection.. Photons entering at large angles have to reflect too many

At some wavelength fiber/ glass may become 'better' than a metal reflector. Our tapered tubes where for collecting light from a source that had already been collimated (sp) and sent through a FIR scanning interferometer. With an extended source (LED into ~120 deg)* collecting ~1/2 the photons is hard enough... if you want to focus them you've got to give up more.

George H.

*Hey LED's are giving up 1/2 their photons to the back... anyone putting a reflector in the back? (might be a little cooler too?)

Lasers are way better than LED's as light sources. Above threshold, except for losses at the end reflectors, every extra electron turns into a photon... hmm, I think that's right?

George H.

On a sunny day (Tue, 13 Aug 2019 17:28:52 -0700 (PDT)) it happened George Herold wrote in :

Open up a LED, some have a small hollow mirror like structure that the emitting part is mounted on:

.)----------- | __________

Jan Panteltje wrote in news:qivo1q$lvc$ snipped-for-privacy@dont-email.me:

Those are the lame low current traditional design jobs that have been around for decades.

The high output high wattage jobs are not configured that way (no tiny dish) and all their light is directed forward.

$1.82 each

A nominal 120 degrees of spread includes most of the light, so a 5mm 'pencil' beam would result from a lens greater-or-equal to 5mm diameter, at a distance of its focal length (so the lens acts as a collimator). Minimum focal distance would be the distance at which 120 degree spread from a point hits the 5mm diameter, about (5mm * cot(120deg/2) )/2 =1.44 millimeters

That doesn't seem practical (it'll waste lots of light if you let it spread farther), but it gives you a pencil beam for a long working distance. A light pipe will put the

5mm spot at end B after cementing the LED to end A, but light spreads at the 120 degrees starting at end B (so do you really want a light pipe butted against the receiver?).

A laser with a telescope-like 'beam expander' can give collimated light, too, but it's nice to have some diffusion (because the 'spot' might have more light/dark modulation than you'd like, and only the small receiver area is sensed, not the average).

The best light concentration, alas, results from focusing NOT to a pencil beam, but to an image of the emitting die (and that, too, doesn't usually look like a uniform illuminated spot).

ns?

Sounds more and more like you need a cheap laser diode w/driver then, not an LED. Does the spot of light *have* to be green?

Mark L. Fergerson

I guess you need some sort of lens then, but you cannot focus the spot any smaller than the LED die because entropy.

--
  When I tried casting out nines I made a hash of it.

Sounds to me like you are down to either an acrylic cylindrical light tube in contact with led and detector and just take whatever spot you get on the detector, or define what you mean by "indifferent to the length from the object" to see if the depth of focus you can get from a collimation lens is sufficient and accept that the spot will always be larger than your detector.

--
Regards, 
Carl Ijames

I wrote

Like this:

and