Ok guys my frustration level is at an all time high. Here’s the deal. I’m trying to print station platforms with an Ender 3 v2 and can’t seem to get a good print. So I guess I will start off with my printer. Ender 3 v2 with Teflon feed tube, upgraded hot end, .4mm nozzle, auto leveling (yes I did manually level first), yellow flat springs, PEI magnetic bed. My print started flat and was going perfectly until the infill webbing started getting larger then it started pulling up the print. Also a more recent thing I have noticed is the first layer seems to have more of a gap between the lines than it used to. Even when printing an older file made with an earlier version of the slicer. I have tried these platforms printed with the Curality slicer and Ultimaker Cura with similar results. I have tried changing bed temp, print temp, rafts, no rafts and even changed the shape of the first print to go around and around instead of diagonal. That actually made things even worse and the gap between the lines was much larger. Here are a couple pics of the print I aborted (3rd attempt since I installed the new bed). Other items have printed fine, such as the ballast spreader in the last pic.
I'd be interested in seeing the print stopped at the first layer of the skirt and seeing a picture of that. My guess is that it is printing the first layer too high. Might be level but print head is too far away from the bed. Sumner
Here is the first layer underway. I changed some speed settings and a couple other things as per a recommendation from a video on optimizing your Ender.
Thanks but still hard for me to see what I would like to see. I'd like to see what that very first layer of the skirt looks like. It should be very tight to the bed like a fishing line that's glued to the bed. That very first layer is what makes it stick. If the head is too close it can't come out and might even start clicking as the extruder can't push it. That of course is too close. If you use the YouTube I posted you can get to where you use a piece of paper and it should drag on that paper. Not rip it but drag on it just a bit. It is something on gets a feel for overtime. The problem is I think with the auto level you cannot run the same test prints to use the paper to check. He has gotten around that in the video but it is a lot slower than if you don't need to do that. Sumner
I think you need to lift your bed a hair (esp on the right side oriented to the picture.) The lines don't seem to have much squish and there really should be no gap at all between the lines as they cover the bed. Like Sumner said don't get too close, I'd say 2 tenths of a mm would be about right. Edit to add...not sure how your auto-level is set up, to accomplish what I mention above you may need to adjust the z-offset value. Mike
I agree, looks like all the prints lifted on that side indicating that the first layers are not really sticking to the bed and holding the print onto the bed. If you haven't done it I would for sure take the time to follow that video and make sure the bed is level and the first layers are going down properly. Then the auto level ought to be working fine after that. If down the road things regress then I would go back through the same procedure again. If not printing will be very frustrating. Sumner
I did follow the video, in fact that is the video I used to learn how to level the bed when I first got the auto leveler. I did change the Z offset during the last print and by changing it from -1.16 to -1.17 and the head hit the print and took it off the plate. The print below is from the same printer a couple prints before the original print head died, with the auto leveling. The toes of the owls did come up a little but the print is tight and the resolution is good.
There are a few rules to 3D printing, among them I'm sure is don't change the Z offset during a print (exception: it's the first layer) Among the mess of rules there is the "first rule" however in 3D printing, with both resin and FDM, and that is: your print's success starts at the first layer. Learn it. Love it. Live it. If your first layer isn't as close to perfect as you can get, just stop the print right there because there is a high likelihood of failure coming your way. So back to the z offset - that is something that you want to adjust before (or during if you're fiddling) the first layer has completed. Using your print as the example, this should be what you're shooting for on that first layer: It's a fine line getting there which is why I adjust .1 mm at a time. Other thing I wanted to ask - with the temperatures outside dropping, is your printing area considerably cooler than it was before? That could cause long flat objects to cool too quickly causing your object to curl. Bed temp of 65C should be a good starting point for PLA. Once you get it dialed in again you shouldn't have to fiddle...fingers crossed. Cheers -Mike
I was on the first layer when I made the adjustment to the Z-offset and it hit a part of the print that was already laid out, pulled it off the bed and dragged it around as it tried to continue to draw. The nozzle was so close to the bed the feeder gear wasn't able to push new filament onto the bed. That was with a .01 adjustment. So I downloaded the latest version of the Creality Slicer and dumped my old config files and started over. I added the G29 to the start code to get a new level reading for each print, and left all other settings alone in the default. I re-sliced the platforms and now I am having it print them again. So far the first layer of the first platform looks good and tight and much better than the last time. My border print looked more like the "perfect" above than the "Close" this time, but we will see if it comes off the bed in 3 small lines or one solid one.
I will be heading to bed here soon, but I just checked the printer and we are at about 20% of the print and this is where the lifting was starting to happen before. So far all is well in the world and I hope it stays that way. There are a couple of differences this time over last time. First is the software. I downloaded the new slicer version from Creality, Dumped my old config files and started from defaults, and I also set the G29 to cause a leveling event every print. Yea so it takes a minute or two longer but when the print is 7 or 8 hours... Who cares? Next is the Bed, I releveled it and reset the Z offset, Everything looked better on the initial layer and so far so good... Yes I am knocking on wood... or my thick skull... either way... Well off to bed and we will see what happens in the morning.
Update: I now have 4 perfect platforms. It appears the main issue was some setting in the slicer. This last print was done at the same time, in the same room with the printer in the same place as before. We keep the house at a constant temp even through the night so a night time temp change was not part of the issue. This time I used the Creality Slicer with all default settings except the addition of G29 in the start code to cause a leveling event prior to the start of the print. Right now I have another set of 4 on the printer let’s see if I can get lucky twice in a day.
Great to have results. I've always used Cura with no problems associated with it as far as I can tell. If the ends of those prints started on the build plate and warped up later as the print went on my guess is that you fixed it when you went back and re-leveled and set the z axis and got good adhesion with the first layer, as per Mike's comments. Only way at this point to check that out would be to go back now and reuse Cura if that is what you were using and see if the problem is fixed. For me I'd probably just go on but since I like Cura I'd go back to it on the next print, make it a smaller one, and see what happens. Sumner
Hi Massey and all My students found that the best adhesion promoter for the FDM machines (MakerBots for freshman, Stratsys single and twin head, 8 material Objet Multi-material systems, and our really high end machines, and the laser fusion powder-based systems. We had 25 plastic and 2 huge Hoffmann laser fusion systems for superalloys) was AquaNet (the pink can, cheap stuff). They experimented with all kinds of stuff to get consistent first layer adhesion, but this one worked every time. I also had two students that investigated the materials characteristics and part strength when employing different writing algorithms/patterns. Issues arose when sharp corners were formed- the adhesion of the writing layer to the base layer was compromised. Internal stresses created by the writing pattern can vary widely. You might try warming the build plate a bit after the 5th to 10th row so the material can relax. Also, I cannot tell- but are you running your printer in a closed box? We had issues with open frame FDMs when the AC kicked on/off during the summers. My son has a couple Prusa's and had to build a cover for the whole machine to prevent writing errors consistently.
Looking more closely at your base bottom image- are you leaving bits on the build plate or is the writing head skipping or lifting? When my students build a 2ft x 3ft x 4ft FDM machine (basically a really advanced toothpaste-tube squirter!) they had to work hard at getting first layer adhesion, because the ABS and similar materials do shrink a bit... and with their dimensions, even 5x10-5 shrinkage becomes a lot in 2 feet...
Right now the machine is in the open, my house it warm in the low 70’s and we do not lower the temp at night. Believe it or not, it’s cheaper to leave the temp constant than to lower it at night and raise it when you are home. I did this experiment several times with the same results… but back to trains… Right now I am working on set #3 and everything is going good. It’s about 60% done and everything is sticking perfectly. I did run into a slight snag earlier today when I ran out of filament at about the 20% mark on the second set of platforms. I had another spool ready to go but a different color. Same brand and type before anyone says anything. And I was able to feed the new in right behind the old and the printer didn’t even know. Here is the results of that print.
