Darkly Labs Community

Adding Air Assist to Emblaser 1


I’ve just started making parts to add air assist to my Emblaser 1 (A3), following on from a thread I read here:


When I started 3D printing the parts, I realised they’re designed given the user (**Juraj Kosso) **had a nozzle from a spray bottle that he’d attached to the end of his 3D print.

I didn’t have such a bottle, so I ordered a couple of different brass nozzles from China for a few dollars.

As I was waiting for them to arrive, I wondered if I needed a metal nozzle at all, and what benefit it has over just 3D printing the entire thing.  Then I realised that I’m probably putting a bunch of melted plastic right near the path of something that makes things very hot, and this is not a good idea.

But I’m still a little concerned about the brass nozzles (which have since arrived).  If they’re not perfectly aligned, won’t they reflect a bunch of light back inside the laser, possibly damaging the unit?


Finally, I don’t have any screws with which to attach the nozzle unit to the end of the laser head.  What size do I need?  The screws on the back of an old PC seemed too small, and I gather they are M3?  So are M4 screws what I’m after?




Hi Anthony,

The nozzle we use for the E2 is a silicone compound.  This material is very important as it will be close to the beam and also any flare-ups from the material. ABS or PLA will melt and potentially ignite is a very bad way.

For the air-assist to work, you need enough pressure from your pump to produce a suitable amount of air-flow at the point the laser is cutting. The more airflow & pressure your pump produces, the bigger the outlet hold can be for the nozzle.

You will probably have to experiment with this.

On our nozzle, an opening of 4-5mm is what we use for enough airflow to make the system work.


Hope that helps.

Hey Domenic,


What about using a brass nozzle on the end of a 3D print?  I have one of these:

Which fits nicely onto the end of one of these:

I modified the stl file so the length of the cylinder + the length of the nozzle adds up to 50mm, so it should sit 1.5mm from the material being cut.

My compressor is just a small airbrush unit, but I was intending that the opening would be less than 4-5mm, and more like 1-3mm.  If there’s not enough pressure coming from it I’ll look into either a larger compressor, or an aquarium pump.


I also got several 3D printer nozzles, in various sizes, as I was worried about the weight of the nozzle pictured above.  I’ve modified the stl file again so that I can try adding one of these:

To the end of it.  The diameter of the openings that I bought range from 1mm to 2.5mm.  Since I think my laser kerf width is around the 0.1mm range, if I’m lined up correctly the focused part of the beam should be passing right through the hole.


But I know when I run the laser currently (in it’s original form), with my glasses on I can see a lot of light under the guard that is not in the tiny dot that is doing the cutting.  In a brass nozzle, that’s going to be reflected around internally.  Is that something I should be worried about?

I gather it’s possible to darken brass using sulphur, the same as you can with silver - boiled eggs.  Also apparently with vinegar.  I don’t know if the tarnish will restrict the reflection of the laser enough to make a difference, though.




Definitely give this a try but I think it will be a nightmare aligning the nozzle opening up with the laser if you have an opening <2mm.

With respect to your beam bouncing around in the nozzle. If you have it all aligned, then the main beam should pass through the opening. There is always some stray light produced, and this will not cause any issue. Remember that the material you are using is not a mirror, so light hitting any of the sides will very quickly become dispersed.

The only time you would have an issue is if your nozzle was able to reflect a focused beam directly back into the laser diode. This would cause damage to the diode. For this to happen, you would need appropriately angled faces inside the nozzle.

Also, there should be very little heat generated in the nozzle if the beam is not hitting it. You will have some generated from the material, since you will be very close to it, but they should be relatively small.

I tried aligning it last night and you’re right Domenic, it was a nightmare to align it.  It’s not helped by the fact the screws from my 3D print don’t seem to line up directly with the underside of the heatsink, so it’s not even easy to get into rough alignment.  There were other issues with the 3D print - the outer diameters seem correct, but none of the inner diameters match the design, so I think my cheap printer is at fault here.

But I do have a machine that can make very high precision parts - the Emblaser!

My new thought is a 3mm MDF housing that sits around the bottom of the heatsink (for 5-10mm), has an internal plate with a hole for the lens that sits flush with the end of the heatsink, and has an opening at the end where I can insert one of the brass nozzles; probably one of the larger ones, 2mm or 2.5mm.

I can make that to the exact size I require, and it should start off being lined up quite well, since it will be sitting flush and in line with the heatsink from the get-go.

I’ll post updates when I have them.

I’ve designed something in Inkscape using my suite of custom extensions.  I input the various widths/heights/depths/diameters and my code produces the rest.  I don’t think I can attach SVG files directly here, so I’ll post a screenshot of it.

The outer border is an A5 page in Inkscape (148mm * 210mm).  The inner border my extension places around what I’ve created so I can make sure my material is aligned when I’m about to cut (I do a 50% power, 3500 mm/min single pass first).

Any lines in blue are ‘outer’ lines, which I use Cut2D’s “Outset” function by half kerf width, and red lines are ‘inner’ and I use the “Inset” function by half kerf width.

I don’t have a 3D visualisation of how it fits together, so bear with me:

Counting left to right, top to bottom, they are as follows:

1, 2 - Heatsink Brackets

3 - Nozzle bracket (1)

4 - Back Panel

5 - Front Panel (note the hole in the middle for the air hose to connect)

6 - Nozzle Bracket (2)

7, 8 - Side Panels

9 - Nozzle bracket (3)

10 - Lens bracket


The brackets tabs fit into the holes in the front, back and side panels.  The 2 Heatsink brackets go at the top, then the lens bracket, then the 3 nozzle brackets all go together.

There’s a hole in each panel between the 2 heatsink brackets go where a 3mm bolt can be screwed through to hold the cover in place against the heatsink.

The Lens bracket fits flush against the end of the heatsink, and should not actually touch the lens or lens cap at all.

This one was input with parameters for one of the smaller nozzles (not the large one I first posted above, I believe weight will be more of a factor with the larger MDF cover than the 3D print).  The nozzle screw has a diameter of 5mm at the shank, so should screw into the 5mm hole nicely.  It extends 8.25mm out, bringing the tip of the nozzle equal to the bottom of the cover.

The cover extends exactly 50mm below the end of the heatsink, and 6mm of bulk where the cover switch is, so it will completely replace the 3D printed cover that came with the Emblaser.


I’ll try to get some time tonight to fire up the Emblaser and cut this out.  I may even have some time over the long weekend to assemble it and test it out (but no guarantees, I’ve got twin 2 year olds).


What are people’s thoughts?  Folly making a cover for my laser out of wood?




Ok, so version 1 of the new module and I ran into some problems.

Firstly, the 2mm hole for the tightening screw caused a burn that I didn’t notice until after it was all done.  Next time I’ll just mark the circle and drill out the screw hole.

Secondly, my kerf width for 3mm MDF has changed.  The end result was a very tight fit, so much so that I’ll be cutting it again, I don’t think I can make it airtight without more sanding and filing than I’m willing to do.

Thirdly, I may give the nozzle an extra 0.5mm “grace” from matching the height of the cover itself.  That may be due to how tight a fit that screw was, or just a detail of how it sits, not perfectly flush.  

Finally (I hope), the nozzle doesn’t need all 3 brackets, it looks like it only needs 2.  6mm of screw-in is ample.

Here are some photos of it:




I’ve been busy over the last few weeks, redesigning parts of my code and my air assist module.

Also, I had to make a box for my partner to store her essential oils in after I got a bunch of 1.6mm carbonised bamboo.


But onwards with the air assist!

I then cut and assembled my 3rd attempt at the module.

I’ll try attaching it tonight if I get time.



Great work on the essential oils box. You should add those types of project to the ‘Project Showcase’ section.

Your new AA design looks interesting. Let us know how it goes.

I attached the AA this evening, and I’ve just finished running a cutting matrix to see how it affected the cutting of 1.85mm boxboard.  The test was from 600-1400mm/min, and between 6-9 passes. 20 tests in all.

I had to make a couple of ad-hoc adjustments to fit the air hose, but nothing drastically changed, only added. 

The cutting results were good.  All 20 tests popped out.  About half fell out cleanly as I picked up the sheet.

There was a slight amount of scorch blurs around most of the lines, but all the cuts were clean.


I’m now interested in running the same test on 1.6mm bamboo, as previously I cut my boxboard at 1000/8.  It looks like I could choose any of the 20 combinations now.

This is the rear of 1200/7.  They pretty much all look like this.