Cutting acrylic for inlay work - getting a 0.019" kerf with a 2" lens?

[Mike Lassiter]

I've been working on some ideas, and running into "trouble".:confused:
I am working with "1/8" acrylic from Delvies, which measures 0.112" thick just the acrylic and total thickness for paper on both sides is 0.127" thick.

Trying to work out doing some inlay where I cut something out of say a white sheet, then cut out maybe red to fill the spaces cut out of the white.

I understood the 2" lens our ULS laser has approximately a 0.005" spot diameter. For good tight fit I have read to contour about 0.002" outward on filler pieces so they would be theoretically 0.004" bigger to allow for kerf that laser cut out in base material. Yet I kept having very sloppy fit. So I created a circle in Corel Draw 1.0" in diameter and cut it out. Focused material before cutting and measured the circle cut out and the hole left. The circle cut out measured 0.989" diameter more or less (it would vary slightly when rotated in caliper) which is 0.011" smaller than the 1.0" circle and the outer opening measured 1.008" for a total kerf of 0.019"

Can anybody offer insight into this? If the laser beam diameter is 0.005" how am I getting a 0.019" kerf? I have used air assist, but removed the nozzle and no air and had the same results. Getting same results anywhere on table. I have been using a 12x24" piece that I already cut some pieces out of, and using the areas around the cutouts. Some on right side and some on left side. So this puts cutting on left edge and center of my table. This just doesn't seem right based on everything I have read from others here and elsewhere.

Any thoughts?
Thanks

[Joe Hillmann]

That .005 is for one kerf. You have two, one on each side of the circle. So now you just have to figure out where the other .009 is coming from. Are you over powering the material? That could cause the kerf to be wider. Being out of focus could be an issue and the material being 1/8 thick could also be a problem. Have you ever actually measured the diameter of your beam?

[Richard Rumancik]

Or to put it another way, your kerf is (1.008-.989)/2 = .0095". This is almost 2x larger than it should be but not as bad as you suggested.

My first thought was that the lens was installed upside down, but since you have a ULS I don't think this is possible - the lens and mirror are mounted to a plate, correct? Are you using a probe to focus or are you using an autofocus function?

As a quick-and-dirty test, maybe try to draw a circle "in focus" and then raise the table 1mm or so and draw another; do the same 1mm below focus. The "in focus" should be the thinnest line - if it is not, something is wrong with the focusing.

Can't think of any optics problem that would cause what you are seeing.

[Mike Null]

I believe it is a wrong assumption that you can determine kerf size by spot size. If there were no extraneous factors then that might be possible but in this case the material poses the first issue. The material, during the cutting process, burns and melts, a situation that you are unlikely to be able to scientifically calibrate. The focus can be another factor as can the depth of the cut. It is also possible to overpower the material thus enlarging the kerf even more.

I believe your solution is the old fashioned trial and error method. In this case I would enlarge the drawing until I achieved the desired result. It is also my practice to make the cut from the back side.

[Mike Lassiter]

I guess I got caught up in the total difference thinking that was the kerf, but as pointed out I needed to remember to divide by 2; which I overlooked. That makes me feel better - now not as bad as I was perceiving it to be.
I worked on a few things last night as I posted this thread, and I noticed something a little odd.
Cutting red or orange it seems the kerf is wider than cutting turquoise.
Something else I checked, and present for thought. I measured the focus tool that goes with the 2" lens and from the bottom of it to the edge of the sloped spot that the bottom of the lens is measured from measured 1.95x". This is the tool that came with the lens. So I am asking myself if this should be 2" considering it is for the 2" lens. I sort of figured each tool was set for a specific lens, not any measuring tool set to work with any lens.

Is the measuring tool suppose to be specifically set for a specific lens, or work with any lens in the focus length it is set for (say any 2" lens)?

[Scott Challoner]

I agree with Mike. Trial and error will probably be your best bet. It's been my experience that some colors absorb the laser energy differently than others. That's why you may have a larger heat affected zone with some colors. Some say that clear acrylic (and possibly certain colors) actually acts as a waveguide. That is, the polished edges after cutting help to focus the beam into the cut. You may also consider doing a beam alignment. You're beam can be oval shaped if not aligned correctly. I noticed a big improvement after aligning mine last year. You should be able to get help doing a beam alignment and adjusting the focus from your dealer.

[Dan Hintz]

You need to ensure your focus point height is exactly where you think it is... if you're off by even 50mils, that could account for another several mils of kerf each side. Once you have the focal point height correct, focus into the material halfway... that will make your kerf as thin as possible.

[Mike Lassiter]

I tried to email Edward Williams who was at ULS in the past, but it seems is no longer there now as I got the email returned unable to be delivered. I had ask him if the focus length is from the top of the material being cut to the bottom edge of the lens holder that the focus tool is used to measure at, or the true focus length is different.

My point I guess is the tool used to focus doesn't measure 2" from the bottom to the top of the sloped area that the lens holder contacts to be considered in focus. My tool measures 1.963" from the bottom to the top of the sloped part. I laid digital caliper flat on top part stopping at the slope or ramp the lens holder touches to the bottom of the tool that sits on the material. Should this measurement be 2"? The bottom is tight on the rod, and doesn't move any; so I am pretty sure it is still the way it was originally when we got it.

[Dan Hintz]

Don't rely on a guessed measurement based upon where you think the focal point should be... measure it and be sure. By "measure it" I don't mean pull out the ruler, I mean run multiple test lines on a slanted piece of anodized aluminum (or other suitably high-res substrate).

When you start trying to measure things in mils, you need to start paying close attention to every measurement at the mil level.

[Mike Lassiter]

Dan,
please explain.
I always use the focus tool that came with the lens. Never use the auto Z or auto focus to focus. I don't consider that to be guessing. If you are suggesting to run multiple passes and raise/lower the table each time to see what gets the narrowest kerf, I guess I understand what you are saying. I have been trying to ask is the tool suppose to have the lens in focus when the bottom of the lens is 2" above the material I think?

[Scott Shepherd]

Mike, what he's saying is who said the tool is calibrated correctly? After all, it's a set screw on a piece of delrin holding it all together. Draw yourself a hairline about .5" long. Vector mark it on something like plastic or anodized aluminum. Now move it 1/16" in one direction (the line), and adjust the focus by .005". Run it again. Do that a number of times, .005" at a time, going up from your "zero" and then going down. You'll find a spot in there where the line is the thinnest. Move it to that same Z, then loosen the setscrew on your focus tool and set the tool to your machine. Then your focus tool is calibrated to it's thinnest line.

[Mike Ireland]

Ed Williams is still with Universal - just out of the office.
You might want to look at the Vector Scaling feature in the driver. You can find it in the Manual Control Driver under the Vector Tab. That will let you make adjustments so that you get an accurate sized piece when cutting.

[Dan Hintz]

Vector scaling is not the appropriate way to deal with kerf...

[Mike Lassiter]

investigating now. Had to go buy a set of feeler gauges to measure with tonight. So far I can say using 100% power and varying speed only 2% speed changed the kerf from 0.007" to 0.012" with cut lines 0.1" apart. I have ran lines side by side increasing focus depth 0.005" at a time and saw a 0.001" increase (from 0.007" to 0.008") on the 6th line. The last line cut was with speed decreased 2%. The reason being I noticed the material database has 2 cast acrylic materials listed. One is under Acrylic called Continuous Cast Acrylic (which seems to have faster speed setting than the other) and the other is under Acrylic > Cast Acrylic > Cast Acrylic. Using the same material thickness for both the Continuous Cast Acrylic has a 2% faster speed. Not sure what the difference is between the two, but I used the setting the material database put there, and the material cut good. The extra 2% speed reduction is needed if the paper is left on otherwise the cut is not completely thru everywhere.

[Mike Null]

Unless we are talking about two different things I disagree with Dan about vector scaling. I believe it is the proper way to adjust the inlay drawing for the kerf.

[Dan Hintz]

Unless we are talking about two different things I disagree with Dan about vector scaling. I believe it is the proper way to adjust the inlay drawing for the kerf.

Draw a 1" square in the middle of your table, then a 12" square surrounding it. Adjust the drawing scale by 1.003 to adjust for kerf on the inner square. The outer square just shifted by the same percentage, but it's 12 times the inner shift in total change. Suddenly tabs don't line up and things go to pot.

But it's quite possible we're discussing different settings...

[Mike Null]

I'm talking about increasing the size of the inner square to 1.003. In my scenario there is no reason the large square would change in size.

If you are talking about scaling the drawing as in 1"=100' for example, then the whole thing would change but that isn't relevant to this point.

[Scott Shepherd]

investigating now. Had to go buy a set of feeler gauges to measure with tonight. So far I can say using 100% power and varying speed only 2% speed changed the kerf from 0.007" to 0.012" with cut lines 0.1" apart. I have ran lines side by side increasing focus depth 0.005" at a time and saw a 0.001" increase (from 0.007" to 0.008") on the 6th line. The last line cut was with speed decreased 2%. The reason being I noticed the material database has 2 cast acrylic materials listed. One is under Acrylic called Continuous Cast Acrylic (which seems to have faster speed setting than the other) and the other is under Acrylic > Cast Acrylic > Cast Acrylic. Using the same material thickness for both the Continuous Cast Acrylic has a 2% faster speed. Not sure what the difference is between the two, but I used the setting the material database put there, and the material cut good. The extra 2% speed reduction is needed if the paper is left on otherwise the cut is not completely thru everywhere.

Mike, don't do this checking cutting thick acrylic. Do it with thin material like engraving plastics or anodized aluminum, then vector mark it with just enough power to mark it. That's going to tell you the sweet spot. Trying to figure the focus while cutting thick acrylic is a wild guess. The thickness of the acrylic is going to cause you some issues with that, since the beam, by design, is shaped like an hourglass.

If you are cutting thicker colored acrylic, you're ALWAYS going to have a wider kerf. ALWAYS. There's no way around that.

[Richard Rumancik]

I think Mike L. originally posted because he was seeing an unexpectedly large kerf. Once he gets the kerf thing sorted out, then he can address the clearance issue so that he can get a good fit. I agree with his approach - get the smallest kerf and then address compensation for best fit. I don't think he was asking about how to do do the compensation part . . .

But Mike Ireland said:


. . . You might want to look at the Vector Scaling feature in the driver. You can find it in the Manual Control Driver under the Vector Tab. That will let you make adjustments so that you get an accurate sized piece when cutting.

I don't know how this info relates to Mike L's problems. What Dan was saying is that you can't use vector scaling to compensate for clearance when doing inlay. (I hope Mr. Ireland was not suggesting that.) To compensate for clearance on inlay, you need to use a contour offset on one or both parts, not a scaling function on the whole part.

Scaling works (sort of) on circles and squares but that's about the limit of where scaling could possibly work for inlay.

[Mike Null]

Richard

I have done more than a few and enlarging the drawing as I suggested earlier works every time.

[Richard Rumancik]

Mike, how do you do the "enlarging" - what Corel tool do you use, and how do you decide the magnitude of the "enlargement"?

[Dan Hintz]

For objects that are "well spaced" and relatively small, so to speak, enlarging in this manner will appear to resolve the problem... but do the experiment I outlined above (or just think about it by drawing on paper and ask yourself what happens) and you'll see it's not the correct method to solve the problem. Fixing a 1" item's kerf by enlarging 0.6% to remove a 6mil kerf does bad things to large items around it. That same enlargement will increase a 12" item outside by 72mils. That should be quite noticeable.

[Mike Null]

First, just enlarge the inlay vector. Lock the aspect ratio and increase one of the dimensions by a few thousandths. That will equally increase all of the drawing and offset the kerf. The amount of increase will vary by material and may be a trial and error process but it has always worked for me.

[Mike Lassiter]

Sitting at work waiting to start and checked thread. Mike, if I follow what you are saying right when you do this you cut your inlay pieces first then send the file again so the base piece is cut but adjust the vector scaling and this will give you a smaller opening for the inlay pieces to fit into.
My thinking is by doing this not only will you change the OD of the material but the ID of any openings cut out in it. Sort of like inflating a tire inner tube out of a tire. Everything swells up for lack of better description. Is that the general idea?

[Mike Null]

Mike

For almost all of my work I first raster the opening for the inlay (as was done in the maple box shown) then I vector the inlay using the enlarging method and cutting from the back. The inlay in this case is veneer about .020" thick.

In the case of the abalone in the guitar I followed the same technique but vectored the inlay from the back at least 10 times at low power so as not to fracture it. The abalone is about .050" thick.

[Dan Hintz]

My thinking is by doing this not only will you change the OD of the material but the ID of any openings cut out in it. Sort of like inflating a tire inner tube out of a tire. Everything swells up for lack of better description. Is that the general idea?

Correct, which is why it's not the proper way to do things... you can get away with doing it on relatively small items (as we've seen) because the percentage of change is within the limits of repeatability. But once objects get larger, that method will bite you, hence why I suggest not using it.