LPI and Engraving

I was trying to see what was really happening and started thinking in terms of the kerf (0.003 for figuring). The laser is hottest in the center of focus, and less so outside particularly where it is not in perfect focus so it is hottest at the point of focus but still affecting the piece to the width of the beam. That is a reason that cutting half-inch stock even with multiple cuts is so hard as the beam is still cutting the material wider with the lower power because it is wider and that shades the bottom where the cut is needed.

With engraving, it is the opposite problem. the cut is not deep but you want to remove the whole layer of material so near the center it is cut deeper than the edges. by inverting the 240LPI for example to 1/240 you get ~0.004" which is a hair less than one per kerf any bit left between kerfs will also burn but there will be obvious lines as the path of the center will be deeper than the edges. 340LPI is 0.0029" or just under a kerf so the kerfs will overlap some but not much so the ridge between cuts will be less stark.

450LPI is 0.0022 or almost two passes per kerf. This puts the center of the next cut well into the ridge of the next making the ridge between them about half as high as with 340LPI. 690LPI is just a bit more than two passes per kerf lowering the between kerf ridge to just over a third the height of the 340LPI ridge.

1355LPI is 0.0007" per pass or about four passes per kerf making the ridge one-quarter the height, (or less as the ridge is slightly concave) and almost indistinguishable even in acrylics.

Wood will have grain that will make their own ridges from huge ones in oak to much slighter ones in Walnut and still less in some more exotic woods. I have found ways of reducing this effect, or rather not making it worse, first by making the fewest (1), hardest (full power) passes possible with the highest LPI. This leaves only speed as a variable and for some things, the highest speed is not enough. For this reason, stepping down the LPI vs lower power, having lower power means there is less range in a variable engrave, And even in a solid engrave less power increases the percentage difference between the hardest and softest grain increasing the grain effect. Taking multiple passes will also multiply the effect with each pass.

There are many opinions about LPI but I have tried to lay it out by math and logic to help those trying to figure it out. Doing one pass per kerf is around four times faster than four times per kerf although higher speeds help lower this, and bigger engraves have less waste time speeding up and slowing down per pass than smaller things so the relationship is not linear.

Still, when cutting time is in minutes (or seconds) and engraving times are often in hours, one is tempted to save time and go for the faster engraving. The Math will always be there however and your choices will be reflected there. :slightly_smiling_face:

Edited for clarity :slightly_smiling_face:


Obviously you have thought a lot more than many of us about how this all works. Thanks for this explanation. I will have to read it many times and see if I can get it to sink in.


So, one armchair laser physicist to another, here’s my theory on why I think you’re right, but your reasons why might be off:

While this is the basic gist of what is happening, I don’t think the logic makes sense – I think you reached the correct end result without the correct underlying causes. Diffuse light sources and laser sources are fundamentally different. With collimated laser light, the light density is much more uniform than with a point source like a typical led or filament, and using a lens to shrink it down to a pinpoint makes for a very uniform end image – there is no “out of focus” or “brighter” areas, it’s all one very small, very intense circle of infrared light.

Where I think the logic gets back on track is that you do see more power in the center of your laser pass versus the edges. I think that’s more down to the fact that it is a circle. I did an illustration of my take on the physics of that situation, let me see if I can find it. Here we go:

So yes I think the individual passes show power fall-off on the edges, but I don’t think it has much to do with the focus. it’s more about laser residency time on any given point as the laser passes by. In the end, I’m not sure that this is anything other than academic, we know this is how it behaves so the reasons why are less important.


Here is an example that I was working on…

in the first case the variation per cut is very obvious and I think 340LPI the variation by time is an interesting alternative consideration. going very slow and low power should by that be more flat as the time on point would be more even. that would make an interesting contrast to very fast and high power, perhaps with multiple passes?

1 Like

I’m not sure… because fundamentally the same problem is in play, you get more laser residence time no matter how quickly you move. There’s probably a formula for cross section of a chord of a circle, let’s see:

Chord Length = 2 × √(r2 − d2) where r is radius of the circle and d is the distance from the center of the circle. We’re talking about such small numbers here, it’s interesting… but lets assume the beam width is about 0.008", at midpoint we see chord length of 0.008" by definition, but if you go out from the center by 0.003, you get: 2 × √(0.0042 − 0.0032) and the chord length is 0.00529, which is about 66% of the residence time. go out another 0.0005 and it gets much smaller.

At this point I feel like thermodynamics and material thermal mass/conductivity starts to make a significant impact on the end result, so chasing these tiny numbers around starts to get a little pointless in terms of any measure of accuracy… but like we were saying earlier, we can clearly see the end result via direct experimentation, so the details are moderately trivial.

One thing that might not be trivial is that the conductivity and thermal mass factors are actually probably a bigger concern at slower speeds. Just like we see wider kerfs at slower speeds, I would bet we’d see more thermal effects around the edge of your engrave paths at slower speeds too.

I dunno, lasers are weird.


My thinking was along those lines that the outer areas would get the larger effect than the center. But whatever the cause the ridge/valley effect is easily demonstrated. Unlike the power and speed setting that have no known (to us)real world values, LPI means something in real life and the math works as I tried to reason it, and easily demonstrated empirically.

1 Like

Very interesting ready, however, i would have to read and reread to even understand what you’re saying :crazy_face: But very impressed about the thought that goes into this stuff!


Can someone say this in video… maybe then it will sink in! :woman_shrugging:t3::flushed:

Welcome to the forum kvhcjay! There is a lot of information here, there is a lot to learn that makes more sense the more you do with your Glowforge.

The photo in post 4 tells the tale. LPI are lines per inch when engraving. The automatic level is very few lines per inch and leaves a distinct “grain”. The very high LPI is very much smoother and clearer, but takes a lot longer, and must have the speed adjusted experimentally, so for now does not have an automatic setting.

1 Like

This topic was automatically closed 32 days after the last reply. New replies are no longer allowed.