Hello, I am really looking to purchase my first laser, which will most likely be the Omtech K40 40W CO2, the reason I want it is for watermarking paper, making paper just thin enough in certain areas that the design is not visible under normal viewing but visible when backlit. I’m looking to find an sample online or demonstration of this to help me go forward with the purchase. If it cannot do this, then I will not be buying it. Pleaseeeee help!
I have a 45W machine, currently and make targets for alignment. Although it’s out of TEM00 mode, here is a target so you can see the marks I made on them with the co2. The targets are lased using watercolor paper.
It should work, but will be very sensitive to power adjustments and will be close to the tubes low power end, which could limit you. Most of these won’t lase below 10%, so the lowest power is likely about 3W. Most K40 machines are 30W, not 40W machines.
This will be the tough part. Lase low enough power or go fast enough power to create a non-visible surface mark…
There are plenty of people with a standard type of K40.. hang in there and see what others have to advise.
Good luck
I tried that on Tyvek (which is basically spun polyethylene) with poor results:
Some tests on posterboard definitely thinned the paper (on the upper tags), but left visible results:
Other folks have similar results with very heavy paper using a diode laser.
However, a laser removes material by burning it, which generally leaves a discolored residue. Burning a layer of paper without producing ash seems very difficult.
You must test this with the specific combination of laser and paper you intend to use, because a laser is likely the wrong hammer for watermarking paper.
This is what CHATGPT has to say, Yes — burn marks showing during laser watermarking are often due to the type of paper, and wood pulp paper is significantly more prone to visible scorching than 100% cotton paper.
Here’s why:
Why Burn Marks Occur on Some Papers
1. Wood Pulp Paper Contains Lignin
- Lignin is a natural polymer in wood that burns and chars easily.
- This results in browning, yellowing, or even blackened marks where the laser hits.
- It’s difficult to get a clean, invisible watermark on standard copy paper or pulp-based stock.
2. Cotton Paper Has Better Laser Tolerance
- 100% cotton rag paper has no lignin, making it more thermally stable.
- The laser tends to evaporate fibers (thin the sheet) instead of scorching it.
- You’re more likely to get a “clear” watermark-style thinning — especially if power settings are low and well-controlled.
KEY LIGHTBURN SETTINGS TO TUNE
1. Image Mode: Grayscale
- Go to Cut Settings Editor >
Image Mode
→ select Grayscale - This allows pixel-based power control (0 = max burn, 255 = no fire)
2. DPI / Image Lines Per Inch
- Set to 254 DPI (1 pixel per 0.1 mm) for accurate control
- Avoid going higher unless your laser is finely tuned (higher DPI = more overlap = more heat)
3. Power Settings
Setting | Value |
---|---|
Max Power | ~6–10% (on 40W laser) |
Min Power | 0% |
- Lower if needed: test starting at 5% Max, then increase slowly
- Cotton paper requires far less energy than wood pulp
4. Speed
- Set around 150–300 mm/sec
- Too slow = more burn; too fast = might not mark enough
- For watermarking, start with 200 mm/sec
5. Pass Count
- Set to 1 pass
- Multiple passes will overheat and brown the paper
6. Air Assist
- TURN OFF for watermarking
- Air assist can flare fibers or cause brown edges
7. Defocus Slightly
- Raise your laser’s Z-axis (bed or lens) by 0.5–1mm to soften the laser spot
- Defocused beam = gentler ablation
TEST IMAGE FOR LIGHTBURN
Create a grayscale gradient from white (255) to dark gray (50–70) — avoid black (0) for now. For Photoshop:
- Use a portrait image in grayscale mode
- Avoid areas darker than 70 unless you’re okay with stronger marks
- Save as PNG or BMP
LightBurn Cut Settings Summary
Setting | Suggested Value |
---|---|
Mode | Image (Grayscale) |
Speed | 200 mm/sec |
Max Power | 6–8% |
Min Power | 0% |
Air Assist | OFF |
Lines per inch | 254 |
Pass Count | 1 |
Z-Offset | +0.5mm (optional for defocus) |
IF ANYONE IS ABLE TO USE THESE SETTING TO TEST FOR ME, I WOULD HIGHLY HIGHLY APPRECIATE. THANK YOU. |
Note that CO₂ glass laser tubes generally do not fire reliably, if at all, below about 10%.
Applying this 254 DPI grayscale PNG test pattern I used for some power measurements in Pass-Through
mode:
To some old letterhead that’s as close to cotton paper as I have, in a 60 W CO₂ laser …
This column shows variations around ChatGPT’s settings:
The right bar is the 1 mm marker in the image. The left bar is the darker part of the grayscale.
This column uses settings around 400 mm/s, with the top pattern at Max
=11.0% and Min
=10% (the others have slight variations):
Bottom line: a CO₂ laser doesn’t have much dynamic range against paper …
Good infirmation posted here. If your primary use is for watermarking, I recommend a diode laser rather than CO2. You can reduce the output to less than 0.25%, meaning you can output 0.1w with a 40w diode laser.
I’m kind of confused here about what occurs during a passthrough.
I thought it was a scanning operation, so the head should be at speed before it needs to lase. If what I say is true, then minimum power setting has no use as these are based on head speed.
What does the IN terminal of your LPS reveal?
No matter the laser, this is usually why most people have problems doing grayscale photographs.
AIUI, pixels range from 0 through 255 and, on a grayscale image, get power levels from Min
to Max
, respectively. That’s not the case with dithered images or Fill
layers, where the Max
power level applies to all the active dithered pixels along the entire scan line.
Pretty much what you’d expect.
This gray-bar target:
Produces a linear-ish analog voltage (cyan trace) and somewhat less linear-ish tube current (green trace):
The magenta trace is the low-active L-ON
signal and yellow is the X axis DIR
signal for scope sync on scanning direction changes.
I’ve always used minimum/maximum values in vectors, not like this.
If you’re right, have you tried modifying the start speed to adjust the power range for a more compressed, power wise, image?
AFAICT, the Min
value keeps the corners from fading away:
That’s with Min
=7.1% and Max
=7.5% and the coolant at 20.7 °C.
Running the same test with the same power levels and the coolant at 20.5 °C drops out on the corners even worse:
So, basically, screwing around near the bare minimum power to fire the tube turns the laser into a sensitive thermometer.
Only to discover it doesn’t have the effect we all think it does while engraving.
For even quasi-production work, I’d leave well enough alone …
The moral of the story here seems to be, “You can’t shoot squirrels with an elephant gun.”
It appears to get the finesse you need, a 10w diode laser would be ideal. But I admit I am biased.
The trick would be getting enough power on white(ish?) paper to remove a layer or two, as visible light lasers have trouble with material reflecting most of their energy.
Just another reason why my results, no matter how good or bad, are mostly irrelevant.
My S30 10w has enough power to blast a hole through 80# white bond paper. The real worry would be NOT blasting a hole. A lot of experimenting and incremental power setting would be required. Add to that the variability of the material and I think this will be a challenging project.
Now you’re saying they work on speed?
It doesn’t work with engraving as the minimum value is not needed or used.
It depends on what the layer calls for and whether the controller is Ruida-ish or GRBL-ish; we’ve been discussing Ruida controllers here.
Dithered images are scanned at constant speed. Because the original grayscale image is converted to (lower resolution) patterns of black-and-white pixels, the Max
power applies to the black pixels and the laser is turned off for white pixels; there is no grayscale information in the image.
Grayscale images are scanned at constant speed, with the layer’s Min
and Max
power levels applied to each pixel, scaling the 0 pixels to the Min
power and the 255 pixels to the Max
power. Ordinary grayscale images get resampled to the layer’s specified DPI/interval, Pass-Through
images sluice through without resampling.
Per the doc:
Fill
(a.k.a. “engraved”) shapes are scanned at constant speed, so the Min
power is again irrelevant, except for rubber stamp Ramp
mode where the power ramps down to Min
just outside the edges.
Conversely, each vector / curve on a Line
layer runs at variable speeds as set by the acceleration limits. The Min
power gives the power applied at the lowest speed, typically at corners, per your chart. If it’s too low, the laser shuts off as the head slows going into corners, as happened on those DVDs.
So, as always, it depends.
Hello, do you say this because you have a 10W diode laser, are you able to test for me and show me the results please and thank you, I’m specifically looking to buy the laser for that purpose tbh
Thanks everybody for the replies, most of the answer seem way too technical and over my head, I have no idea what is being discussed considering I am 100% new to the whole laser community. But seriously looking to buy one, regardless of CO2, diode, whatever it is just so I can thin a tiny later of paper to produce a watermark. Meaning the mark shouldn’t be visible in normal viewing but since the paper is thinner in a certain area, more light should pass through when viewing holding the paper up to the sun or backlighting it. Thank you for all the replies!!
No, it would be meaningless because I do not have your paper. Any other paper testing would not give results you can apply to your project.
Take your paper to a local owner of a diode laser. They may perform the tests you are requesting.
You may have more success by stepping outside your parameters and trying out an ink jet type printer that prints with lemon juice or some such.
Should you be able to do this, then the lemon juice should show up when the paper is lightly heated. This is an old schoolboy trick for sending secret messages.
I am sure that other more reliable substances, other than lemon juice, also are available if you look.
Just a thought.