CNC etching leaves thick residue on etched area


I’m CNC etching cast acrylic with a V-bit at 12K rpm and .2 mm depth. The etches come out gorgeous with one major issue that I can’t seem to figure out how to avoid or at least handle without damaging my piece. The acrylic shavings from the V-bit etch harden and stick to the etched area and are REALLY hard to clear off.

I have tried using a soft bristle scrub brush but that doesn’t clear the area completely and leaves small micro scratches that are visible when placed in the LED base.

I have tried using a small flat head screw driver to chip away at the etched area which works however the smallest scrape outside the etched area will scratch the acrylic. It’s so hard to clean the entire piece without slipping once.

I’m not sure what to do, it’s so frustrating as I’ve spent over a hundred hours perfecting this craft and I’m stuck at the end with this hard residue that I can’t get off easily.

Attached is a picture showing some of the hardened residue left after etching. It is there whether I try to clean right after etching has completed. I had just tried to clean it off in water so that’s why it looks like it’s darker and lighter in some areas.

Any help would be greatly appreciated.

I also have a problem with melted chips sticking to the rotating end mill during acrylic processing.

It happens even when I change the processing conditions.

I haven’t tried it yet, but it may be better to cool it down by forcibly blowing air on it.

As a countermeasure, I repeatedly use a hand luthier to remove the dust after machining.

Yes, I do.
I’m trying without being told.
Thank you.

Machining acrylic is best done with a specific acrylic bit, with cutting geometry designed to minimize heat generation. Examples here, also available on Amazon for cheap (your mileage may vary): https://www.amanatool.com/products/cnc-router-bits/plastic-cutting-cnc-router-bits.html

Regarding feedrate, increasing feedrate is a common way to minimize heat generation at the cutting edge.

Increasing feedrate will increase chip load, resulting in:
a) less rubbing and more cutting if the chip load is too light
b) larger chips to carry away more heat.

The snapmaker, with its limited spindle power of a mere 50W, cannot run bits anywhere near their optimal chip load, which will result in significant rubbing as the cutting edge dulls after initial use.

In this particular instance I would think b) is unlikely to be a significant beneficial, but I can’t formulate a reason for this, just a feeling related to acrylic’s thermal characteristics.

a) is likely to be a significant benefit but only if the spindle wattage is not limiting the maximum chip load at the DoC used.


In summary:

  1. Use a plastic cutting bit for better performance
  2. Increase feedrate as close as possible to the recommended feeds and speeds for the cutter being used.
    a) Because of the limited spindle power and other physical limitations the ideal speed may not be achievable in practice.
    b) Limitations on the depth of cut (both axial and radial) may be necessary to prevent overloading the spindle.

For an example walkthrough of feeds and speeds from a “by the book” perspective, to illustrate how quickly you can get ridiculous numbers that are not achievable on this machine.

From here (Chipload Calculator | GDP Tooling), using the assumptions of a 12K spindle speed, 2 flute 1/8" cutter, in soft plastic (targeting the minimum acceptable 0.003"/tooth)
That is achievable with a feedrate of 2000mm/min, approximately as fast as the machine is physically capable of moving. Due to the vibrations induced from moving so quickly while machining this would likely result in failure.

At a more modest 600mm/min that is only a chip load of 0.001", which is likely to generate significant rubbing, resulting in melting through the workpiece instead of cutting as the cutter is trying to remove only 1/1000 of an inch (0.025mm) per tooth.


That company has a tooling guide here: https://cdn.dynamixse.com/blog/wp-content/uploads/sites/3/2018/06/CNC-Tooling-Guide-by-GDP-Tools.pdf

Page 47 describes a method to find the optimal chip load:

FINDING YOUR OPTIMUM CHIP LOAD
When the chip is too small, the cutting action will generate heat in the cut
and will cause the cutting edges of the tool to deteriorate prematurely. A
larger generated chip within the reference range for a particular material
will achieve the longest tool life.

You can find your optimum chip load, which will allow you to maximize
productivity and get the best tool life and lowest cost per panel, if you
follow the steps below:

Start using the recommended chip load and slowly increase your feed rate
until the finish quality becomes unacceptable. Then slowly decrease feed
rate again until desired finish is restored. Make note of your feed rate.
Next, decrease the machine RPM’s until the finish deteriorates. Once that
occurs, increase RPM’s until finish is once again restored. At this point,
you have found the “sweet spot”.


While this is good practical machining advice, it unfortunately does not translate well for new users on a snapmaker. This is standard practice, however, for new materials and new tools.

1 Like

I think I see where the confusion arises.

Greater rate of travel would make a difference if the travel were a significant source of heat in this scenario. From what I understand, it isn’t—pretty much all the heat is coming from the friction caused by the rotation of the bit. So a higher travel rate may increase the absolute temperature of the bit slightly, but it’s already so hot that reducing the amount of time the bit spends in contact with the plastic it’s milling makes more of a difference in terms of total heat transfer.

You’re right that there is presumably a break-even point where friction from travel becomes the most significant heat source, but I’d bet it occurs at such a low spindle speed/high travel speed that milling with those settings isn’t realistically possible.

I’m pretty sure this is caused by the toolpath used. Looks like the standard Luban cutting approach. Just go from the left to the right. And not a toolpath that nicely follows the contours. So the jagging you’re seeing is not because of the reasons you describe, but just the diameter of the milling bit. (little overlap between toolpaths)

But @Borinkito will have to give more information on that. (toolpath & software used)

I personally would not use a v-bit for milling acrylic either. Again, much more force, rubbing, thus heat build up in an tiny little tip. At least use a flat end mill or better a specialized one @brent113 refers to.

And with regard the heat. I do agree with most other comments that going slower will only cause more melting of the acrylic. So use a bit with smaller number of flutes, higher travel speed and less depth of cut. Using multiple passes if needed. A V-bit will probably never result in a nice surface finish for something like this.

@Borinkito, besides all the other recommendations with regarding to choice of bits, workspeed etc. I would also recommend to try some kind of dust collection. There are some threads about a vaccuum adapter or just manually babysit it and vacuum everything away.

It will have two advantages: the chips/dust will be sucked away, which means it won’t get trapped between the milling bit and your stock, causing less friction and again less heat, so the dust and chips can’t melt anymore.

Constantly sucking air away will cause airflow which again will help with keeping everything cool and this not melting as fast.

(But milling bit and work speed will have the biggest impact I think).

1 Like

Wow. THANK YOU ALL for this really informative and animated conversation. Lots of great information for me to soak in and try.

To answer @brvdboss, yes just general Luban generated tool path, left to right…I am not that advanced to use Fusion360 yet. I tried on something and I was spinning my wheels so I gave up on it. I should go back and try again though.

I’ve been thinking about trying to hook up a vacuum for months but haven’t pulled the trigger. I just vacuum it out after every run but having a vacuum connected to the head would be nice.

In terms of workspeed right now I’m at 500mm/sec. I guess I’ll try and up that too 600 and see what I get. I’ll play around with that and use a flat mill as well to see what kind of results that gives me.

Appreciate the feedback! Happy making!

BTW…that picture was from an old project, this was the final result using the same settings referenced in this thread. I was happy with the results, just the cleaning to get the chips out is what I’m trying to avoid for future projects. Thanks again!

1 Like