Does the laser beam pulse ?
I thought it was a continuous beam, but does the power supply/control have an input, especially on the presence of any rippling effect ?
That sub looks perfect.
How did you do the texture in the logo? Of course, as long as you donāt mind me asking.
Is it engraved noise?
Have you tried that techinque (whatever you did with the logo) with the Glowforge?
What were your power/speed/LPI/focus distance settings? I only have white PG so Iāll need to order some black.
But the lines show on the white one I did - itās just harder to photograph with my phoneās camera.
I scanned the logo (itās actually from a decal of the plaque that was on board). Then I turned the color jpg into a b&w image and then dithered it. For the large sizes I stopped there. For challenge coins I flipped the colors (black to white) as at the small size of the coins the dither resulted in too much fill.
I did the engraves - boat emblem, sub, lettering in 3 different operations because I wanted the emblem & sub to be layered and to constrain the length of the rastering passes - the sub was wider than either of the other two engrave objects. Doing 3 different operations allowed me to layer the sub & emblem and also kept the X-rays travel to just the width of the object being engraved vs the full width of the sub for the entire height.
Iāll dig out those files and give it a shot on the GF. Theyāre DXF ones so it wonāt be an exact comparable test since Iāll be translating them to SVGs for the GF. Iāll give it a shot this weekend when I get back home (moving the daughter back home now from school in NC to home in CT for the summer - 11 hours of driving tomorrow).
Trying to remember from this morning, (which feels like a week agoā¦)
I probably used my standard acrylic settings for the engraveā¦ 20/200/0.116. LPI was 340 for the black pieces. I ran another one at hi-lpi and darn near burned through it.
Will be interesting to see how it comes out.
It might be related to the accuracy of microsteps on the stepper motors. If so, the pattern should repeat with spacing equal to the circumference of the toothed drive wheel.
Huh. Way different than mine. Iāll have to try these when I get back home.
Definitely another possibility. But thatās rapidly getting out of my comfort zone.
If itās stepper related, (ie: mechanical), there probably isnāt anything they can do to fix it except upgrade to better steppers, (I believe they use 3 in this machineā¦they need two just to move the gantry along the Y axis), and the odds of that happening now are exactly zip. Theyāre just too close to launch at this point, suppliers have already been selected, and with any luck, the parts are on their way, or here.
Certainly something they could look into for the next model though.
(Canāt wait to see what that one will do. This one is already unbelievable for a starter laser.)
Most of whatās in the thread covers it, and there are a few other things we know about that arenāt mentioned. Ripples always present in laser processing, as each pass has some overlap with the last, and energy distribution isnāt uniform across the beam. Further, thereās a āsplashā effect where you can get small ridges to either side of your beam if you deposit enough energy. Plus all the other reasons mentioned here.
Most back-of-the-envelope settings create ripples. Careful tuning of settings minimizes them. Itās extremely sensitive and nonobvious (higher resolution and power can make them worse as they worsen the splash effect). In the very best case, you should still not expect the results to be smooth under magnification.
Should that not be what youāre looking for, Iād recommend taking a refund and looking for another tool - likely not a laser, although other laser technologies (of which Iām not aware) may be capable of engraving thatās even flat under magnification.
i think the better question isnāt whether the glowforge is capable of eliminating them but where on the spectrum you can expect it to deliver once the settings are nailed down (using the above examples as a rough guide, say).
Those are two things I hadnāt considered. The first is one of the āduhā moment things. Obviously the center of the beam is the highest focus energy and the edges less so - not enormous differences perhaps, but significant. Like the way a flashlight beam fades to the edges.
The second one is what appears to be biting me - I used the standard engrave settings. @Jules is seeing better results but using radically different ones. Iām going to do some settings variation testing to see if I can tweak it down some more. I know they wonāt necessarily be useful once we hit production but the process to get them will be.
Thanks for the insight.
To add to jrnelsonās question: Are you expecting the unreleased low power settings to be key to ameliorating the ripples?
I expect this will get dramatically better, but in terms of setting expectations, you should assume only whatever weāve put in our product specs. So if the current engraving performance is a showstopper for you, Iād recommend looking for a different tool. If not, hopefully youāll be pleasantly surprised with what you actually do get.
Yes a single score will cut a U shaped channel and may have some splash each side but when you repeat that side by side with a large overlap I donāt see why it results in much courser ripples much further apart. Something must vary with the Y position.
For example if I milled it with a ball end mill, that also makes a U shaped channel and could leave a burr each side. However if I cut a pocket with a step size one tenth of the bit diameter it would make a pretty flat bottom with ridges only at LPI frequency and burrs only at the pocket edges. If there were wider spaced features it would be due to uneven motion.
Plus @jamesdhatch has a laser cutter that seemingly doesnāt create ridges so it canāt be something fundamental to the laser cutting process.
Your spec states āPositioning precision to 0.001ā (0.025mm)ā. That is about 10th of the kerf, so why does it make ridges that are something like two orders of magnitude bigger and an order of magnitude bigger than the beam width?
I would guess that under just the right circumstances āsplashā is cumulative in some way. That is, instead of the next pass vaporizing the splash from the previous pass, it essentially just shoves it over a bit and adds additional splash (remember also that the splash means your beam is cutting through a thicker piece of material where the splash is, hence wonāt cut as deep). Answer is almost certainly in speed/power adjustment.
An analogy from the milling/drilling/cutting world might be those speeds and feeds where the swarf doesnāt depart neatly but instead welds itself to the edges/bottom of the current cut or to the cutter and does bad things.
I just donāt buy that. Yes splash could be cumulative but why does that give ripples that are fixed relative to the Y axis?
@cynd11ās test showed that if a circle is engraved and a second circle engraved next to it with different focus and power the ripples are basically in the same places.
@jamesdhatch showed that in a large engrave the horizontal ripples extend for several inches at least. Something different must happen at different scan lines to get ripples. It canāt be a chaotic physical process as that would be random, not tied to Y ordinate.
The product specs state it can engrave at 1000 LPI and with 25um precision. If the scan lines have that precision why do ripples occur a few mm apart? I can accept there might be splash that builds up line after line but after 100 lines it should have stabilised, started oscillating, or behave chaotically. None of these would give position dependent ripples largely unaffected by power and focus settings.
It seems that the maximum engraving LPI is about 195 and then the results get worse.
Give me access to a PRU and I am sure I would be able to get to the bottom of it and solve it for you.
Next time you are in my area, swing by and Iāll be happy to show it to you.
You seem like a scientifically trained person so I probably donāt have to advise you not to draw too many conclusions from not nearly enough data, but Iāll say it anyway. (One of my favorite Sherlock Holmes quotes). As I mentioned when I first did the test, I was not using the optimal recommendations from GF for those engraves, since I was trying to test defocussing. Iām still interested in defocussing; ripples, not so much. And I have a lot of projects Iām waiting to get to and not enough time to do them, so I probably wonāt be looking into this in detail any time soon.
Youāll have to pay his way ā¦ He is in Europe (Great Britain I think)
I know. Hence the wink. I did say if he was in my area. Who knows?
Reminds me of Mark Twainās quote from Life on the Mississippi:
āThere is something fascinating about science. One gets such wholesale returns of conjecture out of such a trifling investment of fact.ā