Acrylic on LR2

Checks out.

I cut a ton of acrylic shelves for a friend. He polishes the edges with mapp gas. I have the same experience as you with propane.

1 Like

No, I haven’t tried them on acrylic, I think I just surprised myself that I haven’t managed to break one cutting wood (yet). :grin:

I think you’re correct, they do look like 3 flutes. They list plastics as one of their applications but I know all too well how different the promise can be from reality.

1 Like

Happy to hear this!
will need to look for mapp gas then. Do you have an idea how sharp the edges remain after this polish?

:smiley: I totally understand, still waiting for that moment to happen too :slight_smile:

yeah, as most posts on the web point to a single flute, I believe that a 3 flute might work but is far from perfect. So will keep looking for dedicated Acrylic bits.

Does anyone have some insights on optimizing jobs in estlcam so I could put multiple squares next to each other in 1 job?

I haven’t delved too deeply into it but doesn’t the ‘tile’ function allow that?

1 Like

I should look into it, but isn´t this just for duplicating the part? I would like to cut the overlapping sides in 1 go, thus saving machine time too.

As I said, I haven’t really looked too deeply but if you’re cutting duplicate parts I’d guess you could adjust the spacing for one pass to cut the sides of two parts? Might be self defeating though if conventional vs. climb milling has a large affect on finish when cutting acrylic.

I get your idea but not sure if estlcam is able to just cut that side once. Because if I select parts, it will try to cut all sides. But will experiment with it.

It kind of goes against the idea of roughing and finishing cuts. I am sure you can do it, but it would be a pretty manual process.

1 Like

A few problems with the idea…

  1. This would mean that the one edge of the part would be “climb” milling, while the corresponding edge against another would be “conventional” milling, and Estlcam doesn’t like to mix those up. (Neither should we, as they have different characteristics in the finished piece.)

  2. The reason for finishing passes is that the tool path isn’t precisely predictable under load. We could reduce the load via DOC and speed so that a finish pass isn’t necessary, but I think it would be a very rare event where you would end up with a faster cut sharing a side that way. The idea of the finishing pass is that the load on the tool is very low, so the tool path is precise and predictable and you end up with the best precision for the final part possible.

  3. Even if you only need to cut 3 sides of a rectangle, the tool still needs to traverse the 4th side for every pass (See above for climb versus conventional milling) and while this might save a few seconds for a “rapid” pass versus a milling pass, it’s really not that much.

I’ve looked at trying to eliminate 1 or 2 sides for a job where I wanted to use the original material edge, and in the end, I ended up just taking a small amount of material off, because I wanted the accuracy, or else I did the interior features only and cut the outline on something else, like the table saw.


Yeah. It all depends on the desired precision. It won’t be that far off.

I have pretty good luck with some .5mm ball-end bits.

The top engravings came out clean with no “fuzz” at the edges.

1 Like

Forgot all about the engraving question. Depending on the carve, drag bits might be a solution. I use one on acrylic too. Takes several passes, but looks nice.

1 Like

weird, what I use is a similar bit. But I see you go nearly twice as fast. Will give that a try too.

Good idea! It´s just some text so that might be doable. Do you use a diamond or steel engraving bit?

I understand what you´re saying Dan,
However if I have a look at both sides of the current trochoidal cut, the result looks the same (by eye). So although two methods of cutting might be in place, it isn´t any other then how it works today - no? It cuts a piece and by doing that you have the same line on the other side.

I wouldn´t mind to do the rough cut first with only 1 pass for the aligned sides, then do a finishing pass on all sides. That would shorten my time with approx. 1/6th.
That´s roughly 40 pieces x 4 minutes = 160 * 0.84 or: 134 minutes instead :slight_smile:

I suppose that what you could do is define a second tool with the same specifications, set all of the trochoidal stuff to zeros (disable it) and the Z depth to be full pass, set up the finish pass speed, then cut the outline with the other tool…

This would take a lot of care witht he CAM, you’d have to use “manual toolpath” for the trochoidal cuts for ines that aren’t complete outlines.

1 Like

I’m not home to look at speed and feed, and I was never able to get a glass look edge like on that video. I do have a single flute bit like that and I will try if this weekend if I have the time.
The results I get, you can see on my LR3 Build thread, which is very similar to yours.
If I’m doing a part that needs the sharp and glass looking edge, I usually wet sand it starting with 400 grit paper all the way up to 3000 grit and a very, very, very light flame polish with MAP gas.


So you´re doing both then? sanding plus gas polishing?
Do you use a polishing disc for the sanding or rather sanding paper?

I know it´s been a while since my last post on this topic, but I´ve been busy :slight_smile: though I got a small update for those interested in Acrylic cutting.

due to the above advice, I´ve added a finishing pass as an experiment. I landed using the following parameters for now;
3mm 1 Flute Upcut bit
6mm Z+
60° plunge rate
20mm/s feedrate
2mm plunge feedrate
6mm Z+ finishing pass
10mm/s feedrate finishing pass
2mm plunge feedrate finishing pass
30% Stepover
20% Trochoidal stepover
50% Trochoidal width
0,05mm Trochoidal oscillation

On my first examples, setting the Katsu Router on speed setting 2, I got exactly the same results as my previous setting without finishing pass. So that´s with small random cracks in the edges (tear out I belive it might be called?).

Throttling down the speed setting to setting 1, these cracks have been eliminated by 99%. I believe it´s not perfect because of small machine movements. It seems hard for the machine to follow the same path, especially visible when carving text. Wonder if an LR3 might be better at this.

The edge result is like 20% more transparent with my new settings then before, although I must say it depends on what edge you´re looking at. All 4 sides have a different look which is odd for a beginner like me:)

It´s on my todo list to buy a specific acrylic 1 flute upcut bit. So hope to test this in few weeks to see the difference between dedicated acrylic cutters and regular single cutters, but first I need to find a 3mm collet for my katsu, so the new flute will fit the machine as the shank is 3 or 6mm not 3,175.

I’m sorry I couldn’t do the test and will be another 2 weeks or so before I will be able to do anything as I’m in Brazil until July 2nd.
Yes I do both and I prefer sand paper, polishing disc can generate heat and is hard to control the angle of the “grinder” when polishing it.

1 Like

no worries, I´ve got time - enjoy your holiday :slight_smile:

Hi, I utilize my LR2 to primarily cut acrylic LED signage and have encountered a lot of the same challenges you are experiencing.

My signs typically include a routed channel / groove in which I inlay neon leds. Depending on the leds, I need I 5-7mm wide channel and I cut that channel a little more than half way into the 6mm acrylic. And once the channels have been carved, I move onto the profile cut.

For my channel cuts I like to use a single flute upcut bit. I initially tried trach cutting, but it was annoyingly slow, and today I typically just grab whatever single flute endmill I need for the channel (I keep a good selection of single flutes handy). If I need a 5/32" channel, I grab that bit and setup estlcam for an egrave an set it to 3.5mm deep, 1 single depth pass.

For my profile cuts, I found switching down to a 1/8" single flute worked the best. I seem to be able to push the smaller bit through the 6mm acrylic much faster than a 1/4 end mill. I typically do 2-3 passes (depending on the shape), with a single finishing pass.

I found my best results by slowing my router down to near its lowest speed setting (1.25-1.75ish on my Dewault, my LR3 will use the Makita for better selection of RPMs). Any faster and I notice that the acrylic gets a bit gummy and wants to fuse back together.

I also start to get a bit nervous of the outcome when I try running past 8-10mm/s. This might just be me or my build, but I’ve had numerous instances where steps were missed leading to a lot of scrap acrylic over here.

For the actual engraving of text onto the acrylic, I haven’t personally used it, but I’ve seen many videos where a spring loaded diamond engraving drag bit is used. The spring loaded element helps keep it in contact with the material. The router is just holding the bit while the machine moves, router not powered on for this.

For finishing the edges, I ended up buying a belt / disk sander. I also start somewhere around 400 and keep increasing (2000+). I then use a cutting compound and work that in. This is a lot of work, flame polishing is WAY quicker and looks beautiful…when it works. My flame polishing results are a mixed bag. When it works, its beautiful, but propane cant do it, the fake mapp gas equivalent provides mixed results. Shy of getting a better torch & fuel set, consistent good flame polished results still out of reach for me. Protect your front and back surfaces from the wrap around heat.

Thanks for sharing your efforts and results, it was a good read and has me wondering if I can go push more speed. I may have to go revist the troch method to see if there speed to be had there.


Happy to share my results, it might help and give a “jump start” for anyone willing to cut acrylic :slight_smile: plus I learn a lot of your input too!

I was surprised to read that you´re doing just a regular cut and no trochoidal approach after reading a lot of issues by doing so. It´s my understanding that this way of cutting is mainly focused on keep the bit cool, so that it doesn´t melt the piece…

Did I understand correctly, that your regular job is like 2mm DOC and feedrate of 10mm/s? Might try and see what that result gives compared to mine (and taking total job time into account).

As it looks for now, you might want to try trochoidal too, as my feedrates are much faster then yours, plus there are less loops to do so it might be a faster job for you, no? I don´t have any issues of missed steps, I only notice a small shifted text while engraving.

The spring loaded bit is on my short list. Only doubting if I really need a diamond version, or that a metal one would work too because of the costs involved (plus, setting op Z with a diamond bit will become difficult with a touchplate :P).
But for now I am waiting on delivery of my dedicated Acrylic bit samples, to see what´s the result after using those. Hope to have them in the coming weeks.

With regards to the flame polishing, just like you, my results are really mixed, so for now that´s not the way I´m going. I believe I´ll just buy a belt sander too.