1064nm laser is producing poor and inconsistent results on metals

Hey guys,

Have been playing with my 1064nm infrared module for a few months, and I have a two major issues with it.

The first that there appears to be an offset when compared to the diode laser, somewhere in the order of -4mm in the y direction. This means when I want to use the diode laser to cut a jig, then swap to the infrared, there’s misalignment between my elements. I haven’t been able to find any documentation to support this - anyone else experiencing the same issue?

Secondly, most of my metal engraving results are quite poor. The picture below shows a piece of brass that I’ve surface prepped, then tried etching a Charizard image on to. You can see how portions of it came out ok, but other sections don’t even show up. I decided to try a material test (lightburn) on the same piece, and again you can see some of the text, but nothing else. For the record, the Charizard was etched at 100% power, 6mm/s.

I desperately want to stop wasting materials and start getting clean, reproducible images on metal for sale. Can anyone help me out with this?

It looks like it is going out of focus as it moves to the right.

I have a fiber 1064 and it’s focus is critical, especially with a shorter lens. If you move a short lens a very small distance out of focus, there is no effect on the material.

I’d double check that everything is flat. Measure the distance from the laser to the material and ensure there is no change. A focus block is better than a ruler, as it’s easy to tell if you can wiggle it up and down or can’t move it.

Both co2 and the 1064 are both in the IR spectrum… the 1064nM is 10 times higher in frequency than the IR co2 machine … both are IR…


I’d also expect more …

Is this a filled line or what ?

:smile_cat:

Hi.

Probably everyone with both the “blue” and IR xTool heads :wink:

There’s a recent (short) thread with one -perhaps the only one- solution:

The solution is to measure and adjust accordingly.

As @jkwilborn said, the focus is most likely off.
The focus arm on mine is (or more precisely was :slight_smile: ) ~1.5mm too long.
I haven’t had the possibility to find out exactly, but I will in the near future.
And I will also reveal my take on an adlustable focus arm in form of a technical drawing.
In short, I hacked the tip of the arm and drilled and tapped the stub for a M2-0.4 allen bolt.

Another IR head user reported 1.6mm of excess lenght.
While 0.1mm might not sound like much, trust me, it is with this IR head.
At least if high accuracy and good contrast in photographs is required.
And furthermore, if any kind of speed is required, 2W optical power means there’s pretty much no power to spare with out of focus beam.

That very narrow focus range means also that a special care have to be taken with the workpiece to be absolutely flat, and the working surface has to be absolutely parallel with the laser head in every point.
Otherwise inconsistency is all that can be expected.

To find out the correct focus, just a simple ramp test is needed.
200mm and 3mm (or eqvivalent imperial units ~8" ~1/8") produce accurate enough results IMO.

How You decide to correct the focus arm being too long is entirely up to You.
For my tests I simply raised the workpiece (aluminium offset sheet) by 1.5mm.

Regards,
Sam

Both of these points were great. Thanks for the alignment thread, I didn’t come across that one in my search.

I spent a lot of time yesterday measuring and squaring everything, and ran a ramp test. I even removed the rubber feet on the risers to make sure they were sitting flush on my spoil board. End result is that the focus arm is 0.6mm too long, but even worse, I have a 0.8mm rise across the length of the x-axis gantry.

Really starting to regret buying this machine, and contemplating switching to a glavo 1064 instead.

They are a tad bit more expensive…

Good luck

:smile_cat:

Only twice as much as this PoS!

Hi.

You’re welcome.
While LB forum engine is one of the best (of the few :wink: ) that I’ve came across, it still has its peculiarities.
But I feel that us xTool owners are in the same proverbial leaking boat, so I’m confident that when problems arise, there’s always someone who can point us in the right direction to patch at least some of the holes.

So the folks at Makeblock must’ve done something to rectify the issue :slight_smile: .
Or the focus varies a lot between the units :frowning: .
In any case, on Yours it’s better, but obviously not perfect.
While the IR head (or any laser for that matter) will work to a degree when out of focus -either intentionally or unintentionally- power has to be increased and/or speed reduced, sometimes rather dramatically.
And as I said earlier, contrast always suffers.

That’s worse, but if the rise is linear, all You have to do is to shim the legs or to fabricate adjustable feet.
If the rise is not linear or if a straight edge test reveals that the gantry is bowed, a warranty claim might be the best solution.

While the xTool frame does look very pretty and well thought of, there’s always room for improvement and fine tuning with anything budget oriented.
My only experience is with xTool frame, but it did require some squaring up in assembly stage.
Very little, but still.
From what I’ve read over the 'net, most of the budget machines/frames need much more tinkering to get everything absolutely right.

Because of the way I wanted to fabricate the enclosure for mine (ATA style), I did turn adjustable hard plastic feet on a lathe for mine to fix the machine rigidly onto the bed for transportation.
All the initial testing was done with the 20W “Blue” head, so I didn’t find any problems with the feet lengths measured and equalized only with a vernier caliper.
I assumed that I have to revisit them anyway when I start to use the whole bed with the IR head.
And again when I eventually make the bigger ATA style enclosure to be used with the extensions.

I somewhat regret choosing Makeblock as well, but for different reasons :wink: .
I was a bit blinded by the pretty and well thought of looks and features, and the praising (paid?) reviews.

But, for the more budget oriented hobbyist like me, I feel that D1 Pro still offers a good -perhaps better than the competition- platform for improvements.
If I’d bought the machine solely for making profit, I would have looked for industrial machines with 10 times the budget.

Could the marketing be more truthful: Yes.
Could the price be lower: Always.
Could the work area be larger: Definitely.
Could the XCS work with older less powerful laptops: Heck Yeah.
Etc.
But…
Given the fact that both the 5W+ and especially IR diode lasers are very much in their infancy, I for one gladly take what is offered, for a price I can afford.
I would really much like that there wasn’t any tinkering required and I could just design and cut/engrave/mark since for me the laser is just a tool, but at this price point it won’t happen.
A degree of realism is always needed when diving into something brand new, no matter the claims the marketing departments or paid reviewers make.

OTOH, It’s been a quarter of a century since I studied industrial scale lasers more throughly while at Poly, and those beasts did require a lot more work to optimize the productivity.
And both the cost of the system and the running costs were astronomical.
And any downtime for whatever reason did cost the company dearly.

Also, no-one in their right mind would’ve assumed that in a few decades first the CO2:s and then other lasers capable of precisely cutting anything tougher than a sheet of paper would for a reasonable price be available to the general public either.

So, even if it doesn’t always feel that way, I for one think we do have everything very well indeed :slight_smile: .

Regards,
Sam

Money is a driving issue… I wouldn’t have a laser if it wasn’t for the Chinese producing them at such low cost compared to those made in a more open environment.

My $7K fiber is comparable to a US made $70K models, so I can’t complain too much that I could actually afford one of these… Much like the China Blue 50W co2, if all the parts were high quality with good support it would be 10 times the $2k I paid for it… and I wouldn’t have any of them… :sob:


A number of us purchased a jl1 from Amazon for < $80. It has a relatively low cost laser module, 500mW, I think… so I purchased a NEJE40630 laser module (about twice the cost of the jl1) that is now mounted on the machine. I actually purchased two of them, one is still stock on the left, in the photo …

They could make a fortune with these if they would just use firmware that would support any software… most of us had to upgrade the firmware (supplied by the maker) to use Lightburn…

Here’s one of the threads, if you have any interest. When you see these for this price you have to snap them up, they don’t last long.

There are also a couple of firmware builds from other laser people…

:smile_cat:

I’m also seeing terrible results, even after Xtool support saw fit to swap out my 1064nm module for the same issue. I’ve walked through focus settings numerous times and the results clearly aren’t cutting it. I’m not convinced after reading many online “fixes” that focus or leveling is the issue The work piece is level, the rig is stable and tight, and works beautifully for all my wood uses, but the example image is what I get using any setting deemed suitable from a materials test on aluminum. In this piece I tried 79% power and 167mm/s speed as that was the best results on the materials test, but as one can see, the engraving results varies across the entire piece.

No idea what to try next as the results never change and even vary from one run to the next.
The one thing I’m noticing is that the unit seems to slow down in the areas that are faded and the beam seems to drop in intensity as well, when it clearly shouldn’t be.

Any suggestions are appreciated

The controller will ramp the power down on the laser during deceleration to keep from burning, that’s why the edges fade out. You can turn on continuous power to prevent this. The option is in the window that comes up to setup your layer.

Screenshot 2023-10-04 062647

With this option checked, the laser power will remain at what you set it even when decelerating.

Before you mortgage your house for a Galvo, consider this. This is a mechanical issue that can be fixed. Making sure you are on a flat surface, loosen all the leg screws to let the frame settle flat. If that does not get you close enough, you can shim the low side (be sure to check the gantry both front and back) with brass or stainless shim stock, or even Mylar. The plan is to make sure the frame is the same distance from the table (or spoil board) on all 4 corners. As a thought, is the spoil board flat?

If this does not work, then go for the Galvo machine.

After many testings on coated aluminum, I’m convinced that Xtool’s IR module is just not suitable for the task. I’ve performed many material tests and see the same issue at just about any speed and power setting. Inconsistent engraving, which leaves blotchy results throughout the job. I obtained a calibration gage from Xtool, that clearly shows they’re built in gage on the module is not accurate, but the results and overall performance is in question.

It does seem to work far better on plain brass or copper, although so slow, I wouldn’t recommend it for anyone thinking of using it for anything beyond a retirement hobby.

On coated aluminum, there appears to be no adequate setting to burn away the coating (even on Xtool’s supplied cards) evenly, so the results don’t look bad.

I’m reluctant to even try on the rotary setup for cups. I’m very disappointed with the results or support from Xtool, although after many back and forth, they did replace the module, but the results are no different.

Woodwork has proven to be great, so clearly the problem lies in the overall weakness and very strange behavior of the IR module’s movement along the X/Y axis. Instead of crisp movements like the standard 20W module, the IR module is remarkably sluggish and slows way down as it approaches the edges, like it’s afraid to overshoot. The regular module does not act like that at all. Even moving across the material from one element to another it slows down on the approach. It’s just strange. Understand this is NOT just a LightBurn software issue. The machine behaves the same using Xtool’s software with the IR module as well. sluggish and slow.

That’s a lot of mass compared to the relatively lightweight visible dpssl module.

When you add mass, all the machines operation must generally be slowed…

Blame Sir Issac Newton:crazy_face:


Very sorry to hear such a poor report. I think these are new and will change, but they are very limited in power and controls that make this frequency for more widely functional.

Good luck

:smile_cat: