1.4W 462nm Laser Diode

[steve-o]

I love sausages

me too. I had one for breakfast this morning with 2 eggs. Yum.

[laserist]

There was a study that found the people in the lowest 1/4 in terms of humor thought they were so very funny. So in a sense they were so very funny...

[danielbriggs]

how are you implementing the extreme precision that is needed in adjusting the beam optics and diode positioning

I have not run into any issues to date (I've setup 120 diodes in the last two days using this method) however I am not using 2mm FL lenses in which the alignment would be more sensitive.
Actual alignment is made using a modified tool I milled with castellations to lock onto the lens to provide unobstructed, fine positioning.
The interference patterns reveal a lot about where the lens is qualitatively w.r.t. the waveguide. Perfectly straight fringes indicate no coma. When slight (I have not measured qualitatively) off axis alignments are present distinct "S" and "C" fringes are created. The lens is first adjusted in "Z" axis for parallel fringes, and then if necessary, in X/Y to minimise coma and spherical aberrations. The lenses are then permanently secured.

Good images of expected interference patterns:
http://jkerwin.users3.50megs.com

However, if you can accurately measure the beam profile of a few of the more common multimode diodes then generate an inverse optical element (in software), a corrector could be fabricated (refractive or deffractive because these are monochromatic light sources) to improve these beams in a way that rivals single mode diodes.

Which diodes specifically? This can be measured with ease. My tools go to about 4 micron accuracy.
I mainly (almost exclusively) use single mode diodes, as although the power is lower, the beam quality and apparent brightness is so much greater in compact uses.

[planters]

Interesting. You are adjusting the lens relative to the diode and not the reverse. With the simple, threaded barrel mounts that I am using for the lenses this is not possible and the diode has to be adjusted in its mount.

The two diodes that come to mind are the P73 and the 9mm Nichia 445nm blues. The red is the bugger as the blue, even with its multimode stripes can be packed into a 1/4 mrad with a 5mm aperture at the scanner. Nevertheless, the blue and the 520nm greens will probably have similar, if not identical aberration.

This area of aberration correction is important to me. I have gone about as far as I can with low order optics even though, unlike you, my approach has been empirical. I doubt I am going to see a significant improvement without higher order correction. The difference between a single mode diode (even aligned so-so) and a multimode diode is substantial.

[danielbriggs]

The two diodes that come to mind are the P73 and the 9mm Nichia 445nm blues. The red is the bugger as the blue, even with its multimode stripes can be packed into a 1/4 mrad with a 5mm aperture at the scanner. Nevertheless, the blue and the 520nm greens will probably have similar, if not identical aberration.

This area of aberration correction is important to me. I have gone about as far as I can with low order optics even though, unlike you, my approach has been empirical. I doubt I am going to see a significant improvement without higher order correction. The difference between a single mode diode (even aligned so-so) and a multimode diode is substantial.

Sure; I can let you know about the 9mm 445nm's - there is a tray of 20 or so which need correcting on my to-do list.
The only P73 I currently have is non-operational - it was butchered to inspect the die at the micron level.
Using the interferometer to provide qualitative collimation and spherical aberration information does require a coherence length on the order of several mm. All the diodes I've used (even the multi-mode 520nm's) satisfy this condition, at least at low injection currents.

As for the long term goal you may have... MMvsSM... disclaimer: No free lunch, physics limits are present.

[Stiffler]

My good fellows, do any of you care to translate Mr. Planters & Mr. Briggs conversation for me?

[planters]

This is not at all complicated, but the jargon can make it seem like its more sophisticated than it is.

With single mode diodes, the beam quality can be extremely high with very low divergence even with narrow beams. The limitation is usually not the optical quality of the small, often molded optics (1/100 wave is not uncommon), but rather the positioning of the lenses to minimize off axis or misalignment aberrations like coma and astigmatism. Or, sometimes the optical elements will introduce a significant amount of spherical aberration (like when a plane window is placed in a converging or diverging beam). An interferometer can be set up to allow one part of a beam to interfere with another part ( the details CAN get a little heavy)) and the shape of the resulting interferences can be used to measure or to null (minimize) these aberrations. To date, I and most others have been empirically adjusting and substituting optics to minimize far field spot size. Daniel's interference technique is a significant improvement. It is the way most high end optical assemblies are dialed in, commercially.

Multi-mode diodes have a beam that is much more complicated than the round, Gaussian spread of the single mode diode. Even perfectly aligned simple optics will not be able to correct aberrations that are not continuous across the beam. Although not the best example, imagine how a simple lens would deal with the parallel stripes of the multimode P-73. It can't. These do not result in a mush of blurred light in the far field. They focus just as sharply and are as well defined as the main stripe (these are small coupled lasers in their own right residing within the cavity of the diode, but unaligned with the main stripe). Fabricating a reciprocal lens element that introduces negative aberration, equal, but opposite to the existing aberrations will improve the beam quality.

Daniel is right there is no free lunch, but my experiments with spatial filters has shown me that there are residual errors present in a well corrected P-73 beam. For example, the focus at the filter is not universal. Some of the aberrations focus at a position different than other aberrations.

My search for a corrector requires a map of the existing aberrations and would have to be placed at a specific location such as a replacement for the spatial filter. It would not be cheap and would only be worth it if fabricated in quantity, but the improvement could be very significant.

[Stiffler]

Thank you Sir Planters!

[kecked]

any chance such a lens could be cut from a laser interference pattern itself and then used to make a lens by pouring a liquid plastic and curing it in the mold? Would not work for high power but should tolerate a watt or two. Maybe use it to make a negative by curing the plastic by depth on power and then making the inverse?
just wondering if there is some way maybe even 3d printing based on a profile? There has to be a hobby way to do it.

[planters]

I think you need to inset a computer in this. Analyze the interference and then produce a computer generated inverse hologram. I think that the need to render an inverse would make an all optical solution difficult to specify. How do you establish the null? But, I do think a holographic route may be the least costly for small numbers. Maybe it is time to visit with the holographer's page? Printing, mold making or contracting for a shaped lens might be tough.