What the kerf?!

Okay. So I’m new to this whole kerf-adjustment thing. Tried making a case on http://www.makercase.com/. It asked the kerf as (beam width / 2) inches. As I found Dan stated the beam width is 5 mm. So! That told me I should tell it my kerf is 0.098425 in. Okay… Apparently not. Using Medium Maple Ply, my slots didn’t fit together. Just couldn’t quite do the trick. So I said "Hey! I’ll just cut that number in half and see what happens. This time instead of running the entire box, I just did 2 slats. WAY off this time. Like 0 way these things would fit together.

So now I’m tempted to go the opposite and try 0.19685 and see what happens. But, frankly, I’m sick of wasting my :proofgrade:! How the heck do I determine what value I should use here?

6 Likes

5mm? I can’t imagine that’s correct. Maybe .005mm? It’s going to depend on the material and possbily feed/speed as well.

1 Like

Cut a 1" square out of whatever material you are using… Measure the cut out square with your calipers… Subtract said measured number from 1". Divide by two…thats your kerf…
It’s the amout of material removed by the laser…
Many of my woods are .005" and acrylic is .006"

27 Likes

Thats pre-focused beam width. regular beam width will be a lot smaller, and kerf will depend on the material you use.

1 Like

Look… What I don’t know about lasers can fill a warehouse. But I re-re-confirmed what I’d read…

Very logical. I’ll do that!

I get what you’re saying. I’m just trying to figure out how to apply it.

What I’m very confused about here is why did halving my kerf entry, seem to make my slots smaller? I would have thought it would have been the opposite?

2 Likes

5mm (half a cm) is way too big for kerf. Here’s a thread from 5 days ago stating 0.007-0.008in

IDNHMGF(Y) but I do have one of these lying around for kerf testing:
https://www.amazon.com/gp/product/B00T86ADZ4/ref=oh_aui_detailpage_o00_s00?ie=UTF8&psc=1

9 Likes

Maybe this will help…

8 Likes

I don’t know what I’m dong either, but I think I used .006 when making an inlay from solid maple PF to go into a piece of walnut PF and it worked.

6 Likes

@Tom_A, the kerf is quite subjective when making box joints. Do you want really tight joints that you almost have to force? Or do you want loose joints that almost have a gap and need support until the glue drys?

Here is what you do. Go into MakerCase and make a 1" box. input your measured thickness of your wood, for :proofgrade: this will be right around 0.120", enter your kerf as an educated guess, I find :proofgrade: kerfs out around 0.21mm or .0083" or so. Download your plan and discard everything but two sides. Cut, test, adjust as needed. You will use a whole 2 sq. in of wood with each test. Once you find your sweet spot, WRITE IT DOWN!

This method has worked GREAT for me and I’ve been doing a lot of box joints.

Now note, every material kerfs out differant, you will have to do this one time for every new material.

13 Likes

Actually, that was a great explanation. Which only re-affirmed what I’d thought. So I’m still baffled by how halving my kerf parameter made things so much worse? Those tabs would have had no chance in hell of ever fitting ever ever ever. So now I’m wondering if I didn’t flub the decimal when I’d halved it. It just doesn’t make sense to me otherwise…

The other thing that thread made me think is… “What ever happened to jdodds?” :slight_smile:

Yeah. I love this idea. And I have what’s left on this :proofgrade: sheet should be just enough.

5 Likes

The kerf for the 1/8 maple ply is running roughly .007 to .01 inches. (Or about 0.25 mm)

Depending on where the beam is focused, either the surface of the material or the middle, it’s going to give a different kerf if you measure at the top of the material or the bottom.

(We’ve got tutorials on kerf in the Matrix. Check out the 2D Vector Programs Matrix or the Laser Matrix.)

Or you can just run a couple of test cases (small ones). If I kerf adjust, I generally just adjust by 0.1 mm ( .0035 inch) using an offset. No idea how the auto-program does it, but if you start with those values, you should reduce the gap by half. If that’s not enough, do it again. Two iterations and you’re probably tight fit.

9 Likes

Thanks! I guess I should have done some testing first. For some reason I thought it would be more intuitive than I’m finding it to be.

1 Like

Oh no, the kerf is a pain to mess with. :face_with_raised_eyebrow:

I should probably throw a caveat out there too - I don’t generally use the default settings. It might be slightly larger than that with the defaults.

Try a kerf of .01 inches and see if that works.

2 Likes

In addition to the suggestions above, my cut calibration tool is handy for calculating kerf as well as power. The boxes are 1/2" x 1/2" so you can measure the cut out square, subtract from 1/2" and divide by 2. If you scale the calibration by a factor of 2, the squares should be 1".

The other tool I use is a box that is about 1"x1.5". I put another box that’s 1/4"x1/2" centered widthwise but aligned with the top inside the larger box. That cuts a U shaped box that I can use to slide onto a piece of stock. I make the internal box larger or smaller as needed to get the fit I want. I usually do a couple of different sized ones at once so I can zero in pretty quickly. This let’s me use up some of my extra bits & pieces that I wouldn’t otherwise use in scrap from other projects.

BTW, the reason you halve the kerf in the box programs is because you’re getting half the kerf on either side of the line (which is the beam center. The tab is going to be narrow by a full kerf vs the size in the drawing. Likewise the slots will be wider by a kerf than the drawing measurement.

2 Likes

Couple of things i’ve learned as i’ve jumped into this recently as well.

  1. Make sure you’re using the right mode in illustrator, some modes will favor pixels over exact measurments. I found that making a 10mm square often rounded in weird ways until I played around with the art board settings.

  2. Watch your conversions, everyone seems to use inches and the GF works in inches, but i’ve found millimeters to be simpler to deal with in designs. 0.01" is very different than 0.01mm. oopsy. hahah.

5 Likes

I did a small-scale test and, at least for this experiment, telling MakerCase my kerf .0025 gave it a very nice snug fit. So I’m gonna go ahead and give my full box a try again now.

10 Likes

He reevaluated what was important to him and went epilog:

3 Likes

Okay… So now I’m just annoyed… Somehow adjusting the kerf screwed my box dimensions?!

My first attempt, while the kerf didn’t work to allow the pieces to fit together, was the correct dimensions. This one, the kerf is pretty nice, but I lost like .2 in! So now my stuff doesn’t fit in it!

Anybody have any experience with that MakerCase site like this?

EDIT: It looks like the site needed to be refreshed. I’d been playing with it a while and I presumed that as I made changes, they were all applying. Turns out they weren’t. I just generated with the exact same settings and got a different result. So, in the words of President George W. Bush: Fool me once, shame on — shame on you. Fool me — you can’t get fooled again.

9 Likes

most if not all of this could be avoided if we had precision in a kerf and a spot in the UI to put the kerf and then designate waste side or center cut etc… instead of all our fudged designs to compensate :frowning:

7 Likes

I think we will see that “someday”.

1 Like