Kerf setting most fundamental approach?

LightBurn Software Questions
,
1

I’ve gone through quite a few write ups, conversations and videos. I hope to spend some serious time locking in my approach for fits across my machines as part of a comprehensive overhaul of how I approach projects for better efficiency and quality. If we’re talking about tabs and slots joinery, to avoid explicit work to manually adjust dimensions, I would like to strive to use exact material thickness and tab width (established through tests) that are then adjusted implicitly through the established kerf measurement.

I spent HOURS today using three different “systems” without satisfactory or comprehensible results.

For this thread, let’s assume circular dot sizes, as I’m dealing with an xTool P3 CO2, their F2 Ultra UV Galvo and their F2 Ultra CO2.

My understanding on the P3, which I have 3 lenses for, the dot sizes are ads follows":

  • Small = .18 x .18

  • Medium = .2 x .2

  • Large = .23 x .23

I’m hoping to walk through the procedure starting from the absolute beginning. Do you apply kerf settings on both cuts, or just the tab or the slot?

Thank you!

2

I think it would be best for you to share an example file, that way we can answer accurately. I’m confused by your question.

Have you reviewed our documentation for how to determine what your kerf should be?

1 Like
3

Jess was a quicker with the link to the very nice documentation/explanation that LB has on this topic. At the same time, I would also like to note that a general search query here on the site also gives tons of good posts on the topic.

That said, kerfsetting is very easy to do in LB. You need to find your spot size and the desired fit. Because materials can vary a lot, I do an extremely simple kerf test - every time. You should let LB distribute the kerf, it works best, I have good experience with it, and yes, LB distributes the kerf evenly to both parts.
Small but important detail, if the shape is not closed, kerfsetting does not work.

Here is the kerf test I use often.

Kerftest.lbrn2 (21.0 KB)

You can scale the test up and down, it has no effect on the chosen kerf setting. You just set the kerf setting that fits your needs.

1 Like
4

I did go through that page and I made several of those gages. Not one of them resulted in a # anywhere near the stated dot size of my machine.

I also used this method, which raised more questions than answers.

image
image1319×193 4.16 KB

So the first approach uses no kerf and is designed to find the kerf. The second approach uses kerf for mating pieces.

So my most fundamental question still stands. Unless I overlooked it, that article avoids any mention of which part(s) to apply the kerf to. The second approach above applies it to both.

Tabs and slots = tabs and square slots for them to fit into. Joinery 101. I purposefully avoided examples because I wanted this discussion to start with the fundamental concept of approach to using a kerf value. Which cuts do you use it on? Do you have a solid kerf system that your system is centered around for perfect fit each time?

5

I don’t understand why you have those questions/problems with kerf settings. LightBurn assigns kerf to both parts
It is possible that you can do it in other ways, but by distributing the kerf evenly to both parts, works perfectly for me, and I use LB’s “standard” procedure for assigning kerf daily.
Your test is very similar to my own kerf test and if it doesn’t work, there is something wrong with the way you use the test or kerf, or your shapes are not closed.

My laser is a CO2 type with approx. 0.1mm spot size, plus/minus. For most boxes I make I use 0.075 mm in kerf setting. This gives me a relatively tight fit, i.e. I have to use force to assemble and separate the parts again, that’s how I want it.

1 Like
6
  1. I can’t use LB directly with an xTool, unfortunately. I still use it for vector work because it’s supperior.
  2. I disagree that LB automatically applies a kerf setting across multiple layers. I always use separate layers for outer and inner cuts for tab purposes and order of operations to help prevent every piece being 95% complete before disaster vs 95% of the pieces being completed before. I love LB and wish xTool could catch up.

If I were to make both parts on the same layer with the same kerf setting, that suggests that tabs and slots are using the same (+/-) setting which exactly matches the image that I posted, right?

I find it quite interesting that you are using a different setting than your machine’s dot size. I 100% expected that my kerf setting should match the machine’s dot size. I was able to find settings that worked for me, but they were not aligned with my machine’s claimed dot size. That’s really the major sticking point for me. I expected this to be more closely aligned with the specs. Maybe I should just run with what my tests show and not be concerned with whether that matches the mfr specs.

7

What is a KERF?
Kerf refers to the width of material that is removed by a cutting tool – typically a saw blade.

You cannot not have a Kerf if you make a right angle turn. You either have Kerf (as in a Kerf setting), or you do not. @JessN and @bernd.dk provided info on how to apply the Kerf, which is normally 1/2 of the cutting path.

8

THAT is fascinating. I need to digest that. Do you have more insight into right angle turn kerfs? I could 100% be misunderstanding this topic!

9

Sure! The Kerf Offset stays the same, but the cut path widens or narrows depending if you are following an X or Y direction. Laser diode manufacturing physics.

10

I don’t know what you mean here.

I’ve used kerf in most of my box type projects. I never apply kerf to a design… A few times I change the material, so I need to change the kerf without going through all the editing to recreate the same object for different build materials.

:grinning_cat:

11

Depends on how you process the project. Boxes.PY has a setting for it. I just apply something there and not worry about doing it in lightburn. If I was a production house, I might do it differently. :grin:

12

It is of course a pity.

If I set the layers that I use for a specific project to my desired kerf, they are respected. That is, if the first part of my test (from the file) is on layer 00 and the second part on layer 01, then they match up in this way the shell - kerf is correctly assigned.

That’s how it works.

…not clear to me what you mean here.

I knew you would “notice” it. Here’s the thing to say, theory and practice are not always 100% the same :wink:
Unless you have the opportunity to approach the matter like Russ Sadler, that is, have the machinery, the time and the necessary knowledge, you (I) can only come relatively close to the 100% from the theory, as we mentioned before.
When laser cutting, several things play into how big your kerf will be. In addition to the environmental parameters, the power distribution through your optical system is one of the most important parameters, this together with your cutting speed/power and type of material ultimately determines how big the gap you leave with your laser in the respective material.

Simple example, set your power to a value you know would be enough to cut through 5mm plywood/acrylic…just something that can be easily seen. Set your machine to give a pulse of 50 milliseconds and shoot. Afterwards you make a cut in the full material and use a finder blade or calipers to determine the gap thickness or compare both results optically with a magnifying glass or if you have a microscope. I of course do not know your result but here, especially if I look at the back of the test, I can only see a small hole at the pulse shot which I cannot even precisely determine how big it is. At the cut I can fairly precisely determine how big the gap is and it certainly does not match my official spot size of the laser…

IMG_3678
IMG_3678624×339 22.3 KB

In this picture you can see an example of the time factor for kerf, note that it is the small inner circle in the pulse holes that is the actual penetration, the rest is “impact crater”

IMG_3680
IMG_36801920×2397 377 KB

This “should” be my kerf - theoretically, but of course it isn’t. It’s again a short pulse with the same power setting that I use to cut this material.
For a 2" lens the theoretical spot size is 0.1mm, in the picture I would judge the hole to be a quarter of that…
That’s what I mean by theory doesn’t always follow praxis and vice versa.

What I want to say is, that whether it says 0.075 or 0.2 mm kerf, is “irrelevant” - as long as the result is what you are after.
Of course you should “know” the kerf of your laser in the different materials, it makes the work process a bit easier, but it does not necessarily have to reflect the “official” spot size of your laser.

2 Likes
13

I never heard that term or considered the possibility. Super interesting!

I hadn’t been aware of Russ Sadler, either. Just subscribed and will take a look.

Lots of great info in this conversation and I really appreciate all the helpful insight.

After some more experimenting I believe that I finally got a result that comes close to what I was hoping for. In the fit vectors that I posted I got a good fit with the +.18 components, which doesn’t match the claimed .2 for this lens, but it’s close and, as mentioned, if I can get good results with this, that’s all that matters. Going to spend more time today experimenting. I really want to be able to take any design and apply the machine’s kerf, vs factoring it into the design.

PS: The kerf on my Ultra UV is insane. It can require effort to remove a cutout due to the tight fit. Crazy.

1 Like
14

I never do that, it only causes unnecessary confusion and complicates the process for me.

I still dream of a UV laser :wink: but if I find enough jobs for it to pay for itself then I’ll buy one too. It’s the technology that interests me most about it. A kerf from such a machine is a whole other world, hope you find a good way to deal with it.

Todd, are you from the “metal industry”?, it strikes me that different “laws” apply here. My K=0.07mm added 3mm MDF, corresponds almost to an H7 press fit if we are talking about metal parts + here it is usually only one part that gets the tolerance or the task of ensuring the fit.

I also just played a bit with different layers/materials and the same kerf, because it is just interesting :wink:

In the picture you can see 3mm MDF black and not colored HDF. To assemble the round circle I need to use a small hammer - almost an H7 fit :wink:

IMG_3683
IMG_36831320×2053 895 KB

15

I’m actually from the software field, which is binary. Get it?

I just made a series of fits that brought a tear to my eye. I generated them without any burn or play value and used kerf only. No glue needed.

The one joint I’m not convinced of is the snap style. Very difficult to insert. I experimented with a combination of tabs and snaps (one snap on each end with tabs between). Kind of cool, but labor intensive to edit the vectors. Maybe I don’t need snap joints in the first place.

1 Like
16

Ok, so forget it so it doesn’t get even more confusing. :wink:

I’m not quite there, can you show me one?

17

image
image389×438 3.9 KB

1 Like
18

Ok, now I understand what you mean, it’s probably trial and error until you find the right value for the specific material. I haven’t worked with that type of connection, I think it’s best suited for large boxes with material from 5-6mm and up. If I could cut plywood with exterior glue I would like to try it, but my machine can’t handle it and the plywood I use and my machine can handle is too expensive for large boxes or furniture.

1 Like
19

THis is a good video on Kerf. Google Search

2 Likes