Tiny 2.4mm diameter lens with 1.45mm FL, what application might a lens so small have?

[Laser57]

Found this lens on ebay, it has such a small diameter and focal length it caught my curiosity, anyone have any ideas on what applications a lens this small with this short of a focal length might have?



Geltech Aspheric Lens:
-Diameter: 2.40mm
-Focal Length: 1.45mm
-Numerical Aperture: 0.58
-AR Coating: 600-1050nm
-Material: D-ZK3
-Refractive Index: 1.580±0.002
-Clear Aperture: 1.60mm

[andy_con]

single mode red

[Laser57]

Would't the divergence be much higher if collimating that close to the diode due to so little beam expansion?

[andy_con]

try it and see, then tell the forum your results.

you have the lens, so why not do some testing and tell us, rather than asking

a 2mm fl lens works very well

[Laser57]

Primarily I was hoping someone might know what type of applications such a tiny diameter short FL lens is normally used for....

...but I also asked the question about divergence to see if my thoughts are correct, that letting the raw output of a laser diode expand a few mm more than usual with a longer FL lens would reduce the divergence more than if collimating with a short FL lens close to the diode, due to the thicker beam. I have two people telling me the opposite, one yes, another no, the last opinion I received is that it should be collimated with a short FL lens first and then up-collimated with additional lenses for the least amount of divergence.

With all of the other irons I have in the fire right now, it may be a year until I start taking measurements with it and was considering buying a couple of others before they are gone, but not sure this lens will have decent enough divergence to keep it down around 1.5 mRad or lower, thus the question. If they have significantly higher divergence due to the extremely short FL placement from the diode, as a general rule, then I won't bother buying more. Anyone with experience with such tiny short FL lenses?

(PS: I already tried to collimate the output of a laser diode with it earlier, but the heat sink the diode is firmly mounted in prevented me from getting right on top of the diode, need to get a different HS to hold the diode to do that).

Edit: Here's a diode I might use with this lens, if it truly is single mode at 1 watt out, I haven't seen any single mode diodes at any wavelength at this high of a power output, anyone think it could really be single mode?

http://www.aliexpress.com/item/2014-...027333732.html

[planters]

Assuming that the lens is an asphere (at this low an F ratio it is hard to imagine that it is not), this lens should work with just about any diode for which its coating will transmit. Look at the reflection of a white light from this lens. A blue color will tend to confirm the listed coating range.. There is the possibility that with a tall, canned diode that the lens will not be able to sit close enough to the junction, but this will probably not be the rule and certainly will not be a problem with bare diodes.

It doesn't matter if the diode is single mode or multi mode. The collimated beam from this lens will be very small, but the divergence will be high. There is no magic, however. If the beam is 1/2 as big then the divergence will be 2x greater. Who is telling you the opposite of that statement? One hassle with the short FL will be the required precision in your alignment to avoid astigmatism and comma. But, we use 2mm FL lenses with great success and so, this isn't tremendously different.

Now... I agree with Andy. I doubt that you have more irons in the fire than I do. Maybe I am wrong, but if I went ahead and purchased one of these lenses I would follow through with the time it would take to put this lens in front of a red diode (at low powers the diode could be glued over a hole in an aluminum plate). The lens is already installed in a small brass or copper block (it was probably intended for an 808nm diode with some power) and so it too could be attached to a metal block and moved, even by hand, to allow you to get some measurements.

I'm am not suggesting you do this just to make work. There are subtle aberrations in the beams from these diodes that could benefit from counteracting aberrations from the collimating lenses. The company that makes that lens is a good company and these lenses will be available in the future. If your results proved this is a successful application, we could all benefit from this.

I don't mind you asking a lot of questions and I'm happy to try to answer them, but now you should start trying to discover more and pass it on.

[Laser57]

All I have for a laser diode mountings are the copper versions of an Aixiz module DTR sells, but the little copper square this lens is mounted in is wider the the round rim of my Aixiz module and all of my diodes are already firmly mounted in them. I have to buy another laser diode or get an extractor and remove the diode, or perhaps just cut into the copper with a hack saw and then file the edges down so I can have a way to place this copper holder down on the diode.

The reason I was questioning whether the diameter of this ultra small lens would cause the divergence to increase, was because I was told by a well known seller of laser electronics that a large amount of expansion of the raw output of a laser diode prior to collimation reduces the divergence, due to the huge ratio between the diode chip output aperture size and the collimation lens diameter. Thus, my belief that using a lens with a 1.4mm focal length would naturally cause higher divergence, due to the reduction of beam spread straight out of the diode prior to collimation from being so close, as well as the smaller diameter of the beam output of such a small lens. All I was asking is if I am correct in such a belief.

For lowest divergence, wanting to use a much larger lens for another project, believing I could just let the output of the diode expand to two inches prior to collimation with a single lens to collimate the beam, when asking if this would work, someone here was telling me I should first collimate the beam close to the diode, then use a beam expander for best results. I believe his main concern using the method I preferred to reduce the number of lenses, was because this method could possibly exceed the diffraction limit of my collimation PCX lens (I didn't tell him my lens would be 50mm diameter). Is the whole idea that letting the output of a laser diode expand prior to collimation to reduce divergence wrong? I don't know enough about optics to know where the diffraction limit is reached with a 2 inch dia. lens between 445-1064nm, or if the limit can be reached with a lens that big between those wavelengths.

[planters]

You do not need a laser diode mount in order to test this lens. I would not hack the the module you have. Leave it alone and buy a very low cost laser diode. DTR may have some of the older, lower cost, lower power diodes that are not worth cluttering up his site, but in any case if you have a diode to work with there are so many systems, and optics that you will be able to test you will find it worth the investment.

I would like to know the name of that seller, however if you don't want to post it, please PM me. He is not wrong, however, with very high divergence, you will not be able to place the collimator very far away before it will have to be enormous to avoid vignetting the beam. In some applications with the high divergence Mitsu diodes, even after collimating, the beam will diverge sufficiently to allow you to skip the first lens of a beam expansion telescope because the beam has expanded enough on its own already! So, to answer your question, yes you are correct. But, because we routinely expand the collimated beam with further optics, the short FL of your lens will only change the required magnification in these later optics. The magnification will have to become greater.

The diffraction limit will not come into play with this optical application. And, aside from my comments in the earlier post about subtle effects of lens choice for reducing residual aberrations, there are many ways to accomplish what you want. If, as I guess, you want the lowest divergence then as Steve said get a good quality beam to start with. If you want a diode laser get one of DTR's single mode diodes. Drive it at low power. This favors the single mode and reduces spurious modes that are not as favorable, but given enough pumping power can begin to emit. This also reduces thermal effects in the junction. Now, use an aspheric collimator with a 4-10mm FL because this will minimize the alignment demands by at least the square of the FL. Accurate alignment and spacing is frequently overlooked when the merits of these various lenses is considered. Finally, beam expand to the limit of your PCX lens' clear aperture. If you also wanted the cleanest far field spot as well as the lowest divergence then use a pair of PCX lenses for this final expansion stage and place a spatial filter at the focus formed by the first lens in the pair.

Now, if you would please ask Steve to send me some of that $25 that would be OK and meanwhile, you need to get to work. Lets see some data.

[Laser57]

Planters, I have 12 of each color R,G,B single mode laser diodes mounted in copper modules I bought from DTR I was intending on combining the power together with using a knife edge and TTL driving their CC regulators to produce different perceived colors with but then I started to think about combining the powers together using a long cylindrical line lens to produce a high powered R,G,B line output, so have plenty of those diodes to experiment with. Thank you for your advise to run them at low power, I was not aware that driving a 50mw 520nm single mode diode to higher than the rated output would cause diode to produce a spurious output or depart from TEM00, if I understand properly. The line lens R,G,B project is just one of three I want to build and it's just for fun.



Another project I am planning, trying to get someone to build a host for it now, if he can get the time to start on it, is using a single mode diode with a 5um aperture output chip I bought from Junktronix, I wanted to first partially collimate the output using a G2 lens to reduce the beam expansion of the diode to match the clear aperture of a 2 inch diameter PCX lens placed about 4 inches away at its focal length in a sliding tube. My thought was with a thumb wheel on the G2 lens to adjust its focus to vary the diameter of the spot put on the PCX lens, and then using a sliding tube with the PCX lens on the end to move its focus in and out to fully collimate the beam the rest of the way, I could build a host with a variable beam width output. Perhaps this might be a good method to produce a low divergence pointer?



I was working on combing the output of three diodes, R,G,B to produce a beam which would turn white after each of the individually collimated outputs diverged into one another miles away, and the host was built and it puts out three single mode beams of each color, but I can't see a white spot when on a cloud because the alignments of the individual beams are fixed, I need to have a host with three individual swiveling heads built for that, it seems, or some way to align the beams. I knew that might be a problem, and it was. I tried to then let the raw outputs of all three beams combine together into a single lens by expanding their outputs together to overlap the three beams, but they still won't come together, they just depart out of my PCX lens in three different beams. My next thought was to mount the individual diodes close together in a single head, like 12 of them or more, up to 21 of multiple colors and then try to collimate them together after letting their outputs expand quite a bit first, so they would overlap onto a single PCX lens but the cost of having that built is a bit high for my budget right now after Sinner quoted me 1800 dollars to build a pointer like that for me and that's only the metal, no optics or diodes. Maybe that is another flawed idea, when I ask around on LPF about such ideas I don't get any one telling me if it is another unworkable idea or not, so I'm still considering it.



Above: Terribly rough sketch of multiple diodes mounted closely together in a pointer host I was wanting to have machined and the hole pattern I had decided upon to mount the individual diodes into.



Above, photo of tri-color pointer I had made.

So, that's what I've been working on building, I'd be happy to pay someone to coach me through some of these ideas, I am sure it would save me far more than I spend for help. I understood what you last wrote, after a bit of googling for some points, but I don't know if I fully understand the mention you made of using two PCX lenses to collimate the output, why two? Won't one fully collimate the output? If you have two, does that mean the first one partially collimates the output while the second then completes full collimation and through using two you can have a lower divergence beam? Are the two PCX lenses you mentioned different diameters?

I know the direction I am going, and that's to projectors. For a start, I bought this LED projector to play around with. I want to replace the LED's in this ADJ Inno Pocket Scanner with a expanded 532nm laser to make a simple sweeping projector, nothing fancy but will get my feet wet:



I want to build my own huge scanner mirror (albeit slow) using CNC dual shaft stepper motors for use with a six inch diameter Special Optics beam expander I bought on ebay, it is just too big and heavy to use that way so I need to use it with a scanning mirror. The ADJ Inno scanner will help me learn how to build a simple X-Y scanner. To manually control the beam pointing I am searching for a stick X-Y controller for stepper motors, but can't find one for sale anywhere:



There's all my irons in the fire for this year. Thank you for your kind help and time with these questions. If anyone sees an obvious "that won't work you dummy" in any of this, please let me know.

[planters]

Looking at your projects, it seems that you should have plenty of raw material to perform the test on your small lens. Let us know what you get for beam size and far field divergence.

Your tube idea will probably work, but keep in mind that if you use the aspherical collimator well away from its designed FL you will introduce spherical aberration. This may or may not be compensated by the PCX lens. This will need some experimentation. The mechanical precision as well as the stability of your set up will be challenging.

Why not combine your RGB diodes from the start with a dichroic filter just like it is done in a projector?