Laser Flare Ups

I helped my dad build a Heathkit TV when I was 5 years old. I remember taking boxes of tubes to the drug store to test them and get replacements.

2 Likes

Going back to discussing the p’s and q’s of air assist, I ran a basic test to compare low flow vs high flow in cutting 1/2" birch ply. It was pretty clear that the higher flows can help:


The first image shows the untouched cuts after the test. The top row of 3 squares was cut with high flow (~40-60lpm), and the one just below it was cut with low flow (<10lpm). Same gcode was used for both (feedrates on cubes, power and z height on top). The second image shows what happened after giving the cubes a light tap with my fingertip (think I transposed the 90 blocks though).

So the aquarium pumps can cut 1/2 ply, but notice the diff in the kerf at the surface. The low flow looks much wider. The high pressure appears overall sharper, thinner, and cleaner… and most importantly it dropped all 3 cubes vs just the slowest one.

This isn’t a fully controlled experiment… first off it’s natural wood ply. My flowmeter is not calibrated, and I have not measured loaded flow of the aquarium pump (I suspect it is close to 10 though… very little pressure drop on the 5’ 1/4ID line). I could run this again using a known 10lpm via my compressor, but I think the results get the point across; more airflow = deeper cleaner cuts. I can’t say whether it’s worth the effort and extra noise for everyone. It was easy for me already having the compressor/solenoid/needle for a mister (aluminum milling). For someone without a decent compressor, that would significantly increase budget.

[edit: I’m now curious where the point of diminishing returns kicks in on this. At some point we can start to actually cut the ply with airflow alone lol. It would be cool if we had some data to show a sweet spot where hobby diodes should be.

Also, for anyone interested, here is the silent compressor I have used for most mpcnc operations:

I actually purchased a smaller one and returned it. The one linked above has an alum tank; so very light and rust free! It’s also powerful enough for light/medium construction work. It can’t do roofing/siding/fencing, but just about anything else on my jobsites (trim, framing, etc). I can recommend it for those looking to get a smaller compressor these days. Now… it is also very much falsely advertised mind you! No way it puts out anywhere close to 2.2scfm@90… closer to 1scfm@90. Even my 60lb Hitachi 15A 4gal oil/cast iron lunker can barely do 3.0@90, and that is a very respected compressor in the industry (and it actually does meet it’s advertised ratings).]

1 Like

Last night I did spent several hours cutting some 1/2" Sande Plywood to make some simple speaker boxes. I thought this type of plywood might do well laser cutting, since it’s meant for applications where you can’t have voids after cutting (routed edge stuff for example). I figure that means more consistent core, so less chances for a flare up (which seem to happen more near voids in plywood). So far so good… no flare ups unless I go too slow.

Speed I’ve found is pretty much the key factor in flare ups (besides voids, but no controlling those). Of course faster cuts thinner, but it also gives less time for heating that leads to flare ups. I saw a video on YT that showed lots of cutting data using the Neje 40W. The guy did everything at 300mm/min 75% power, and just added passes as needed. I was cutting thicker stuff in the 100-150 range before, so I ran some tests between 200-300 (at 100% power). You get mostly reliable flare up free cuts at 200, but finer details where the laser doubles back on itself could cause problems. At 300 though, even smaller details rarely flare up.

Also, as my ears started to ache from my 2.5scfm pump, in the middle of a cut I swapped over to the amazon aquarium pump (also it was starting to get late, had about 2hrs of cuts left to do, and didn’t want to annoy my neighbors). The result was I had to add one or 2 more passes, vs 6passes @200 that consistently cut with big air (30lpm). Not bad I’d say, and well worth the noise savings. I did not think about the fact that I had to endure the noise with no escape, if I was to be close enough to see a fire as soon as it starts anyways. The aquarium pumps make babysitting the cut way more practical, and those extra couple passes IMHO are well worth it.

Also I’ll just throw this out there… I have found a few YT videos of folks who say that sometimes using high flow from a big air compressor can actually do worse than a big aquarium pump. These claims were made with the same equipment we’re discussing here… big loud compressor running 90psi to output over 30lpm, vs the amazon aquarium 10lpm pump. Of course, those claims were not backed with science, or even conjecture.

I’ll take a shot here… I can definitely hear oscillations coming out of my air assist nozzle, and feel vibrations in the tube leading up to the nozzle. Given a similar nominal pressure delta, I think adding a vibration component to the airflow would help to knock off more material vs steady state airflow. The waves from vibrations can directly work chunks of burned material loose. Also vibrations increase turbulence, which reduces boundary layer and can drastically increase shear near the surface. With that idea in mind, I now wonder if I rigged up both my compressors such that I get high flow from my tank, and an added vibration component at the nozzle from the diaphragm pump… best of both worlds… hmm. I think the aquarium pump would overheat if I just tee’d it… probably have to have the aquarium pump air recirculate in the line just before it goes to the cutter (so it has flow to cool, but effectively is not flowing to the nozzle, just transfering vibes and heat as it’s flow smashes into the smooth flow coming from the tank).

1 Like

Depending on the compressor and air distribution system, there may be oil droplets suspended in the air stream, or significant amounts of moisture vapor, either of which could have detrimental effects on cut quality and/or end up on the finished piece.

A good point-of-use air/oil filter should be able to deal with the issue, but adds complexity, cost, and efficiency losses to the compressed air system.

1 Like

Can I just point out that cutting 1/2" plywood with a diode used to be thought of as impossible (even just a few years ago). Just crazy.

1 Like

@ttraband I am running a fairly large air filter/dryer just before my solenoid valve; put it there a while back when I first added flood/mist… to avoid getting oil on fancy wood. I’ve since changed my vacuum shoe such that flood is not longer required for wood (unless the shoe gets in the way… rare). I have read similar things about moisture etc increasing power needs; makes total sense to me, plus contaminants could deposit on the lens.

@jeffeb3 I have to say so far I’m very happy with the purchase of that laser. I think 18mm may be a stretch, but it really depends on the species of wood etc. I’m just happy that I’m able to now produce more structurally significant parts with the laser, now that 1/2 sande ply is reliable for me.

I’ll post about my simple speaker boxes in a new thread when they’re done. I have a few days of finishing work planned before they’re ready for photos. I have to be careful with the poly so I don’t drag any soot onto the wood surfaces, and careful with sanding so I don’t dig too deep in black areas, since I’m trying to preserve the black for contrast/design.

3 Likes

Those are some really good results. I’ll have to do some more test cuts and try to get some better results.

My air compressor pushes out over 90psi. I use a small air brush regulator at the laser to pull it back to ~15psi. The air compressor is a large one with an 80 gallon tank, oiled pump, and a 3hp 220v motor. When cutting with the laser it only runs for a bit every few minutes to top the tank back up to full.

2 Likes

Thanks, yeah if you have a Neje 40W and haven’t done well on 1/2" stuff, then sande ply is a good place to start.

I don’t read pressure after my needle valve, but it is usually very low anyways. I got the volume meter and needle valve just to ‘do it the right way’. 30lpm will always result in the same 30lpm out of the nozzle regardless of what’s going on. OTOH 90psi at the guage may give 60lpm, but drop to 40lpm if I say add an elbow after that guage, or lengthen the from the guage. A needle also makes it a lot easier to control flow rate, especially when you get down to very low flows. I like to use 5lpm while just etching, as that kept the compressor from cycling as often vs 10lpm or more. 5lpm is pretty much impossible to get consistently without a needle valve.

The only thing after my small regulator at the engraver is the air line going to the laser head, but that’s something to think about.

I use a needle valve on the co2 going to my aquarium for consistency for the same reason you’re using one. I’ve thought about putting a flow meter on it instead of using the ‘bubble count’ method.