Improving A Laser Engraver's Resolution And Accuracy

Baby steps are important

Four years after I wrote about some improvements brought to my Eleksmaker laser engraver, I made quite a lot of progress on multiple fronts.

Laser module

I was regularly annoyed by the too irregular laser beam and finally acquired a NEJE A40640 model that's supposedly 15W optical, made of two diodes and that contains lenses offering a Fast Axis Correction (FAC) to reduce the beam's divergence. The result is an almost square beam that's roughly 60x80 µm by default and can be shrunk to even 60x60 µm at a shorter distance, or be made less than 60 µm wide if made taller.

This constituted the most important improvement because with a poorly shaped beam you can't do any good work. I must say that as of now I think I will never buy again a module without FAC. Think about it, previously if you wanted to cut a circle in wood, half of the circle was completed (e.g. horizontal direction) while the other one was still burning large areas due to the line-shaped beam. Now the same amount of energy is spread in all directions and can be made very narrow.

Air assist

I had been trying various approaches using aquarium pumps to implement some form of air assist to blow the smoke and make a better cut, but these didn't work well. The air flow was made of many irregular pulses and you could hear "puff puff puff" at the output of the laser's head.

I finally decided to order NEJE's air assist accessory for my laser module. Strangely it didn't fit well, it was forcing against the lens' screw. I suspect that the module evolved a little bit since I acquired mine and maybe some dimensions were slightly adjusted on new ones. Nevertheless, I could enlarge the opening of the air assist head so that it fits on my module.

And to get rid of the bad pumps, I finally ordered an AtomStack air pump. The device is really great. It takes 12V input, has a potentiometer to set the air flow speed, and has an output. I just assembled their hose with the one I already had. It further smooths the air flow and now at the output of the module, the flow is super regular and can vary from tiny to quite strong. In practice, most of the time, it's sufficient to set it to 1/4 to 1/3 of the speed in order to blow the smoke away:

 

The results are definitely better, but contrary to what is often seen on ads, it's not on wood that I found the most impressive results, but when engraving images or cutting acrylic. Previously the smoke was causing the beam to diverge a bit, making the result less precise. Now it's super accurate. It's important however to make sure that the piece being worked on doesn't move, because a moving air stream on top of it can make it move (which is another reason for not setting it too strong). That used to be true already for the module's fan anyway. I found that using a few tiny neodymium magnets can be very effective.

Changing the motors' resolution

OK now we're having a nice laser head with much better precision and less disturbances caused by smoke. Isn't that enough ?

In fact I'm using my engraver mostly for PCBs, sometimes for cutting stuff (wood, acrylic), and sometimes to engrave drawings or photos.

For PCBs you do want to have good resolution, otherwise you can't make a track pass between two integrated circuit pins. Or it will touch one side, or be too thin and disappear while etching. With 0.1mm resolution, when you have 0.6mm between two pads of an IC chip, that leaves a single space of 0.1mm on each side, and 0.4mm for the track. Or 0.2mm on each side and 0.2mm for the track. You don't even have an option of 0.15mm each and 0.3mm for the track. And it depends how these are aligned with the motors' steps.

For photos, you generally use Floyd-Steinberg dithering which further reduces the photo's resolution, and when working with 0.1mm dots, that becomes quite visible.

I had been thinking for a long time if it would be possible to find motors with more steps per round. But there was another option that suddenly came to my mind and that I had not yet been considering: what about finding pulleys with less teeth so that it requires more steps to make the same distance ? I searched the net for a few hours and found that the type of belt I'm using is designated as 2GT or GT2 and has a step every 2 mm. My pulleys had 20 teeth and a 5mm axis, and I found others with 16 and even 12. There is a 10-teeth model as well, but only for 4mm axis. So I ordered these, and managed to install the 12-teeth on my motors to replace the 20-teeth one. Here's the photo from the vendor's site:

One assembled, it looks like this (replaced, and with the old one as a comparison):

Doing this requires to adjust the number of steps per mm. It changed from 80 to 133.333 in GRBL's settings, but that's all that needed to be adjusted. I feared that the head would travel slower, but that's not the case. Apparently the speed is more limited by the head's weight and the motors power. However, instead of having a reproducible resolution of 0.1mm, I'm now getting 0.06mm, which precisely is the default size of the beam. Converted to DPI, that's 423 DPI.

This new resolution now allows to export PCBs as images and print them, the result is now good enough. And that's quite convenient because it also means using the same tool for all exports, with the same coordinates system, giving the ability to produce multiple images of various planes, such as the cream plane used to remove varnish and expose solder pads:

In terms of images, this has tremendously improved the result. The dog photo at the bottom is 675x875 pixels and is printed on black-painted aluminum at 0.06mm per dot, resulting in an image of 41x53mm. The result is really impressive, that's 16.6 pixels per mm, or 278 pixels per mm², vs 100 before. That multiplied by 2.78 the pixel density hence the possibilities of nuances in an image.

One problem however is that an image is almost twice as long to print now because there are almost twice as many lines. That was the opportunity for another improvement.

Bidirectional printing

In order to improve print performance, one possibility consists in printing in S form instead of Z, that is, printing even lines from left to right and odd lines from right to left, effectively avoiding a slow return-to-home operation after each line. The software I wrote to convert PNG images to GCODE, png2gcode, already supported such bidirectional printing, but this had always been ugly at high speeds. It was quite visible that there was an offset between each direction. This is not surprising, for three reasons:

• the startup acceleration is not necessarily the same in both directions. However this was addressed long ago with an option to add an acceleration margin to both sides, that I'm typically setting to 3mm ("-A3") so that the beam arrives at full speed on the first pixel to be printed.
  
• micro-stepping is used to control motor positions. Despite using the high-quality TMC 2209 drivers, which take the delivered energy into account to make steps homogenous, it's understandable that the belt's elasticity will not reproduce the exact same position when pulled in one direction or the other, and that it will depend on its tension. Here the belt is quite tight, but tightening it too much can also make it difficult for the motor to make it move in micro-steps.
• the instruction processing time in the micro-controller counts as well.

Till now, png2gcode would offer an option to set an absolute offset for right-to-left printing (that corresponds to the belt tension), and a time-based offset as well for the processing time. However, approximations that had been used till now were hardly reproducible.

The new laser head combined with the new gears was a great opportunity for trying to improve the situation by taking new measurements.

The test consists in printing on anodized aluminum, a rectangle that's 7 pixels high, 40 pixels wide, with a vertical line at 0, 5, 10 and 20 pixels. When printed only left-to-right ("-Mraster-lr"), it's perfectly regular. When printed in bidirectional mode ("-Mraster"), it's visible that every other line is shifted right by one or a few pixels. The same rectangle was printed at speeds of 600, 1200, 2400 and 3000 mm/min from top to bottom. The pixels are 0.12mm wide. The expected pattern is easier to understand on the top and the deformation is increasingly visible as speed increases. The photo had increased contrast to better see the dots:

This allows to see how much variation there is between them, explaining what is dependent on time, and what is fixed. After some calculation and tests, it appeared that the pixels when going right to left had to be shifted left by 0.12mm and delayed by 2.6ms. This delay is converted to mm depending on the travel speed so that in the end it gives only a distance.

With the right adjustments it's possible to align left-to-right and right-to-left and almost double the print speed. Here's a capture of the final results. It's still visible on the large rectangle that there can be around 10 µm variations in positioning because the vertical lines are not always perfectly straight, but that's very hard to notice on the microscope, let alone to the naked eye! The one on the right was printed at 0.06mm per pixel, and there the positioning resolution remains imperceptible.

The image of the lunch at the top of the sky scraper at the bottom of this page is 1206x943 pixels, rendered on a visit card of 72x56 mm, and took approximately 25 minutes to engrave. With unidirectional printing previously it would have taken approximately 45 minutes (it's not exactly twice as long because the return can be a bit faster when not engraving).

Beam narrowing

As can be seen on the test image below, the beam can be made narrower when it's a bit taller. This is important because if the beam is as large as a pixel, then when it sweeps an area as large as a pixel, it has effectively engraved two pixels. The image shows a test pattern with one dot every 0.12mm (the ruler is in millimeters). A zoom on the dots shows they're between 25 and 30 µm wide and approx 100 µm tall:

We can then exploit this principle to consider that the beam will imprint larger than desired and consider this for granted as soon as the beam turns on.

Example: let's say we want to print the green dashed line below (each square is one pixel the size of the beam). The laser dot (in blue) by default will scan from the left of the square to the right, and will effectively span twice its size. The part that was constantly under the beam will have received more energy, and the borders which were only a limited time under the beam will have received less. As can be seen as the beam advances from left to right (line after line), the pattern is reproduced but the contrast is limited.

Now if we take that beam width into account, we can make the beam start to light up later and extinguish it earlier, still providing a shadow around the borders but leaving the intervals totally unexposed. This is more efficiently done by having the dots twice as large as the beam and reducing the beam duration in half, as this would then preserve the intervals.

When taking all that into account, it becomes possible to print photos on painted metal after passing them through Floyd-Steinberg dithering and produce such stunning outputs:

When zooming in macro mode, it's possible to see the dog's hair as 1/16mm dots:

Similarly, a soom at the center/bottom of the lunch photo, zooming on the metallic beam close to a building shows some of the details, then we can zoom further on the beam and it's visible how the gray is obtained by alternating clear and dark lines:

Wrap up

All of this is a question of patience and experimentation. Now I'm able to easily print a photo on metal by using bright black paint and have enough resolution to almost always do it well on the first pass. In the past I had to adjust power, speed, and color conversion to compensate for the risk of too wide pixels ruining everything. That's no longer the case, as can be seen above with direct printing of photos at very high resolution!

Improving the Eleksmaker A3 Pro Laser Engraver

After starting to use my laser engraver, I found a few points that really needed to be improved.
[Update-2024-07-30: new resolution and accuracy improvements described here: https://wtarreau.blogspot.com/2024/07/improving-laser-engravers-resolution.html ]

Seeing under the piece

First, when cutting thick material like wood or acrylic, you never know if the cut is finished or if another pass is needed. This is the reason why I installed a 39x30cm glass on 47cm long aluminum corner frames, that I fixed using the extra screws and bolts that were provided with the A3 Pro kit. I had to drill a few corners from aluminum blocks to hold everything in place but aside taking measures and drilling at the appropriate dimensions, there was nothing really difficult:

The first frame I used was quite large (5cm), hoping that it could be used as a ruler, but that was a bad idea since it prevents the beam from reaching the medium beneath, so it is not possible to see if the material is entirely cut or not:

So I finally replaced the large frame and opted for a very narrow one (7mm), which has the additional benefit of having a thin line roughly 4mm from the edge. This further helps aligning the beam with a reference position anywhere on the X axis:

This one is still slightly thicker than the glass so that it can be used as an alignment ruler. However its metal is also thinner thus the frame is quite soft, and adding some weight on the glass can slightly bend it and slightly degrade the focus. Maybe I should have stacked two of them to improve the rigidity.

I bought a cheap mirror mounted on a telescopic stick that's made to look for objects fallen behind furniture, it's perfectly suited to looking under the piece being cut (typically with wood you never know when it's finished because even if the beam passes through it's not necessarily sufficient):

Keeping objects in place

As mentioned above, it is preferable to use paper tape rather than weights to keep objects in place. The tape I'm using is sold in any DIY shop to protect walls and windows when painting, it doesn't stick too much and doesn't leave an trace when removed:

It is very convenient to keep objects in place. But when you discover that the cutting didn't completely go through, it's hard to place the object again at the exact same place and angle. At least having one side of the material wedged along the edge frame forces the angle so that by running the engraving process again at low speed and low power, it is possible to progressively slide the object until the beam perfectly enters the existing trench. For other situations like PCBs where it can be required to flip them over when engraving both sides, I found that attaching another small part of aluminum frame along the right Y axis can serve as a second ruler:

Why not on the left side instead ? Because most of the things I have to cut cross the zero origin. So I prefer to move the head by hand to the right when I need to use an accurate ruler for such rare use cases.

Laser line

I thought that it would be very convenient to install a laser line underneath the laser module, in order to precisely indicate where the laser is expected to pass. For this I ordered a small laser line module working on 3 to 5V, it's made of a small transparent cylinder placed in front of a laser diode, making the beam span a very large size (it covers roughly 45 degrees):

The problem is that the beam is several millimeters thick, so it's totally impractical for alignment:

In the end I think it remains more convenient to keep the laser perfectly aligned with the aluminum edge and align objects there.

Extracting smoke

Engraving and cutting with a laser consist in burning, vaporizing or melting matter. Some materials like acrylic do not produce smoke but do produce a lot of vapors that you can smell and that deposit all over the work place in addition to you lungs. And seeing the amount of thin deposit everywhere even on the laser lens doesn't make me very confident that it's good for the laser module's longevity. Other materials like wood, cardboard or ABS smoke a lot, to the point where the amount of smoke can disturb the beam. Therefore it becomes important to be able to remove this smoke.

There is a small (and noisy) fan at the top of the 2.5W laser module, but it's made only to keep the laser module cool, and it doesn't blow strong enough to even disturb the smoke in the beam, as can be seen below while cutting some wood:

I first tried to place a battery-powered flat fan directly on the glass close to where the laser was working. This proved to work pretty well as long as the laser was no more than a few centimeters away from the fan, and it was not very convenient to move the fan to follow the laser beam:

Using a much larger fan worked better but was not convenient and quite noisy. It also had the undesirable effect of blowing some freshly cut light parts and sometimes moving them under the beam, causing some other areas not to be totally cut. So I tried to attach a small and light PC fan to the module to let it move along with the module:

This first experiment was not fantastic, to say the least... The air flow from the fan barely reached the surface, it needed to exhaust much closer to where the beam touches the material, hence this second assembly:

I started by stealing the 12V power directly from the original fan's pins on the board above but finally found it easier to solder a JST-XH 2-pin connector beneath the 3-pin one to take the 12V from there:

This actually is the one that gives me the best result. It does have one downside though: when cutting very light materials like thin cardboard, it tends to make the finished pieces fly over the place even though it doesn't blow strongly. And sometimes such pieces either need an internal cut (one more reason for etching from inner to outer), or they can deposit a few centimeters away on a place that was supposed to be cut, which is more problematic. I found that catching them with a piece of paper tape without stopping the work is easy enough not to care too much about this issue.

I tried a completely different approach by sucking the smoke instead of blowing it, using a small pump and a very soft silicone tube ending in water inside a jar. Long story short: it doesn't work, the suction is too weak. While the tube can stick to your finger, the flow is too small to collect the smoke from a few centimeters around. Maybe it would work with a stronger sucker, but it would also make too many bubbles in the water and would also suck the water out of the jar:

Motor driver replacement

During the first tests, I noticed that the motors were lacking a bit of resolution and that when the laser beam was very thin, some parts could be missed:

I've read on various forums about the SilentStepper family of drivers which was supposed to render 3D printer motors totally silent and to significantly improve their accuracy by taking into account the amount of energy sent into the motor's coils and not just the ON/OFF time. Given that these ones are designed as drop-in replacements for the A4988 that my board is using, I thought about trying them. Most of these modules are programmable but I didn't want to go into this, so I picked the 2209 which supports 1.7A, doesn't require a heat sink with my motors, and defaults to 32 microsteps like the A4988. I just cut the unused programming pins and replaced existing modules:

The first thing I noticed was that the motors would go backwards. I just had to change the direction mask on my GRBL config for both axis (X=bit0, Y=bit1):

Grbl 1.1h ['$' for help]
$3=3

Once fixed, the positions and distances were working as before. However I was absolutely impressed by the total silence when the motors move. You don't hear them anymore at all, sometimes you can believe that etching is finished and you really have to look at the bream to be sure whether they're moving or not.

The second gain I got from this is that horizontal lines are evenly spread all over the distance. This is very important for raster images because before some lines would be left and others could be etched twice by being too close. Now the steps are perfectly regular. I tried to move the axis by sending various commands. The board is configured for 80 steps per millimeter, so we're supposed to be able to move by steps of 0.0125 mm. To be honest, I'm unable to say whether or not the motors move when I ask them to move this little, and it looks like when repeating the operations the motors move smoothly.

The third gain is that when moving fast we still keep some precision. I noticed in the past that when moving too fast I could sometimes fail a piece, because the laser head would miss a few millimeters on a large move. This has never ever happened anymore with these drivers. So they're really worth the money.

Enabling "laser mode" for raster images

The first time I tried to engrave a raster image (i.e. a bitmap image in black and white or with levels of gray), the result was awful. The motors would start and stop for every pixel with a different darkness level. It used to take a lot of time, to heat up the X axis motor, and to make burn spots at the beginning of each new segment of a same color.

I then thought "why not modify GRBL not to stop?". I just figured that it had already been implemented in version 1.1 under the term "laser mode". My engraver came with version 0.9, requiring an upgrade. I was a bit scared by horror stories on the net about some of the arduino-compatible modules there employing fake ATMEGA328P chips and not supporting being relfashed. I wasn't sure whether mine was affected or not:

In doubt, I ordered a small bunch of extra Arduino Nano compatible boards to flash the new version on. The problem I had with the Arduino Nano board is that it came with a Mini-USB connector while I mostly had Micro-USB cables. So I also tried my luck with a RobotDyn NANO V3 board, which comes with micro-USB and a good voltage regulator, but which is also much thicker due to the crystals and the ISP connector, preventing me from closing the cover:

I flashed the latest GRBL image (1.1h then) to the new device, and could start some tests. After about one week I decided that I wouldn't revert to version 0.9, so I decided to try my luck by flashing the original module that came with the engraver, and it worked. This allowed me to reinstall the acrylic cover on the controller board.

Laser mode is enabled by setting variable $32 to 1:

Grbl 1.1h ['$' for help]
$32=1

In this mode, the M3 command turns the laser on at a fixed power level, while M4 turns it on at a power level relative to the motor's speed during acceleration, to avoid burns. And more importantly, the motors do not stop if you don't send a new M3/M4 command while changing the spindle ratio. This means that you can produce very compact GCODE code using series of G1Sxxx commands describing segments of different intensities. This makes the laser go left to right and right to left and engrave line after line, exactly like an inkjet printer. Furthermore, G0 (travel) automatically turns the beam off so you can even optimize the printing using a fast traveling over white areas. It's worth noting that the X axis motor tends to heat up after a while but it remains OK. The resulting quality is extremely good. However it depends a lot on the material and on the software used.

I noticed that I often had to manually edit the output raster files to remove extra M3 or M4 commands placed there so that there's only one in the file.

I was finally fed up with constantly editing the files and started to write a GCODE fixup script. Over time it grew into a full-featured GCODE manipulation tool that allows me to also adjust offsets, intensity, gamma, feed rate and even reverse an image. With a bit of experimentation it becomes convenient to figure the right parameters to apply to an incoming image to make it print exactly as desired on various materials (e.g. have shades on wood without burning), as can be shown here with this well-known Relativity drawing from MC Escher that I engraved on white 5mm thick plywood (which was then cut in 6 passes at 240mm/min):

In order not to burn too much material for testing, I figured it would be nice to be able to emulate the behavior of some materials and represent the expected result on screen. That led me to write the laser-preview utility. It takes a GCODE on input and produces a PNG file based on some parameters (width, height, and diffusion ratio into nearby places). It's not that good and I'd need to redo it differently to improve the modeling of the physics of the material so that we could take into account the variations of reflection at various depths and the fact that some materials melt and that their absorption or diffusion changes with the phase change. But once you get used to validating the output on it for a given material, it does save some time and some material, which is already great.

The manipulations required to do all this and to change settings were quite long and painful, so I decided to write a PNG to GCODE converter which does what it says, it takes a PNG on input, a resolution and some argument such as fixed size, cropping on each side, setting the spindle speed and feed rate, setting the number of passes, quantizing the values, normalizing intensity and applying some gamma/soften factors. The argument are placed in the output GCODE file so that it's easy to try again changing one parameter.

This significantly eased experimentation with new materials. For example with it I managed to print this sketch of me by Frank Tizzoni on a painted metal business card (full one on the left, close-up on one part on the right) :

This was printed with a resolution of 200 microns. Experimentation shows that it's better to use a slightly larger step than the beam width (which was about 180 microns here). It's the best way to avoid over-burning some areas, this results in thin ~20 microns wide black lines that are barely visible. Achieving this requires a very fine focus adjustment, and burning a few cards with fine patterns. A macro close-up with a ruler (0 to 1 centimeter) is provided below showing this:

<

The card above was made using the "NEJE 20W" laser module, which has a much larger beam than the original EleksMaker one, but which is more powerful, and thus engraves faster. For raster mode, there is a sweet spot to find between fine level of detail and engraving speed. For example, engraving a large logo on wood or plastic wouldn't need this level of detail and would benefit from having fast passes. Engraving on very smooth and shiny paint like above achieves and extremely fine level of detail and definitely benefits from a smaller dot, even if that means using a less powerful module at a smaller speed. The card above is 5x7 centimeters and took roughly 40 minutes to engrave. With a finer dot, using a resolution of 100 microns, it would require twice as many lines, hence would take twice as much time at the same power level, or 3-4 times more with a less powerful module.

Trying a new laser module: NEJE 20W

I found a very appealing laser module that was easily adaptable to this engraver, it's the NEJE 20W module. The module is well documented and easy to attach since it's the same width as the original one (30mm).

However, while initially advertised as 20W, it's "only" 5.5W optical power (the datasheet was since updated to reflect this). Despite this, it's still very powerful. Its fan blows inside and the hot air exhaust is close to the lens, resulting in it automatically blowing the smoke out of the engraving area. This is way better than what I've done to date with my other fan experiments.

I initially had difficulties finding the connector's pinout until I figured it's printed on the solder side:

It uses a JST-PH 2mm 4 pin connector, and is provided with a cable that can be cut if needed. The main thing I've been missing on it initially was the alignment button which enables it at very low power to see a harmless dot. I figured that most other modules would suffer from the same limitation, so I decided to address this by creating the laserdot project (board + code).

The principle of this tiny module is to use a micro-controller (ATTINY13A here) to detect the pulse from the laser controller, and send a very faint one when there is none for more than a few milliseconds. This ensures that the beam is never totally off, and ends up being even more convenient than the EleksMaker module with the button since there's no risk to start with the switch on the wrong position. The module was made so as to also serve as a connector adapter, so on one side it uses the NEJE 4-pin connector and on the other side it uses the EleksMaker connector.

Once you have the ability to work with a faint beam, it becomes obvious there's an issue with this one. It's extremely wide. It's not a dot but rather a line roughly 10 times as long as it's large. Plus it's oblique, making it even harder to use in the least problematic direction.

The photos below were taken through the protection goggles with the beam sent over a ruler composed of chip footprints. The beam length was roughly 3 cm. The first one shows the beam is roughly 0.8mm x 0.1mm. With the ruler aligned to the beam direction it's easier to measure it.

This means that even at 1 cm, which is very close, the beam will still be 0.25mm wide. The problem with very short distances like this is that the smoke directly enters the lens and risks to damage it.

At 5cm distance the beam is very wide and unusable for anything requiring a bit of precision:

I sent the beam to a wall to compare the EleksMaker 2.5W module on the left and the NEJE 20W (5.5W) module on the right:

As can be seen, most of the energy is packed into an almost square area with the 2.5W one, while it's on a 10x1 area on the 5.5W one.

It's also visible that the very wide aspect-ratio beam makes it impossible to use it to cut anything small, because some directions are cut wider than others, as can be seen below trying to cut small 1cm circles:

I found that it is slower at cutting cardboard than the original laser module, because by using a much larger beam, a lot of energy is wasted. In fact cardboard can be cut with a very thin dot, which the original 2.5W module easily achieves faster.

It is possible to narrow the beam further by putting the piece to cut closer to the lens (1 cm), but as mentioned above, the smoke enters the lens and deposits dust on the glass that needs to be cleaned up regularly using isopropyl alcohol. The effect of smoke on the lens is that the spot becomes blurry:

There is another issue when trying to cut thick material, which is that the beam diverges much faster and is either unable to cut thick material at all, or manages to do it but with an oblique form, and as such it cannot be used to cut wood to make tabbed laser boxes for example:

I noticed that it has become increasingly difficult to cut acrylic with this laser. Initially I wondered what could have been the reason, maybe I initially managed to very finely adjust it and cannot do it anymore. But I came upon NEJE's Wiki for this module which explicitly mentions that modules like mine bought before last October could be subject to aging and attenuation, so I suspect it might be the issue I'm seeing. I tried to contact them but for now got no response. I'll see.

With the default lens, this laser can still cut quite fast and engrave painted metal at 1cm focus, but the cut remains quite large, even when cutting balsa wood like below:

And the large spot combined with the high power tends to inevitably make balsa catch fire. I never managed to perform a small cut of a 3D puzzle with this one. The lower the speed, the higher the risk it will catch fire. The larger the spot, the higher the risk as well. For this reason it's extremely important never to leave it cut with nobody watching, as can be seen on the failure below! I though about wetting the wood a little bit, but that doesn't work, it bends a lot, which is understandable given the nature of the wood cells as seen above in the close-up.

Also it is important not to use too short a focus to cut wood, as the beam will have a cone shape and be very large at the top, burning wood a lot:

The high power from this laser module had a side effect on my operations, which is that it managed to engrave the glass support. It's pretty visible below with the puzzle being engrave in the glass. It's still uncertain how the laser should be adjusted to do this, it seems that it only happens when there is some matter above. So probably that the concentrated light heats the smoke above the glass at a very high temperature at which the glass melts. This would deserve more experiments, because it might be useful to cut glasses to precise dimensions, for example by first painting it or just placing some duct tape on it.

Similarly it engraves brick. What's fun is that it melts and crystalises the matter beneath the surface, making it look like quartz, which thus becomes transparent and stops absorbing light. So only the first millimeter melts but it doesn't evolve at all anymore past this point, regardless of the number of extra passes:

Trying new lenses

The extra power available in the NEJE module is quite appreciable but the loss of precision is a real issue for some work. Since a lot of lenses are available on the net at different prices, I accepted to invest in a few of them to run more experiments.

First, it was visible that a number of them are not suitable due to too large or too short a focus. The one from the 2.5W module for example is not suitable for the NEJE module.

I then tried a single-element lens. I found that the spot looked small but it didn't seem to burn very well. Then I understood. It needs not to be inserted deep enough, and for its small internal aperture, a significant part of the beam is wasted on the metal parts inside. Thus what appears as a small spot is in fact just a small part of the emitted light.

I found many references to 3-element lenses, so I tried two of them (they ended up being the same so a single one ought to have been enough). These lenses have a slightly longer focus than the original one, allowing to get approximately the same spot size at a slightly larger distance. Here at 16mm I get a spot of about 0.2mm matching what I used to get at 1cm:

Interestingly the simple lens that's found in Aixys modules gives reasonably good results, in that the spot is quite sharp at a long distance (29mm) even if not very small (~0.25mm):

But it still cannot cut acrylic after several passes, and it's visible that the beam is too wide:

Overall even for cutting cardboard it's still not enough, a single pass at 400mm/mn doesn't cut while it does with the original 2.5W module:

Since it's very difficult to adjust lenses at a short focus, I figured that it's easier to invest time adjusting them at a well-defined distance and use a wedge to adjust the module on top of the piece to etch instead:

However this requires that the lens' thread is firmly secured, which is not often the case. I found that the best solution for this is to put some Teflon tape around it. It even makes it much less sensitive to vibrations and should improve overall quality:

The difference can be important, as can be seen below trying to engrave 5mm circles (the circles borders are distant 0.5mm each). The one on the right was made with visual alignment. The one on the left was made after many attempts were made on top of a 11mm wedge.

I found the DTR-G8 lens which seems to get a lot of good reviews. It's quite more expensive than others ($34 shipping included) but the results are actually impressive. The spot is by far the thinnest obtained from this laser module, around 0.1mm wide at 11mm, which is how the circles above were etched. The lens has a wide internal aperture and collects a very large part of the beam:

And it delivers a very thin and sharp spot (the pads have a 0.4mm pitch, are 0.25mm wide and spaced by 0.15mm):

Making your own module

I tried to reuse an old module I built myself from a kit. It's made of a 1W blue laser diode, a PWM driver and a heatsink+collimator lens.

It required some adaptations to fit into the actual support, but nothing too difficult:

This diode being less powerful, it's more compact and is actually what managed to deliver the thinnest spot of all modules. It is visible on this test on a PCB's top side with the concentric circles where the cut is much sharper and around 50 microns wide only from 6 centimeters (the rest around is burnt):

This module doesn't have room for a fan, though it doesn't heat too much thus it seems it would stand working like this. However it also means that the smoke is not blown and hinders the work. It really requires at least a lateral fan on the work piece to blow the smoke away.

I suspect it would constitute the best solution for very fine PCBs, working at moderate speed, and would require less adjusting time than the other ones.

Printing on cylinders

It is possible to print on cylinders such as toilet paper rolls by connecting a stepping motor to the Y axis connector. It requires adjusting the Y scale to accommodate for the cylinder size and to firmly attach the piece to be engraved to the motor:

I thought about improving this with rotating axis to place anything on them, and while looking for mechanical parts to build such a device I discovered that some vendors were already making some such inexpensive ones like Eleksmaker's Eleksrotate so I decided not to waste my time hacking with mechanics since I've never been good at this.

Final words

That's already quite a long article, showing the extent of what is possible to do with such a fun device. I'm still a bit disappointed by the NEJE 20W module because while more powerful it has a very impractical spot shape and seems mostly useless for my use cases. And with careful adjustments the original Eleksmaker module remains the most versatile one.

Every user should absolutely upgrade to GRBL version 1.1 to benefit from the laser mode. I'd like to modify the code to integrate the low power on for centering, which will allow me to drop the laserdot module.

I also found during tests that printing on top of PCBs works very well and could actually be used to print the components references. That's probably something for a future article.