DIY UV Laser suggestions - 'my first laser'

Community Laser Talk
1

The ‘first laser’ is actually only kinda true; I’ve used/repaired lasers in an industrial setting before, but never owned one (other than a diode attached to my cnc)

I’m looking at UV lasers as they seem to complement what I do most, and the JPT Seal-355 seems a good source for this. Has anybody built a UV laser before, and is it worth the savings relative to say the ComMarker Omni X - Looks like I could get 20w from JPT with plenty to spare for the same total price, and the kinematics look pretty simple. Essentially mounting the laser on a column which can move up and down, and building out the optics, electronics, pneumatics, and enclosure.

Are there any gotchas here? Anybody with experience? Are there any other lasers worth considering in the same price band?

2

Is that cost comparison valuing your own time at $0/hr?

Alternatively, would it bring joy to hack your way to a solution?

It’s certainly possible - but I myself don’t have specific knowledge that could assist with a source>up build, and there likely aren’t many that fit that bill here, but we’re eager to see how it goes if you choose to!

1 Like
3

Yup, of course! Valuing my time properly would make the idea insane. :slightly_smiling_face:

We (at least I) generally do this for the love of the process, and to learn how the machines work, with the added benefit of learning how to do other stuff along the way.

I don’t like my solutions looking hacked though; tend to spend all those extra hours at $0/hour making it look nice and resilient. Surprised we don’t have more DIY’rs, I’d have thought this’d be one of the places to ask. I’ll have a look around.

ps. Of your lasers, which do you find most useful?

1 Like
4

I’d say there’s a lot of diy Co2 lasers here. But as stated, the time and energy for fiber and UV diy builds kind of make it a harder undertaking.

But I’d definitely want to see your progress and thoughts behind it if you do go for it!

2 Likes
5

I’m currently cadding and pricing it up, but I actually kind of see it as a bit easier (and certainly safer) than a CO2 laser build. No HV supply, no oxy assist, fire risk is a lot lower.

If interested I can let you know the cost of the head/source once I get final quotes back from JPT and CNI. I should probably ask cloudray too. But this is what I think I’m looking at: (GRBLHAL controller, lightburn for software)

  • £4000 25W 355nm DPSS/UV laser.
  • £300 Galvo scanning head
  • £300-500 1000W chiller (I think I’m capped at around 30W with a chiller this size, I have to check. There’s nothing interesting about building one, so will likely just buy at auction.)
  • £300 Linear components and materials for column/Kinematics (I have loads of ground flat stock for the base already. Probably also motors and rails etc spare)
  • £100 Other electronics which I can’t scrounge.
  • £100 Enclosure
  • £100 Gas control with vfd
  • £100 rotary axis

Around £5500 roughly, the same as the ComMarker 12W.

And probably another 200-300 for random mistakes made during the build.

1 Like
6

Don’t think there’s a commercial grbl based board that will control a glavo system. Might have to rethink this… If there is let me know. Most of the gavlo one are seen as grbl and therefore aren’t presented with options such as 3dslice.

If you keep the watts low enough, you can get away with 5 or 10W. Do you have a need for a 25W UV laser?

Don’t know what you need for gas control.

I don’t own one of these, but galvo operate a bit differently, so I’d double check the parts again, especially the control board.

Good luck – We all want to see it. :+1:

:grinning_cat:

7

This is partially solved for digital galvos, though it’s certainly not plug and play and there’s still a bit of work to do to process gcode to xy2-100. But we’re in a position in hardware now, where it should be doable more easily.

Ideally, I think 25w for the type of work I want to do. (cutting watch gears and parts) If going 5 or 10w, the problem is ‘solved’ so to speak, and a commercial system works. I think I’d just buy the 12w ComMarker.

…That said I’ve heard back from CNI now, and the prices for ‘full systems’ are far more reasonable than expected. It needs a bit more back and forth but I believe we shall reach accord. I wasn’t aware that the big manufacturers sold full systems to individuals. (60W comes in at around 14K USD shipped including the chiller, nitrogen control, controller etc etc) - I’d probably still need to build it an enclosure.

8

Good luck… Here’s a pdf you might like to peruse.

If I get something new and it doesn’t work, I contact the vendor… :face_with_spiral_eyes:

Good luck.

:grinning_cat:

9

Many thanks, still learning all about them so this is useful as is everything! To add to the complexity, you also have step/dir galvos and low resolution galvos that you canm :smiley: I think the analog ones are more hacky/hobbyist with direct DAC/ADC control and tuning, whilst digital is easier to control via known protocols.

I am waiting back to hear from JPT again now, who asked me for more information on exactly what I want to do with it so they can suggest a product, and then I have heard back from all 3 major providers and can consider the next step. (Not including trumpf, who I’m discounting on cost grounds)

10

25 watts is a lot of power for a UV laser! The dot size at this wavelength is way smaller than a regular fiber, and therefore more energy-dense.

But note that the 355nm UV ray won’t do much to any ferrous materials, even at 35 W.
You’ll need a 1064nm fiber source if you want to cut metals.

11

Honest answer: I don’t know.

It’s actually quite hard to know what I want in terms of power and pulse. Of UV lasers I’ve only ever played with a 355nm ns 10W hobbyist class, and used a Trumpf 343nm fs 200W.

The 200W does everything I want absolutely seamlessly, including cutting watch gears, faces and jewels and doing every ablation task you’d want with little to no post processing. It’s far too expensive outside industry though. [It also cuts stainless fine]

The 10W does a little of what I’d want, and potentially if it was mine, and I dialled it in properly and added nitrogen assist it’d be cleaner, but I suspect it’d be somewhat a project anyway.

Unanswered questions remain regarding the frequency and how much it matters (800khz on the fs lasers vs 26k on the chinese ns ones I’m looking at.) - My suspicion is this doesnt matter a huge amount for my application as I don’t need to vary power precisely, but I’m not sure.

There’s also the question of beam quality - All 3 chinese suppliers claim to have a superior one to the Trumpf docs. I’ll get a callback from a Trumpf technical guy hopefully tomorrow to discuss theirs. It might just be older but it’s curious.

I think I still need to learn more before committing to any option. One thing that makes this very difficult is that unlike fiber or even 10kw diy laser cladding builds, there’s very little in the diy/diy industry world about UV.

12

If anybody else is looking for information on precision UV lasers, I’ll leave this here just in case. This is what I’ve learned from speaking to people in the know. Though I still don’t know enough. Please do correct if wrong on any point. One reason I’m writing this out is to learn/remember myself.

ps. I’m talking about ‘UV’ lasers, but actually 343,355,515, or 1030NM matters far less than I initially imagined.

Safety first: Reaching out to the various people I know in academia and industry, one thing has become clear. This isn’t a DIY job on the source side. Multiple professors and grad students blind themselves and each other doing this sort of research.

The hobby class claims of near micron accuracy are horsecrap. By a factor of 50x or so. 2 major sets of factors play into small diameter cutting and marking, and these are as follows:

Spot Size

This is defined by a few factors:

Beam Quality [at source] measured in M² : Anything over 1.2 will not get tight spots, and it gets progressively less tight (multimodes fibers have a 0.3mm or so spot size!)

Galvo scanner quality: A galvo scanner can achieve a 5𝜇m spot of even lower, but this is a damned good one [read research quality] - 10𝜇m with a beam quality of 1.1 and good quality optics is achievable ‘normally’ and this is the kind of limit.

Optical focus and quality: F-Theta lenses do have inherent limitations as above. A short focus lens with excellent optics is essential. Because the focus range is sub 0.5mm, you’re also required to devise a way to non destructively probe your object at the ablation points if it’s 3D (Like in my case) to ensure focus.

Minimum hole/mark Diameter/quality

Spot size directly translates to 'hole size, but as the laser works by ablating material with energy - the longer you fire energy into the material, the type of material, and the intensity of the pulse all matter.

Pulse duration (NS = Nanosecond, PS = Picosecond, FS = Femtosecond) - In the standard NS setup, you’ll typically fire a pulse for at least 9ns. This can be reduced to pico or femtoseconds. The speed of the pulse determines the total energy being transferred to the item. A longer (ns) pulse will typically burr and make a conical mark. An FS/PS pulse on the other hand, will transfer far less energy, hence making a smaller colder hole. This is most easily demonstrated by two of the lasers I am considering. A 60W [nd:yag @ 355nm] laser firing for a 20NS duration, will transfer 1000mJ. A 60W [femtofiber @ 1030nm] laser firing at 200FS will transfer under 2nJ.

13

Thank you for your informative description.
Can you document or refer to the many students and professors who unfortunately can never see again?

14

Multiple is not many. And quite honestly, this is ‘grapevine’ information, so probably somewhat hyperbolic. On the other hand, a very quick google does show multiple incidents, so it looks correct. I won’t spam too many as I am sure you are capable of your own searches.

https://physics.aps.org/articles/v15/141 (lol)

15

Please don’t get me wrong, laser machines are dangerous without a doubt! But all (to my knowledge) accidents that are known have happened because people have not observed/respected the stated safety rules. Here in the LB forum, which has gradually gained many members, not a single one has reported a violent accident with their laser. Apart from stupid people who have grabbed into the laser beam…(including me), I have not noted a single eye injury - thank God.
This is not intended as a discussion round, just an observation.

16

I’m just not sure where you are going with this. :slightly_smiling_face: I simply said, that given the increasing complexity and power of this project, I am uncomfortable with DIY’ing the source at my skill level, including harmonic generation, beam collimation, and galvo scanner motion control. I would rather buy the source and galvo scanner as a complete package, ready to add my optics to.

Others on the forum may be more competent, confident, or both. That is fine. For me personally, I feel this is too dangerous.

17

The first one isn’t a laser accident, it was caused by a window in one of the containers shattering when it was being tightened.

All of the others were rather exotic lasers based on a femtosecond sapphire laser or something similarly exotic. None of these are what a normal hobby laser would be. When the device coolant runs at 80C(176F) and takes 30 minutes to stabilize, it’s outside of my league.

Building the source, would require very expert fiber/pump diode alignment instruments and would likely be more expensive than purchasing one already built.

I think most of us that diy, don’t think of building a co2 tube, which is possible, just like most of us would rather distance ourselves from a 30kV power supply rather than a diy approach.

Let us know how you are coming along…

:grinning_cat: