Galvo/Amp Upgrade

[jdavis7765]

A Keplerian collimator can be used to expand or contract the beam diameter, yes. But in this application the idea is to use identical lenses on either end of the collimator so the output beam is identical to the input beam. The idea is not to change the beam diameter but to position a spacial filter (the pinhole) at the focal point in the middle between the two lenses. This allows you to "carve out" the center of the beam. And since the beam is converging to a point as it passes through the pinhole, any stray light that diffracts around the edge of the pinhole will spread out far enough that it will completely miss the second lens and thus will not be present in the output beam.

So, if I am reading this right and understand correctly, there will be two identical lenses and the idea is to first make the beam diameter smaller with the first lens? Then you would carve out the beam with the "pinhole" and then the other lens is put at the other end after the pinhole to upconvert essentially, back to the original beam only with the pinhole having corrected the beam into a perfect circle? Ideally, would i do this for each individual laser or after the lasers have been combined?

Most commercial RGB modules list the beam diameter and divergence, so it's easy to make sure that whatever you buy will work with your scanners. Be sure to measure the width of your mirrors first though, because there is a big difference between a 4 mm wide mirror and a 4 mm *aperture* mirror. (Because the mirror is positioned at an angle to the beam the mirror must be wider than the beam to capture the whole beam.)

So, I have the compact 506's 4mm. Say I were to go the route of buying collimated RGB laser module such as this for instance. Would it work? I am not sure I understand what it means showing the different beam sizes. Could that be for the distance from the laser?

Or, just for reference, this laser module am I looking at the 5.6mm as the beam size?

Thanks again!

[buffo]

So, if I am reading this right and understand correctly, there will be two identical lenses and the idea is to first make the beam diameter smaller with the first lens?

The first lens in a Keplerian collimator converges the beam (focuses it to a point), but beyond that focal point the beam begins to expand again. (The beam is now flipped in orientation, but that's not important.) When the beam has expanded back to it's original size, that's where you should position the second lens. That second lens re-collimates the beam so it's parallel again.

Thus you have a 1:1 collimator. It doesn't really do anything to the beam (apart from flipping it, which is only relevant if we were talking about an imaging system). The key is the focal point in the middle, as that is where we will position the pinhole spacial filter.

Then you would carve out the beam with the "pinhole" and then the other lens is put at the other end after the pinhole to upconvert essentially, back to the original beam only with the pinhole having corrected the beam into a perfect circle?

Yup! Exactly right.

Ideally, would i do this for each individual laser or after the lasers have been combined?

After all beams have been combined, but before the scanners.

I have the compact 506's 4mm. Say I were to go the route of buying collimated RGB laser module such as this for instance. Would it work?

The goldenstar module you linked to lists the beam width at the laser aperture as 2 mm by 5 mm. The 5 mm axis is too wide for your scanners. You can probably clean that up with a spacial filter (as described above), or just accept that you'll have a small amount of the beam spilling off the scanner mirrors when you scan wider than 45 degrees.

I am not sure I understand what it means showing the different beam sizes. Could that be for the distance from the laser?

Yes, the "at aperture" measurement is beam width and height at the output of the laser module. The rest of the measurements are at 5 meters away, 10 meters away, and 15 meters away from the output window of the laser module. The beam gets bigger due to divergence. Also, there are two beam width measurements because the module uses multi-mode diodes with rectangular beam profiles. (See fast and slow axis explanation above.)

Or, just for reference, this laser module am I looking at the 5.6mm as the beam size?

No. On that website they are talking about the physical size of the diode module itself. 5.6 mm is a common diode can size. 9 mm is another common diode can size. Neither of these two measurements have anything to do with the size of the output beam.

When buying a bare diode you need to know the size of the diode can (basically the diameter of the can) so you can purchase the same size diode module to mount the diode in. This often gets confusing, because the module will have two sizes listed: One is for the diode can diameter and the other is for the total outside diameter of the module. Typical outside diameters for modules on DTR's site are 12 mm, 20 mm, and 25 mm.

Adam

[jdavis7765]

So I purchased the lenses from edmunds and will be giving this a try. I also found these lens holders for 3d printing. Not sure how well they will work or even if the optic will fit but I am sure I can rig something up or also take up CAD as another hobby haha. I will follow up here when they arrive!

Thanks for the help, not sure I would be anywhere with this if it wasn't for this forum!

[buffo]

Wow! Those 3D printed lens holders look great! If you can make those you'll have no trouble building the collimator.

Adam

[jdavis7765]

Wow! Those 3D printed lens holders look great! If you can make those you'll have no trouble building the collimator.

Adam

Printing one now to see how it looks. Looks like you can even customize for different lens sizes. Also, looks like Edmunds is in NJ and I am in PA. Ordered the optics yesterday and they will be here today. That was fast!

[jdavis7765]

So, I have a question about this. First things first, lenses have arrived and man they are smaller than I thought (damnit to America for not using metric system). Anyway, the lenses are flat on one side and convex on the other. Does the laser go into the flat side of first lens, then out the convex side moving toward the second lens which would be into the flat side then out the convex side?

Also, do the lenses have to be exactly oriented (not sure of the correct way to say this) so the laser enters the exact center or is there a little wiggle room?

[buffo]

Does the laser go into the flat side of first lens, then out the convex side moving toward the second lens which would be into the flat side then out the convex side?

Nope - exactly the opposite. Put the flat faces towards the middle of the collimator (towards the focal point in the center). The convex faces should be facing the outside. (It *will* work the other way, but you'll get more reflection off that flat surface if it's facing outward, and you don't want that.)

do the lenses have to be exactly oriented (not sure of the correct way to say this) so the laser enters the exact center or is there a little wiggle room?

It depends on the diameter of the lenses compared to the diameter of the beam. You mentioned that the lenses were very small though. This suggests that you will not have a lot of "wiggle room". Try to get the beam to hit the dead center of the lens. If you are off, you'll notice that the output beam will not have the same shape as the input beam.

Adam

[jdavis7765]

Nope - exactly the opposite. Put the flat faces towards the middle of the collimator (towards the focal point in the center). The convex faces should be facing the outside. (It *will* work the other way, but you'll get more reflection off that flat surface if it's facing outward, and you don't want that.)



It depends on the diameter of the lenses compared to the diameter of the beam. You mentioned that the lenses were very small though. This suggests that you will not have a lot of "wiggle room". Try to get the beam to hit the dead center of the lens. If you are off, you'll notice that the output beam will not have the same shape as the input beam.

Adam

That makes sense! Is the idea here that you could hit the optic (not exactly straight on) and it will still be able to shape the beam out the flat side a little better?

I got the optics holders printed and designed my own rail (its not the greatest) to keep then more aligned. I need to figure out how to adjust small increments a little better though. A lot more difficult than I imagined it to be, but here I am lol. Ill post back when (if) I get it setup. lol

Thanks again,

Jeremy

[kecked]

Believe it or not but even with 500mw laser you can punch your own pinholes in aluminum foil. I’ve also used guitar picks. Remember the best place to put the pinhole is slightly forward or back of the focus. I tend to burn the hole then move the first lens forward a tiny.

ive also done it after the dichros combine the beams. Just need to use the right achromatic lens set. One pinhole.

use a worm gear to move the lens. I’ve used threaded rod too. Those cnc rail mounts are cheap on eBay. I get the 12mm ones. Use them for your diodes too and consistent beam height.

[buffo]

Is the idea here that you could hit the optic (not exactly straight on) and it will still be able to shape the beam out the flat side a little better?

No, the lens will perform equally well in either orientation. But having the incident beam hitting the flat surface directly parallel to it is the ideal case for back-reflection, which can harm the laser diode source if the anti-reflection coating on the lens is not up to the challenge. So you want to avoid that if at all possible. (Admittedly this is more of a risk with single-mode diodes though; most modern multi-mode diodes are remarkably robust by comparison, even if their beam quality is nowhere near as good.)

I got the optics holders printed and designed my own rail (its not the greatest) to keep then more aligned.

I admire your dedication to the do-it-yourself ethic, but if you get really stuck you might want to consider using one of the commercial mounts that Mark suggested above. Worst case, maybe you use one of your mounts for the fixed lens and then a commercial one with fine adjustments to tweak the final alignment.

Believe it or not but even with 500mw laser you can punch your own pinholes in aluminum foil.

I hadn't considered using aluminum foil, but this is a great suggestion! The beam artifacts at the periphery of the beam will be much lower in power than the center of the beam, so the aluminum foil will easily mask that while allowing the main power in the center to punch a nice clean hole.

the best place to put the pinhole is slightly forward or back of the focus.

I've seen people start with a pinhole that is much wider than the beam waist at the focal point and then move the pinhole closer and closer to the first (incident) lens until the pinhole begins "carving out" the outside edges of the beam. When you get an acceptable output beam, that's where you lock the pinhole in position...

use a worm gear to move the lens. I’ve used threaded rod too. Those cnc rail mounts are cheap on eBay.

I bought a set of linear translation optical stages off E-bay years ago for exactly this sort of project. But I ended up paying quite a bit for those units, even though they were used and the price was marked down a lot. (I think I paid around $150 each? Retail was over $400.) So yeah, if you have a source for a linear translation stage that is more affordable, please share!

Adam