# Flow, flow, flow, your boat...

A project log for MultiBot CNC v2

A low cost 3D printed CNC that can be built with minimal tools yet is capable of great things.

David Tucker 04/15/2021 at 01:460 Comments

I have been thinking about air pressure (psi) and flow rate (cfm) and what it is we need to make all of this work.  I think I found a relationship between the two.  In our case we are trying to force air through a small orifice that seriously restricting the flow of air.  So to get more cfm we need more pressure.  Or to look at it in reverse, a given level of pressure and restriction at the nozzle will produce a fixed level of flow or cfm.

That means we can work out the ideal flow rate by pushing air through the nozzle at a given psi and measuring how long it takes for the pressure to drop in our tank.  Then with a bit of math we can work out what the flow rate (cfm) actually was.  Finally we can look at the available pumps out there and see if they can produce the flow at there peak psi we need to run the  nozzle.

What this does not do for us is work out what pressure and flow rate cuts best, that we have to work out the hard way through trial and error.

So I have a 6 gallon tank and am using a 2 mm orifice. I let the tank fill up to its cut out pressure (150 psi) then timed how long it took to drain to its cut in pressure (120 psi) at various pressure levels from 5 psi to 30.  I also measured how long it took for the tank to fill back to the cut out pressure while releasing air at the indicated psi.  At 25 psi the tank was never able to refill.  The table below summarizes my findings.

 psi release time(s) fill time(s) duty cycle (%) CFM 5 130 53 29 0.76 10 63 63 50 1.56 15 51 94 65 1.93 20 43 180 81 2.28 25 33 infinity 100 2.98 30 28 infinity 100 3.51

That gives us a good idea of how much flow we need for a given psi of pressure and a given size nozzle.  If we increase the nozzle size we will need to increase the flow rate for a given desired psi, so adjust this list based on your own setup.

Anyway we need to also work out how much pressure we actually need.  To get a sense of this I made two tests.  For the first test I cut a series of lines with the focus point set to the top of the workpiece in MDF at 50 mm/min and 80% power and varied the PSI from my air assist.  I then used a 0.002mm feeler gauge to measure the depth of the cut by inserting into the cut, marking the top with a pen, then measuring the depth with calipers.

 psi cut depth (mm) improvement (%) 0 0.8 (with heavy scorching) 100% 5 1.7 213% 10 1.7 213% 15 2.2 275% 20 2.0 (seems off) 250% 25 2.56 320% 30 2.96 370%

I'm not sure what happened with the 20 psi measurement, it seems off. However we can see that any air is miles ahead of no air, and more air gives better results.  This data is really rough, my measurement method was not the best, however if you plot it out it appears fairly linear.  What is interesting is that we get deeper cuts with more air without the kerf (width of the cut) getting wider.

For the second test I set the pressure to 15 psi, power to 80 and speed to 50 mm/min (as above).  Then I did multiple passed lowering the initial focus below the surface.

 focus depth below top (mm) cut depth (mm) improvement (%) 0 1.8 100 -1 2.3 128 -2 2.9 161 -3 2.8 156 -4 2.7 150

You can see from this data that there is a small improvement at first from lowering the laser but it has minimum benefit.  Eventually the focus at the top becomes so bad that we end up loosing power.  I'm sure that more air can help here.  And of course multiple passes over the same cut are different and should work much better.

The farther I lower the focus depth the wider the kerf gets, using air pressure is a better way to increase the cut depth.  With that said it seems like for now that I will stick with 15 psi of pressure and focus 1-2 mm below the surface or less than 1/2 the depth of the object we are cutting, whatever is smaller.  That allows my compressor to run at a reasonable duty cycle that won't wear it out and yet it gives me a reasonable improvement in cut quality and depth.  It would be nice to run at 30 psi but I would need a much larger compressor that is capable of 5-6 or more cfm at high pressures.

It looks to me like any air pump that can hit 4-5 psi and 0.75 cfm will get the job done, however it will not have nearly as much benefit for cutting as a stronger full sized air compressor can have.  Going from 5 psi to 15 psi had a 50% increase in cutting depth.  On the other hand for engraving we want to just put out the fire and not really remove any material. In that case I suspect that a (large) aquarium pump is more than good enough.

In fact my hypothesis is that any air flow above 0.25 cfm is probably good enough for engraving, including a small radial fan mounted on the laser head and blowing across the work like a part cooling fan on a 3D printer.  However that is an experiment for another day.