Lens vs. Prisim

[Solarfire]

Regarding O-like divergence. I also am getting in the neighborhood of 1.8 mrad raw divergence (this is typical of other posts) which only reduces to 1.5mrad with a 1:1.5 expanding telescope. And this is at the cost of a near doubling of the slow divergence due to the aberrations I already discussed. I am currently using 24 of these lenses and their variation in performance has been small. Is it possible, as I have done way too many times, that your math is in error?

The O-Like lens prism corrected can give a divergence <0.7mrad (see this post http://www.photonlexicon.com/forums/...ion-correction). If you're going to use telescopic lenses then it should be a combination of Plano concave/convex cylinders as offered by drlava but here a shorter FL collimator of about 4mm should be used to get best results. I prefer the combination of the 405-G-2 lens with drlavas cylinders. Gives a nice 3.5mm beam @ the aperture also with a divergence of <0.7mrad. The far field beam of the corrected O-Like @ 15m is almost round. The 405-G-2 cylinder corrected gives a line that can be rounded up by defocusing the collimator slightly.

Note that both beams in the pictures below have the same divergenc, the O-Like is somewhat larger in the vertical due to the longer FL of the collimator the beam @ the aperture is 4.5mm on the fast axis.

[danielbriggs]

How are you getting 1.1 mRad from an O-like without correction?

Initially it was in the order of 1.5-1.6mRad; but a lot of tinkering and adjustment of the lens later; the best I got was 1.1mRad.
Arithmatic is correct. I will however re-check the units at some point; unfortunately the next two weeks is virtually out, as I'm away.

Cheers,
Dan

[planters]

Solarfire, I read your original post and the updates and it has been very useful. I am assuming you have not looked at applying these optics to more than one diode beam at a time The complexity seems to arise when multiple beams are passed through these optics. Logsquared brought up a good point about the FL of the telescope optics. Based on my memory of astronomical optics I believe that within reasonable limits the larger the focal ratio of the positive/negative pair the smaller the off axis aberrations as well as the relative relaxation of alignment tolerances. The effects are proportional to the FL squared. For example, a 2:1 telescope with a 20mm and 10mm pair should have four times the off axis aberrations and alignment requirements as a 40mm and 20mm pair. I believe refocusing should also be necessary as the beamlet entering the telescope on axis, at the center of the stack, was originally focused to the same far field point as was the beamlet at the edge of the stack, but the off axis beamlet passes through a more powerful region of the telescope and will likely benefit from a readjustment of its focus.

[Solarfire]

Yes I did do some multi diode setups, not as many as you are trying to combine, (max 4 x 445nm knife edged) with cylinder and prism corrected versions. The main thing was the initial beam alignment (without correction), getting the beams running parallel, meaning if the beams are 3mm apart center to center @ the aperture then they need to also be 3mm apart center to center in the far field, not diverging or converging. Second but not less important is to stay well within the clear aperture of the lens being used. To be really correct, a set of aspheric cylinders (custom lens) would need to be used but a set of plano convex/concave delivered good results. Prisms are less of a problem and more forgiving with slight misalignments.
Last but not least, it helps to select diodes for a clean and symmetrical raw beam profile, especially when it comes to 445nm. Some are only suited as trashcan illumination and tend to produce splash without end.

[CDBEAM]

Yikes...The above post begs the question..." To what lengths do we go to to acheive the best parallel beam propagation ?? ".... Yoda would state the goal as..." Neither diverging or converging must you be " hahaha.... I have envisioned the use of two cell phones being used...one at the adjustment area....and...one at the farfield plane...say 30 meters away where a beam termination would be conviently located....then again.....it would require alot of over and under adjustments with the 1.5mm driver to find the perfect spot. I am sure Terrawatt would find the scene....amusing !!! It NEVER ends !!

CDBEAM

[Solarfire]

Yikes...The above post begs the question..." To what lengths do we go to to acheive the best parallel beam propagation ?? ".... Yoda would state the goal as..." Neither diverging or converging must you be " hahaha.... I have envisioned the use of two cell phones being used...one at the adjustment area....and...one at the farfield plane...say 30 meters away where a beam termination would be conviently located....then again.....it would require alot of over and under adjustments with the 1.5mm driver to find the perfect spot. I am sure Terrawatt would find the scene....amusing !!! It NEVER ends !!

CDBEAM

Of course the beams are going to more or less diverge/converge to another; the point is to minimize this. I use a 10cm x 10cm first surface mirror @ 15m to fold back the beam to the work area onto a measuring grid, giving me a distance of 30m to work with. After this initial alignment I take on the beam correction with either prisms or cylinders.

[planters]

I perform alignment the same way as solarfire and it works well. With our remote head projector (one floor up) I love using a high quality web cam on a 20M USB 2.0 connector. This has saved a lot of yelling. A set of aspheric cylinders might be better, but I have no idea where to begin with a description of the prescription and suspect going to long FL correctors ie 100mm-150mm may be nearly as good. Selecting the better quality diodes is a good idea and I'm going to look at this more carefully. There always are a few diodes that have larger more irregular spots at focus.

[CDBEAM]

Thanks for outlining the folding technique...I WILL use that. I did not mean to suggest that striving for perfection of alignment is not a constant and admirable goal. I believe that one should always shoot for perfection....even knowing that at some level of inspection...minor misalignment issues will present themselves. Engineering is always a compromise
of solutions. I was somewhat amused at our mutual desire to atain... as close to as perfect optical perfection. I wonder what percentage of the planet has ever given such thoughts towards this goal of perfection ????

CDBEAM

[planters]

At least half a billion Buddhists?

[CDBEAM]

Ah Res Grasshopper...I meant the Holy Grail of Optical Perfection.

CDBEAM