What is the best way to align beams with this knife edge?

[Alaskan]

Can someone please help me with my knife edge project? I have a large number of single mode R,G,B laser diodes in 12mm modules I want to combine using three of these knife edge modules, one for each color. I've read that aligning the beams can be a difficult job, depending upon how you set things up. What is the best way to mount my 12mm aixiz modules for easier yet precise alignment? Should I even use these modules, taking the diodes out and using something else? I still want to use my G2 type lenses, though.

Much appreciate if someone can help, or several chime in with ideas. Maybe I shouldn't even use this type of knife edge? This is for a hand held laser pointer project.

Each of the knife edge modules, as shown in the above photo, has its own PCX lens on the output.

Thanks!

 

[Bionic-Badger]

I wouldn't do the adjustment with your diode mounts, but rather just ensure that the diodes are mounted in a solid manner to the plate you're using. Try to get them as level as you can, possibly at the height you need the beam, but rely on mirror mounts to do the real alignment and beam deflection. I haven't gotten something set up yet on a plate, but I've got some 12mm CNC shaft supports and they seem to do a good job for bulk Aixiz module mounting and for heatsinking. There are other mounts you can get on eBay or through members here on the forum.

For the mirror mounts, these ones are pretty cheap and adequate.

If you need something smaller, for handheld, I'd get your diodes set in the host, and then see what you can do with mirrors held in place in the host. I also wouldn't rely on that projector's knife edge. That was designed for creating a big wall of light rather than something resembling a beam.

 

[Alaskan]

OK, with that advice, this means as long as the diodes are level, you don't need to adjust them, the mirrors can do all of the necessary adjustments of the beam. That would make things much easier, I've seen mirrors which are made from prisms, FS mirrors for all of the visible spectrum and special mirrors for each wavelength. I don't know which to use yet, the advantages or disadvantages of a prism compared to FS mirrors? Can you suggest for that too?

Thanks

 

[Cyparagon]

I've been tinkering with that projector's blue "light source", and I'm not sure if it is poorly aligned diode mounting or mirror mounting, or both, but the 24 beams are not very parallel at all. There's no easy way to align those mirrors, so you're out of luck unless you can spatially position your diodes on 2 axis as well as aim them on 2 axis. (That's very difficult with large numbers of diodes)

In short, you've chosen about the worst tool for knife edging. It's possible, but a colossal pain, and you'll end up sinking more money into the diode mounts than you saved on using those mirrors.

rely on mirror mounts to do the real alignment and beam deflection.

The only problem with this suggestion is that it makes the mirror array redundant. If you're going to use mirror mounts (which most people will agree is the proper way to knife edge), you might as well get rid of that extra mirror array.

I don't know which to use yet, the advantages or disadvantages of a prism compared to FS mirrors?

the "prisms" you're talking about are not used as prisms. They simply provide a 45 degree angle for the mirror surface. Some people find these easier to work with, but the only difference is the angle they mount at.

 

[Bionic-Badger]

I definitely agree with that mirror array being redundant. It's interesting to look into, but was never meant to be a precise piece of optics in the first place. Given the number of blue lasers, some misaligned diodes probably made little difference. I thought that the heatsink with the diode mounts could be useful though, maybe with some external lenses held in place in front.

Laserproject: you might want to use both mirrors polarized beam cubes for your beam. If you ensure that one set of lasers is polarized in one direction, and another in the other direction, you can beam combine two sets of lasers into one beam with a beam cube. Use front-surface mirrors on mounts to do the actual knife-edging. Note that if you try doing color beam combining using an X-cube, they use polarization as well, so you can't use the polarized beam cube method above.

 

[Alaskan]

With the last kind recommendations, I've decided against using that knife edge assembly and am now thinking of using individual mirror mounts to combine the outputs of several diodes (of each wavelength/color) together into their own streams and then combine all three wavelengths together with dicro's, or a cube.

I'm a bit confused though, it appears I am also being advised to have the diodes adjustable in at least two axis too, in addition to individual mirrors for each diode, is that right? I'd rather spend more money and make this easier to do than have a difficult alignment ahead of me.

Your help much appreciated, thanks Bionic Badger and Cyparagon.

 

[starlight]

Hey laser project, where did you acquire the knife edges? I was gonna buy some from DTR but he no longer has them on his site.

 

[Alaskan]

I bought them from DTR early in the year, PM me your address I will send you one. Alternatively, 15 bucks on ebay.

 

[starlight]

I bought them from DTR early in the year, PM me your address I will send you one. Alternatively, 15 bucks on ebay.

PM sent.

Haven't seen any on eBay but I remember DTR was sending as many as he could fit in a box to get rid of them. I should have grabbed some when he had to many

 

[CDBEAM777]

MMmmmm ??? LP....If you have R,G and B....Why knife edge ??? Use Dichroic Mirrors to combine into one beam !!!

Unless you want a Red beam stacked next to a Green beam stacked next to a Blue beam...lika layer cake ???? If THAT is what you want...go with a knife edge arrangement...with the modules mounted parallel to the optical platform to achieve horizontal alignment (Pitch) and your left/right alignment (Yaw) accomplished with the knife edge mirror adjustments.

I should note that the knife edge mirrors will also provide pitch adjustment....and you will need some minor adjustability there also...As far as a prism mirror vs. a flat mirror....I use the prism mirrors....and each prism mirror is specifically coated for a color: R,G and B.

The reason for prism mirrors selection is that the actual edge of the prism is much smaller than a dielectric mirror square. This means that one can position the beams closer together with prism's !! Closer together delivers improved beam density, which in turn provides greater luminosity !! That is what you want !!

Mirror prism's are more expensive....by a little....and as I said....you will need a specific prism mirror for whatever color you want to knife edge !! The reflective coatings are optimized for Red,Green or Blue !! Lasertrac has really good ones !!! See http://stores.ebay.com/lasertack

But...But......if you want ALL three beams centered on the same axis....then go with Dichroic mirrors. Do a search here or go to the " other " site for laser projectors....there is a S*** boat of info on this !!! The attached pic shows beam combining. The first 90 degree bounce is done with a plain dielectric coated mirror. It is not necessary to set things up this way ! The first laser does not need to be bounced ! It could be positioned straight, and the remaining two (2) laser do demand the 90 degree bounce. Simply, the Green goes thru the second mirror and that mirror reflects Red....the third mirror passes the Green+Red beam, and reflects the Blue beam. In the end, all three beams , G+R+B...are combined into one beam.

Depends on what ya want ????? A stacked RGB beam would be....interesting !!! I do not think that has been done...IIRC ???? See pics !!!

 

[Alaskan]

CDBeam, great!

With ten or more diodes of R,G,B for a total of over 30 diodes, my understanding, which is still coming together on this, was that combining them together first before the dichos was necessary, otherwise if I used dicros for each diode, I loose more power. Is that right? Maybe it is moot anyway, using mirrors one way or another regardless of how it is done with the losses adding up anyway.

I wanted to combine all of the beams together to make any color I want with a controller, but stacking does sound interesting and perhaps that is an option if it gives some kind of advantage. I have all the drivers now, all the S1 lenses, diodes and 16 sets of dicro mirrors, no mirror mounts yet though. I am interested in peak intensity, so the prisms are on my list now.

I am still unclear regarding the diode mounts, should I have diode mounts which allow movement in at least two axis?

Thank you for your help! A+

 

[CDBEAM777]

Well.....I missed that you had thirty diodes coming to the party !!!.....SO.....10 of each color.....knife edge 10 in a group....then dichro the R+B+G ....in a handheld !!! Maybe have to settle for 8 x 3 = 24

In a handheld ??? Well....OK.....must have some " BIG HANDS " in Alaska !!!

I would use the mount kits from either LSP or Lasertrack....See

http://lasershowparts.com/store/index.php?_a=viewCat&catId=84

http://www.ebay.com/itm/Laser-Diode...867?pt=LH_DefaultDomain_0&hash=item20d5de306b

Or......Machine an Aluminum bar to accept 4 or 5 modules. While this would be less cost....it does introduce an additional interface between components....so not as thermally efficient to pull heat away from your diodes.

With the Ally bar +Axixz Module + Diode.....you have two (2) interfaces.....with LSP/Lasertrack LD mount + Diode....you have one (1) interface. In this case...one is preferred over two !! With each interface....even with Artic paste applied to the interface....there is an additional thermal barrier to lower the overall thermal transfer. The goal is to always pull as much heat away from the diode, as quickly as possible. Less interfaces means more efficiency in pulling heat from the LD case !! This is especially applicable if any diode ....er....ah....may happen to be....overdriven !!

Lotsa ways to do things !! Always trade-offs !!!!

SO.... group 4 together....then PBS two groups of 4...so 8 in each group.....then Dichro them 8+8+8....24 modules total !!

I can think of no more compact way to arrange all these modules !!!

Such a construct will be very....er...ah....challenging to align !!!

Perhaps, each color could have its own optical plane.....so....all of one color..
knife edged and PBS's on a single plate of Aluminum, then arrange the plates...one atop the other....and combine the beams...vertically....going from level to level to level using the dichroic mirrors !! Then a final bounce mirror to turn the beam back to parallel with the whole plate+plate+plate sandwitch !!

I dunno.....you got 20 lbs o'shit inna 10 bag.....I would call this build...." BIG FOOT gets a Handheld ".....he could handle it !! hahaha Good luck !!:eg::eg::eg:

Final note: Keep the Modules/Diodes stationary and just move the optics/mirrors....just make sure all the beam sources are at the same elevation !!!!!

CDBeam, great!

With ten or more diodes of R,G,B for a total of over 30 diodes, my understanding, which is still coming together on this, was that combining them together first before the dichos was necessary, otherwise if I used dicros for each diode, I loose more power. Is that right? Maybe it is moot anyway, using mirrors one way or another regardless of how it is done with the losses adding up anyway.

I wanted to combine all of the beams together to make any color I want with a controller, but stacking does sound interesting and perhaps that is an option if it gives some kind of advantage. I have all the drivers now, all the S1 lenses, diodes and 16 sets of dicro mirrors, no mirror mounts yet though. I am interested in peak intensity, so the prisms are on my list now.

I am still unclear regarding the diode mounts, should I have diode mounts which allow movement in at least two axis?

Thank you for your help! A+

 

[LaserCo]

does anyone know the specifics on how when you bounce from a mirror the polarization changes.. thinking.. that you might be able to eliminate a waveplate to perform that function?

 

[CDBEAM777]

does anyone know the specifics on how when you bounce from a mirror the polarization changes.. thinking.. that you might be able to eliminate a waveplate to perform that function?

Ya....I think Laserman 532 or Andy_con did it over at PL !! See pic !! Bounce,Bounce and Bounce !!!....I am unclear on the exact position/sequence !!! Seems lika too much PITA for my liking. Takes room....and more alignment BS....and therefore....more things to go " outa alignment"

Waveplate's work....and may have less optical loss as well...not sure....but...every bounce = some optical loss !!! Stink'in laws of Physics (Optics) again !!! hahaha

And....a little seasonal humor for LP !!! OH....and...Bing the term " Harry Buffalo " to understand the true comedic genius at play here !!!

 

[Alaskan]

Thank you, I have some studying ahead of me to understand everything well enough.