Test Prints For 3d Printer | Top 10 Calibration Prints For Your 3d Printer

MatterHackers

Subscribe Here

Likes

1,484

Views

52,446

Top 10 Calibration Prints For Your 3d Printer

Transcript:

Hey, I’m Alec. And today We’re going to talk about the top. Ten calibration prints. You can run on your 3d printer, so let’s head on over to the lab where we test out all our machines and materials. [MUSIC] Here we are in the lab. We test out new materials and new printers against this set of calibration prints. A couple years ago. These were designed by make magazine to test out different. Fff and SLA printers against different geometry conditions of 3d printing. They’re scored on a one to five scale, with one being a failure and five being a perfect score and they have different examples like Z wobble or support removal or things like that, But these calibration prints are used to get your 3d printer to perform in its optimal ability and to be able to print perfect print quality. [MUSIC] This is the vertical surface finish test and it’s designed to look for ghosting ringing or echoing. It goes by a lot of different names. But they all mean the same thing. What you would see if you had ghosting this print doesn’t. I’d give this one a 5 is you’d? See, make would be further along. You’d see a faint shadow of it behind here and the worse it is then the bigger problem that you have, but it just happens when you have a bit of wiggling of the printhead along the corners and like. I said this one’s pretty perfect. As the printhead makes a quick movement, it can oscillate, which creates a ringing effect. The oscillations diminish on longer print moves and the vertical surfaces clear up until a sharp turn is made again ghosting can come from a lack of rigidity like a loosely mounted hot end or a wobbly frame, springy belts printing at high speeds with a heavy direct-drive printhead, a bed that isn’t rigidly mounted or firmware settings for acceleration or jerk that are too high for what the printer can achieve. Some produce can’t handle printing at such high acceleration or jerk. So you’re just gonna have to tune those settings down and see what works best with that printer. If you’re getting a really low score on this test, what you can do is you can see different ways. You can more rigidly mount your printer or the simplest is to just turn down acceleration, jerk and print at a slower speed. It’s not the ideal solution, but it is the simplest. This is the horizontal surface finish test and looking at the top surface of it. I’d give this one a five because all of this is really really clean with three different sections, flat slope and domed. You’re able to see any sort of artifact or ridges from where the perimeters start and end a more noticeable. These points the lower the score. If this print didn’t get such a high score, some of the settings. I’d be looking at are the infill overlap percentage, which would just be this top layer. How much are each pass overlapping itself or extruder steps? Because if it’s over extruding, then you’d notice that here too. I’ll also look at some of the perimeter and retraction based settings around the dome and the slope here because that’s where you’re really going to notice those so that you’d want to use start and end overlap or retraction distance or how much overlap there is for each perimeter different things like that will affect these different areas. This is the dimensional accuracy test and it’s really designed for functional printing. If you’re printing something ornamental like a coaster or a keychain, just some model you find online. It’s not really gonna matter unless you have some interfacing parts because this is looking to see that when you tell it to print something 20 millimeters wide, that’s the size it comes out with some room for error there. It can be just a little bit over or undersized and still be okay, but if you’re trying to design something, that’s going to fit nuts and bolts in it. You need to make sure that those holes are the right size, so they’re not going to just slip out or if you have some multi-part assembly that everything fits together the way it’s supposed to. The second level of this pyramid is supposed to be 20 millimeters wide and deep and loses points, based on how far the averages from 20 millimeters if the average is between 0 and 0.1 millimeters, a full five points is earned, but if the deviation is between 0.1 millimeters and 0.2 millimeters 4 points are earned. And so on. Now, if you get a low score on this one, what you’ll want to do is first start checking your extruder steps for millimeter. If you have that too low or too high, you could get these layers printed, larger or smaller than you intend. We have another video an article on how to calibrate your extruder, so be sure to check that out to get that right, but if you’re still finding that you’re not receiving a good score after making sure that’s properly tuned, you’ll want to take a look at your motion axes, so your X Y&Z. If those have some sort of backlash to them where you’re able to move the printhead without actually moving these Stepper Motors, you’re getting a little bit of over swings, so it may move a little bit further instead of being 20 millimeters, That’s moving 20.5 because it’s just got a little bit of slop. So you want to make sure that your belts or your motors? Your lead screws are all properly tensioned so that when you tell it to move a distance, that is the distance, it actually moves. [MUSIC] This is the overhang test and it’s designed to see how well can your printer cool down? The hot plastic as it’s extruded. So it’s really looking for. How does the bottom surface look? This specific test print gets a one barely the way we score is based on how well does the bottom surface look for each section? We have five here. We have one two, three, four, four, the numbered angles. And then this section down Here is almost a gimme. Although this one was close to losing points on this one, the corners are just barely okay now. When you’re printing this, your print speeds will affect it. If you’re printing very slow, like ten millimeters per second, you’ll probably get a better score than if you print this at eighty millimeters per second, just because there’s more time for the air to cool down the molten plastic. It’s also important to consider what type of fan you’re using for your part coin. The Sierra 20 right here uses a radial fan for part cooling and an axial fan for its heatsink. Some printers do also use axial fans for their part cooling, but things can get a little weird there because they’re not really designed to put out air pressure, so it’s just kind of blowing air in one direction, and it’s not really able to focus it. Where a radial fan can. Usually you’ll see. Somebody put a really large blower fan on some sort of ducts that will reach around the nozzle and that can provide a significant amount of cooling, especially compared to just using an axial fan. So if you’re getting poor tests with this, try out, printing it in different directions because whichever direction the fan is coming from will also affect it. So you may see that this way. It looks really bad, but if you print it this way. It looks great well. Your fans probably on the back, blowing air this way, able to cool these overhangs a lot better. If you’re still getting bad results, no matter which way you turn it. You may want to consider changing out your radial fan or your axial fan to just a higher power radial fan and printing some ducting so that it gets the air right where it needs to be. [MUSIC] This is the bridging test and its testing similarly to the overhang test, but what it’s specifically looking for is unsupported spans like this, where it has to connect large areas with nothing underneath to support it. This specific print will get a score of three because these first three sections don’t have any significant drooping. You can’t see any bits of filament that are hanging out. But levels four and five do have little strands that are hanging down below and so those wouldn’t get scored well. This test heavily relies on the algorithm within the slicer, which is not something you can change unless you completely change your slicer. In this specific case, the slicer used instead of just printing the perimeters and then doing the infill. It did long passes across the span, and then the following layer had passes like this that covered it. Some slices will make that first layer. Have things go like this, and that’s how you get bad bridging because it doesn’t have enough filament below it to help support those moves, so you can try changing a slicer to get better results or you can try adding more cooling printing at different speeds, there are different bridging settings that can change it. You’re not gonna see something nearly this extreme and most of your printing, usually what it’ll be. It’ll be holes on the sides of your part or grooves or really small details. Usually you won’t see something as big as this that you need to bridge over, but this is a really good test to see how well, your algorithm for the slicer you’ve had tailored to your printer. How well that works out? [MUSIC] This is a negative space test and much like the dimensional accuracy test. It’s really about functional prints. If you’re having any multi-part assembly, you need to make sure that your parts fit together the way you intend them to. So if you have a bolt hole on the side of a part in a space that you can’t drill out later, you need to make sure that the holes large enough to be able to easily slide in that bolt. If it’s too small and you can’t drill it. You’re kind of stuck in general calibrating this out. You’re going to want to check your extruder steps per millimeter. If you’re over extruding, it’s gonna get stuck If you have start an end overlap, that’s too high. You may end up getting little beads that stick to the outer edge of this and the outer edge of the pin that finds them, or if your seam alignment in a weird spot that can cause issues. Basically, you’re gonna know if you’re gonna have a good time or a bad time, pretty much right away, and it will probably take a bit of calibrating to get this to come out, just right. [MUSIC] This is the retraction performance test and for most people will be the one that they start testing first. Just because this tends to be the most obvious error. You’re gonna get this specific model. It’s kind of hard to judge whether I’d give it a three or a four. This is one of the more subjective ones because it’s just. What do you think that it should be scored? Not anything you can actually measure. Really a five is gonna have perfect. Spikes of four is gonna have a little bit of wisps. A three may have what looks like branches where it’s retracting. But just not enough, and you end up with weird bits growing out from the side. A two is gonna start having issues with the tops of the spires, where it’s just not able to retract as reliably and starts just leaving them out and a one is basically just gonna be solid all the way around this one’s actually fairly hard to calibrate because there are so many different settings that can affect this. Whether it’s your retraction distance the extra length on restart whether using a direct-drive. Bowden remote drive. They all have different feedback that will make it retract better or worse, just based on the same settings, even print cooling or print speeds can affect this. Your temperature can affect it. Maybe a printing a little too hot, and that’s what’s getting the wisps, and if you turn into the temperature, it print better or maybe the filament’s wet. If you’re printing something like nylon. Pla doesn’t tend to have that problem but nylon. If it’s wisping, it’s probably got some moisture in it. There are a lot of different things you can change to make it work and it can get frustrating, but there is an end in sight when you can start seeing these wisps and it’s usually just very fine, tweaking at that point to get it to go just perfect. [MUSIC] The support removal test is kind of an interesting one. Because you’re really gonna be able to see when it’s a five or one. But in between gets kind of subjective, a five is almost exclusively for supported materials like PVA or hips that are used to get the bottom surface quality just as well as the top. Where one is going to be either. The supports completely welded or it’s just a total failure for these supported sections. In between though you’re gonna be looking for the supports getting wellö to the tops of these middle sections or the bottom sections are a little droopy here, and you’re getting it really close, but maybe these little beading. These little circles aren’t coming out, Just quite right, but maybe the side over here is coming out, right, It’s it’s a little bit of back and forth, and you should be able to look at it and go. Alright, that’s a three. The weight supports are printed can differ depending on the slicer. Generally, there’s what’s called a Z offset distance or Z distance or air gap and what that’s doing is it stops generating support a specific distance away from the bottom of the model. Some will do it where it’s just support. Stop 0.4 millimeters below it, or it actually will lift the nozzle in sprinting the bottom of these sections here so that it has time to fall onto the scaffolding at the right area. And then you can just trim it off afterwards and it will look good. It really depends how far that air gap needs to be. Sometimes point. Four is a little too far and point two is too close. It is something that is determined by your layer height, because if you have a point, eight millimeter layer height because you have a really big nozzle, then you’re gonna need a much larger air gap than if you’re printing at eight point, one layer height, which could have a much smaller air gap because there’s just a lot less material and there’s a lot much smaller section that actually is going to be touching the supports. It’s just a really interesting one to fine-tune because you really just have air gap in temperature that can be impacting it, but once you get a dedicated support material. This is basically a non-issue test. [MUSIC] The full bed dimensional accuracy test is basically an expanded version of the dimensional accuracy test where this is a small one. This one tests the full bed size to see. If any discrepancy here it gets expanded. If you make the print bigger, so you’ve trying to print a really large print. That’s multi parts to make one big assembly. You need all the parts to fit together. No matter how big they are or where they are on the bed, The Z wobble test is an interesting one because it’s either pass or fail If these sidewalls of the print are perfect is a pass if they have some significant bumping, it’s a fail, there’s generally two ways. I can think of getting this failure and that’s related to either to the lead screw or the Z-axi’s motion, so the lead screw on the CR 20 right here has an eight millimeter pitch, which means for every rotation of this screw. The whole carriage will rise eight millimeters, so you should see any of this bumping in a cyclical fashion, basically, every eight millimeters You see the same error. If you’re able to get them to move smoothly, but you’re still getting issues. Then there is one big one that I see very often misdiagnosed. And that’s taking your lead screw and seeing this loose end up here and how it wiggles and putting a support up here by printing out a mount with a bearing in it to help constrain this. But what you’re actually doing is over constraining it with your lead screw. You should only ever have two things interfacing with it. The lead screw nut and a couple that attaches it to the motor or it goes directly into the motor and as part of it by adding that bearing at the top. What you’re doing is you’re exacerbating any sort of bend in the lead. Screw if it’s not perfectly straight, which it likely isn’t, it’s pretty hard to get these to be perfectly straight. It could have some bend in it. Then what’s gonna happen is it’s gonna bind and any sort of bend in. It is going to be really shown off here. Where in the middle it goes, okay, but as it gets out here, things just get bad. If you have a bearing at the top of your lead, screw at best, it does nothing and at worst, it’s severely impacting your prints, so I would advise removing the bearing and printing without it and letting this spin wildly because these are what constrain the motion, the wheels, the bearings, the rails, that’s. What keeps it riding smoothly up and all this should be doing is providing the lift to get it to move vertically or if it’s using belts. Then you won’t have the same problem, but if you have a screw, don’t put a bearing on top. [MUSIC] The squareness test has four blocks that are printed at each corner of the build plate and then one in the center. But that’s not used for measurements with these four. You’ll use an angle gauge to measure up against these outside edges and make sure that they measure 90 degrees. If they don’t, that means somewhere within the mounting system of your printer. There’s an error and you’re not having everything perfectly square. If you’re printing something, that’s not structural and doesn’t need that. It’s okay to let that go. But if you’re trying to print something that structural needs to fit together, it is something you’ll want to diagnose and figure out how you can fix that in your machine. So those are the top ten, really 11 calibration prints. If you even keep them track, there’s just no way to cut anyone out because they’re all so important in getting the best print quality out of your 3d printer and having it printing in its peak performance. This is what we do. When we bring on a new material or a new printer to make sure it can print at its absolute best and so. I hope this has encouraged you to try this out on your printer and see if you can get a perfect score on all eleven of these. I’m out for mater hackers. Thanks for watching. Hey, there! I hope you enjoyed that video on the top 10 calibration prints for your 3d printer. If you’re looking for something a little more in-depth, you can check out the troubleshooting guide. That’s also on the channel or some other articles on madder hackers comm. If you’re still looking for something to print, we do have fill a mint. Who is our mascot and the troubleshooting case on himself, which has a bunch of different parts in him, designed to help troubleshoot. Your 3d printer. See on the next one.