Page 1 of 2 12 LastLast
Results 1 to 10 of 15

Thread: Eight Diode Knife Edge Array - best practice?

  1. #1
    Join Date
    Aug 2014
    Posts
    230

    Default Eight Diode Knife Edge Array - best practice?

    I've seen 2X 4 diode knife edge arrays which are both sent (with a rotator on one side) into a PBS cube to combine eight, but I want to combine eight which are all the same polarity using a simple knife edge array. Can someone point me towards a photo of an eight diode array so I can get an idea of how wide the beam will be for 1 watt green diodes? I don't have room for a PBS cube.

    I'm thinking two levels of 4 x in a line to make a:

    ::::

    pattern is the way I want to go. Has anyone done that? Where four are at one level, then another four in front or back at another level, or is it better to alternate each diode to their own level to interleave them that way? Lasertack used to sell 4x arrays which produced a :: pattern doing that, but no longer sell them.
    Last edited by Laser57; 03-26-2021 at 16:57.



    _________________

  2. #2
    Join Date
    Oct 2012
    Location
    Germany
    Posts
    1,320

    Default

    ... I've got a DIY 8x module from a friend - have to test the combined beams, but not this bad

    Viktor
    Attached Thumbnails Attached Thumbnails 2mdmyj3a.jpg  

    Blaues 8x - Modul1.jpg  

    Blaues 8x - Modul2.jpg  

    Blaues 8x - Modul3.jpg  

    Aufruf zum Projekt "Müll-freie Meere" - https://reprap.org/forum/list.php?426
    Call for the project "garbage-free seas" - https://reprap.org/forum/list.php?425

  3. #3
    Join Date
    Aug 2014
    Posts
    230

    Default

    Quote Originally Posted by VDX View Post
    ... I've got a DIY 8x module from a friend - have to test the combined beams, but not this bad

    Viktor
    Ouch, that looks so hard to adjust and keep aligned (if possible to tweak again), but the output mirror method of combining the two 4x strings is unique to me, pondering on how I can use that with this, might be another direction to go for this. What is that, a 8 gball diode array?

    I just added (edited in) two photo's to my last post showing two ways I am looking into arranging the knife edge, my concern is either way is going to end up running out of room for a full eight diodes, either the diodes alternating sides, or in a long string will end up moving the parts off of the mounting platform due to its width being 62 mm. Maybe eight is too much to pack into this laser pointer head. I should just glue the mirrors on and see how much room is remaining, looks tight, but was hoping to find an example of another string of eight diodes being knife edged. I've been googling, not finding what I'm looking for and probably for a reason.

    I want to put them into a pointer. I'd like to expand the beams too to reduce divergence, as a single all beams together using that pattern into the expander input, so all :::: into a beam expander. From what I have been able to gather, the pattern won't be so wide as the periods shown in this text, more like 4 x 8mm

    ||||
    ||||

    (Tall side side 8-9 mm). I do not have a lot of width to work with, that's the problem, only 62 mm wide inside the pointer and my concern is the V of two strings of 4 X knife edge diodes, left and right, might run out of room for the width of the mount, and if a single string not wide enough as each additional diode gets an inset away to allow the beams to each pass by the next mirror. Maybe this isn't the best way to do this for that many diodes.



    Above photo's are two different possible ways to mount the diodes, maybe..... not sure which way is the best to go, measuring how much room remains if the diodes alternate sides seems like too tight. The copper strips are different thicknesses to mount the copper laser diode mounts on to, as one of the knife edge mirror mount assemblies and group of diodes will need to be raised so the beams form a :::: pattern, one set shooting over the other set of four. I could alternate how high each diode is all along the strings, one next to the other, but not sure what that will buy me. Years ago lasertack used to sell blocks for four diodes built that way, but I believe no real need to do so for eight diodes, just have each block of four at different levels.

    This project is very fluid, evolving. initially I was just going to mount four diodes with mirror mounts which had feet, but then I thought about using the nice railed lasertack mirror mounts I had, but they had blocks of aluminum diode mounts on them, so removed them and using these copper ones which have better heat dissipation which can be soldered to the copper strips I can mount to the aluminum block with screws on each end of the strip. With the 8 mm FL lenses I want to use, I need extra room due to how far they stick out from the diode mounts anyway, IF I can end up using those due to the constraints I am up against.

    The divergence for the fast axis of these diodes isn't so good, so I really do need to expand the output, if not the fast axis alone, both. Maybe I shouldn't worry about expanding or correcting the beams inside and just hang a expander on the output and don't use those long FL collimators inside to give me more room. If I limited the number of diodes to six, I could put a 10% loss prism pair inside (horrible!) to get rid of some of the fast axis divergence, but I'd still want to hang a regular beam expander on the output with or without fast axis correction anyway, even with the rectangle shaped beam pattern through a set of spherical lenses. I've seen one YT video of a knife edged pointer using a beam expander on the output, it seemed to work OK, not perfect, not optically elegant, but did reduce the divergence of each beam.

    Yes, I'm bored, rattling on about this project helps, the real issue I need input on is the configuration I should use to string these diodes for knife edging in the limited space I have, but seems I'm able to solve that problem by using only six diodes, less someone has a way to use eight they can offer? The problem I'm seeing is when I add mirrors, and each being 7 mm tall, that is 14 mm total space lost in the middle, maybe I can just make it with a 4 x 8 mm pattern, that's about all of the room I will have for eight diodes using the alternating side configuration, and if I string all of the diodes along one side, I might just make it but that is one ugly configuration.

    Maybe I should reduce the diodes to six, this gives me some extra room for a pair of prims to expand the fast axis and reduces the heat problem, eight of these likely won't run long with the heat they produce. This is typical for me, start out thinking four, then six or seven, then omg, maybe I can fit eight in there! Who needs a 10 watt 520 nm laser pointer anyway and with a prism pair and expander, the losses won't allow it with eight diodes. OK, 7.5 watts out of six 1.5 watt NUGM04 diodes?
    Last edited by Laser57; 03-26-2021 at 18:30.



    _________________

  4. #4
    Join Date
    Oct 2012
    Location
    Germany
    Posts
    1,320

    Default

    ... in my module is one of this 8x diode DLP-beamer heads with eight 4.5W-diodes -- so up to 40 Watts (if slightly overdriven, what's no problem with PWM and temperature check).

    But it's not meant for beaming, but for "material processing"

    Viktor
    Aufruf zum Projekt "Müll-freie Meere" - https://reprap.org/forum/list.php?426
    Call for the project "garbage-free seas" - https://reprap.org/forum/list.php?425

  5. #5
    Join Date
    Aug 2014
    Posts
    230

    Default

    Practical uses for lasers? Huh? People do that?

    Just toys or things of beauty for me Well, I think I have the info I needed to proceed. Just didn't want to start cutting metal before I had a better grasp on knife edging.

    Thanks for your input, after a few exchanges with lasertack.com today I have a new idea for knife edging I've never seen done before, Alex doesn't even know what it is, I will report later.

    Chris



    _________________

  6. #6
    Join Date
    Oct 2012
    Location
    Germany
    Posts
    1,320

    Default

    Hi Chris,

    ... I was "brainstorming" with Alex some years ago for coupling fibers to the blue 6.5W-diodes ... and he has one of my 9A-drivers, I've developed to fire some IR-laserdiodes with up to 30Watts@975nm out of a 0,1mm-fiber.

    Actually my "hottest" diodes are 270W@975nm with a "knife-edged" beam, combined from 35 single emitters into a parallel beam of 5x5mm square -- 0,2mm, when focussed to 50mm FL

    Viktor
    Aufruf zum Projekt "Müll-freie Meere" - https://reprap.org/forum/list.php?426
    Call for the project "garbage-free seas" - https://reprap.org/forum/list.php?425

  7. #7
    Join Date
    Aug 2014
    Posts
    230

    Default

    The new NUGM04 520 nm laser diodes appear to have terrible divergence but are RATED at 1.6 watts output, I have a hand full of them and want to knife edge them into a pointer. The two big challenges, aside I have little experience with knife edging, is heat and battery consumption. I have a pointer which uses 3X 26650 batteries which should last a few minutes of intermittent use and want to try to build an eight diode beast.

    Now, that's too much power, even for 26650 batteries, they will suck down flat too fast, but I can set the power output to low, normal or high by varying a PWM input control line to the drivers, so that part can be dealt with, but where I am still facing a challenge is trying to figure out how to put a beam expander on the output of the knife edge array and have the beam expand properly. The challenge is I have no knowledge in this area and have not yet put a big knife edge array together to experiment to know what happens when a pattern such as, below, are put through a spherical lensed beam expander:

    ||||
    ||||

    I found some remarks made by LPF member "UV Glue" (I might have that ID wrong) in the past when using a spherical beam expander (with a knife edge array, of course) the best pattern to use is a 2x2 pattern:

    ||
    ||

    Actually, his example was four dots tightly packed or arranged together, not these long rectangles. My thought is there might be another arrangement which will allow eight diodes through a spherical lensed beam expander with minimal distortion, and that is if I stair stepped and spaced the knife edge outputs to the shape of circle such as:

    ..............................|
    ......................|..............|
    .........................................
    ...................|.....................|
    ...........................................
    .......................|..............|
    ...............................|

    The above illustration shows the rectangle beams further apart than what I want, but it shows what I mean. I can get the beams much closer with a knife edge this way than individually mounting eight separate diodes in a ring, this is why I want to knife edge them, but I also want to reduce the divergence several fold too. I want the output to appear as a tube of closely spaced beams, don't care if the individual laser beams are clearly evident.

    I would use the double 4X4 arrangement to form a tighter beam but I do not think it can be put through a beam expander that way and work out well, or am I wrong? Member UV Glue didn't think that arrangement works well. My thinking is the beams going into a beam expander need to be symmetrical in placement through spherical lenses or there will be unequal behaviors, the rectangle shape of the uncorrected diode outputs will likely already produce some amount of distortion, but I'm hoping the beams will still look good. This is just for fun.
    Last edited by Laser57; 03-29-2021 at 23:23.



    _________________

  8. #8
    Join Date
    Oct 2012
    Location
    Germany
    Posts
    1,320

    Default

    ... I'm more into focussing with highest possible energy densities, not collimated beams -- but the 200W and 270W-diodes are pretty good examples for "extreme" knife-edging.

    There are 7 modules, each with 5 emitters, collimated per 5x-module, then knife-edged.

    Attached some of the internal images ... the "solely" dot at the ceiling was the parallel 35x-beam roughly in 1.5m distance ...

    Viktor
    Attached Thumbnails Attached Thumbnails Rofin-200W-Dioden - 7A-Karte.jpg  

    Rofin-200W-Dioden - 3A-Karte.jpg  

    offen.jpg  

    Rofin-200W-Dioden - 3A-Decke.jpg  

    Unfokussiert1.jpg  

    Ohne Beam-Guide2.jpg  

    Aufruf zum Projekt "Müll-freie Meere" - https://reprap.org/forum/list.php?426
    Call for the project "garbage-free seas" - https://reprap.org/forum/list.php?425

  9. #9
    Join Date
    Aug 2014
    Posts
    230

    Default

    Uber nice arrays, if I were into cutting I'd want one of those for sure, or one which can be used to SHG to produce visible spectrum beams. I've only had 50 watt FAP800 laser diodes to burn with using IR in the past. I guess those must be water cooled, that is such a huge amount of power I cannot image anything less would have enough cooling ability.



    _________________

  10. #10
    Join Date
    Oct 2012
    Location
    Germany
    Posts
    1,320

    Default

    ... yes, a friend is designing a driver with attached watercooler, fitting exactly beneath the diodes - my tesing equipment was with a big aluminium-plate, 3x 90W Peltier-coolers and a big cooler with forced air cooling ... attached some images of this setups ...

    Viktor
    Attached Thumbnails Attached Thumbnails 4xDioden.jpg  

    152877052_221629389653652_4431218177890209036_n.jpg  

    Aufruf zum Projekt "Müll-freie Meere" - https://reprap.org/forum/list.php?426
    Call for the project "garbage-free seas" - https://reprap.org/forum/list.php?425

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •