Supersonic Air Assist

[Charles Jones]

Esteemed laser operators,
I've been mulling over this topic for a few weeks and wanted to reach out to see if any others have considered it or maybe even tried it.

I wanted to try supersonic Air Assist to maximize cutting depth/speed/quality in plywood. It probably wouldn't work well with acrylic, but I cut lots of 1/2" BB ply and anything I could do to accelerate the process would be awesome. So with that intro aside, here goes:

I have been interested for some time in the improvements in cutting ability that users have achieved by using high pressure clean and dry shop air. Most posts mentioned that they achieved the best gains up to around 30psi on the compressor regulator and no improvement beyond that. I started thinking about the physics of compressible gases and that started to make sense.

Compressible gases such as air will accelerate in a nozzle up to a point. They eventually "choke" in the narrowest point or throat once the local flow velocity reaches mach 1. It doesn't matter how much pressure you put behind it, you will never increase the exit velocity from a standard converging nozzle by increasing the supply side pressure. In other words, once you reach the pressure ratio between upstream (above the nozzle) and downstream (normal atmospheric pressure) that chokes the flow, no amount of increased pressure will increase the velocity at the throat. This pressure ratio for air is 1.893.

https://en.wikipedia.org/wiki/Choked_flow?wprov=sfla1

So... Any pressure above 27.8 psia (assuming 14.7psia atmospheric pressure) will choke the flow and maximize your air Assist velocity. 27.8psia (absolute pressure relative to zero) is equivalent to 13.1psig (gauge pressure relative to atmospheric) which is what the regulator on most shop compressors will read. In other words, any compressor that delivers 13.1 psi to the nozzle is maxed out. Doesn't matter how much extra pressure you dump behind it. This makes sense given that most users supply 30psi measured while the system isn't flowing and then once it is flowing, losses in the tubing etc probably result in something like 13.1 psi in the nozzle. Hence the maximum effect around 30 psig.

The caveat is that if you have a converging/diverging or CD nozzle (also called a de Laval nozzle) you CAN get faster flow. This is how rocket engines and jet afterburners work.

https://en.wikipedia.org/wiki/De_Laval_ ... prov=sfla1

With this type of nozzle, you can't increase the speed of the gas at the throat, but you can increase speed BELOW the throat above Sonic velocity by having a divergent section downstream of the throat.

This is where I got interested. Every laser nozzle I have ever seen is a simple converging nozzle. No divergent "Bell" on the end. If an appropriately shaped laser nozzle was used, then pressure ratios above 1.893 (or above 13.1 psig in the nozzle) WOULD achieve greater exit velocity, and hence would have potential for increased air assist effectiveness.

For example, a compressor that could deliver a full 30 psig to the nozzle while flowing would have a pressure ratio of just over 3.0. this would theoretically achieve supersonic flow and I am hoping... Increased cutting capability! 30 psig is low enough to not burst laser system components, but potentially have an increased cutting effect with an appropriate nozzle.

I found a nozzle online that might be easily attached to a laser system.

https://us.misumi-ec.com/vona2/detail/1 ... wyg_area_0

It has a 2.7 mm orfice which would consume around 6.1 SCFM.

https://www.tlv.com/global/US/calculato ... ifice.html

Anyway, forgive the long post, but curious if any others have considered this option.

[Charles Jones]

Another point to make is it's possible that 30 psig achieved the best cutting effect because there is simply nowhere better to go. High pressure mach 1 air assist already achieved the greatest improvements, and boosting the air assist velocity even higher wouldn't actually achieve any improvement, despite the increased flow velocity.

[Gene Uselman]

I think that this is a good area to explore- at present I am using a nozzle that has been bored out to double the normal size as I had been trying to get my laser beam tamed down and consistent over the entire bed. My 130w RECI tubes especially had very large beams and I struggled with them for several weeks before giving up. The SPT tube had a nice small beam and the current GSI tube seems very nice. I have also seen a separate air tube/nozzle which I would like to experiment with. My shop air is 100+psi but I don't have that much pressure to the machine [yet]. I am also using a 2.5" GaAs lens which has way better cutting potential for .5/.75 wood and intend to get a 3.0" GaAs lens for more experiments.

[Doug Fisher]

An interesting discussion. I guess my question is, and you seem to have alluded to in your post, whether supersonic airflow is actually needed/of benefit. At some pressure level, you will be blowing all of the smoke out of the way all the way through the cut in the material.

[Charles Jones]

Doug,
Agreed. I was mulling over the mechanisms that could improve cutting depth and speed. Removing smoke is one, and I agree that eventually you'll max out the benefit of that particular function. My hope is that there is possibly some heat transfer, or exothermic benefit to be gained by making the gases flow more quickly. Maybe the superheated exhaust gas could actually speed the cut if pushed fast enough? Or maybe the increased flow could actually support combustion, like oxygen assist does in a metal cutting laser. Or possibly, the higher air assist velocity would allow slower cut speeds for cut thicker materials, without sacrificing edge quality because of excessive charing. All interesting hypotheses.

I suspect that the greatest gains would be on cut quality and reducing the char/ambering of the cut edge. This isn't a big deal if you like the look of the browned edge, but the browned edge doesn't support glue as well as a saw cut edge, which is also a concern of mine.

[Charles Jones]

I have been searching for a tapered drill bit to help me turn my own CD nozzle on the lathe, but couldn't find anything until now. I'm going to give these a try when I get a chance. I'm not sure that it will make a nice smooth Bell nozzle shape, but they're cheap and worth a try once I get my compressor hooked up.

[Frank Small]

The burrs might be good for roughing out but might may leave a pretty rough finish. I saw these stepless bits that may make a better finish.

I’m sure you would get the best finish if you used a boring bar on your compound set to the desired angle.

04C317BC-30B3-4D09-A0F6-FB32FA25C202.jpeg

Frank

[Charles Jones]

I looked at that drill bit set and saw that the smallest diameter on the smallest bit is 3mm. I'm trying to shoot for a 2mm throat diameter because as you can imagine the SCFM difference between a 2mm throat and a 3mm throat are enormous. In my machine I can fairly consistently get the laser spot to exit a 2mm hole in the nozzle without hitting the sides. I'd go smaller, but not sure we can avoid clipping the beam, especially because adding a divergent section below the throat implies a few mm more nozzle below, which could hit the work, OR moving the throat higher up in the converging beam, which would mean a wider, more out of focus beam diameter. I probably need to draft something up before cutting metal to ensure we have the necessary clearance.

I did find the attached drill set, which LOOKS smaller, but not positive what the dimensions are. I submitted a question on Amazon hoping to find out.

[Charles Jones]

Also, I would turn this on the lathe, but I just don't have a boring bar that small... Back to Amazon and maybe MSC!

[Charles Jones]

This should work.