Just so you know to measure the e steps; measure and mark 120mm from where the filament enters the extruder. Extrude 100 then measure from the extruder to the mark again and see if 20mm is left. You should have the filament loaded and primed in the machine during the entire test.
Yep! I did that wrong. I’ll redo that test here.
Following your advice, my settings seemed way off. I had to do it a few times to make sure it was alright. It’s hard to measure the filament inside the enclosure. But after I got comfortable it looks like my extruder was nearly 10% off. My new e-settings is 235.17 after all is said and done.
Wow. I’m trying not to assume it’s the settings. But I’m printing the wall thickness cube now. It already looks phenomenally different and far more even with the initial layer. We’ll see how the cube looks once it’s done printing!
Thank you Atom. This was already amazing advice. I’ll give the tie fighter another go after I’m done calibrating it!
So I did the extruder calibration. It seems to have helped tremendously with a lot of the issues. I wasn’t able to do the wall test though, since I couldn’t really change the filament size, and honestly I was getting an average of 1.6 mm for the filament instead of the 1.75 mm. I ended up ignoring that. I also went back to the default settings for the most part. These are the new prints:
There’s far more detail in them! Still having a bit of stringing issues, but it’s no where near as bad as before. The elephant’s feet is damn near gone after I used your suggestion, so thank you again for that. There’s still some striations where I guess the filament got caught on something, but again, it’s not bad. The biggest issues I have is it seems like the print came off of the heating bed on the lower panels.
Besides getting the filament to roll smoother, is this pretty close to ideal from what I can expect from my prints? I’m fairly happy with it, but if I can push the prints further to perfection I’ll keep tweaking.
You can get better, you should be able to eliminate the banding and stringing. But your getting there. Retraction settings should help with both of those issues along with calibrating k value but that is a bit tricky. What retraction settings are you currently using (retraction distance and speed)?
So the warping was a bit worse than I thought. I couldn’t glue the wings together and they would have to be filled if I wanted to salvage this. But I’m interested in getting to know this better.
I turned off the heated bed after the initial layer thinking it wouldn’t do much, but apparently that might help with the panels curling up. This is my own guess, so let me know if I’m right or not, but would slowing down the print on the initial layers help? When it does the panels, it has a sharp 180 degree turn from going on the inside of the panel to the out. I’m assuming that turn is causing the plastic to pull from the print bed once it cools. I also have my filament set to 195 degree temperature since a lower temp seemed to help with details, but I may raise that to 200. Either of my ideas sound right on the warping?
My retraction and speed are the default values, so 5 mm retraction distance and 60 mm/s retraction speed. I’ve read somewhere that retraction speed shouldn’t be too fast with PLA and it’s more sensitive to distance than ABS, but when I started messing with it I didn’t see much change in upping and lowering these values. What would you recommend?
As far as the k-value goes…I’m not sure what that is. When I looked it up, this site said it was the parameters for how the machines deals with the pressure build up when extruding at high speeds. Why is it tricky to calibrate?
You should leave the heat bed heated throughout the print, by default luban cools it down after the first layer but i find that odd, I like to set mine at 55 and leave it there for the whole print (for pla).
Printing at 195 should be perfect for pla
I recommend a retraction setting of 2mm at 45mm/s
Here is information on k value and how to calibrate it.
Let me know if this helps and if you have any more questions
Above and beyond, thank you. I tried doing the Marlin test on my printer. Filled it out as best I could. Yet despite selecting the origin at center tab it still doesn’t. The Gcode looks like it’s going into the negative. I could probably manually change it, but I think I’ll take the recommended setting for now until I get comfortable with it.
I also updated my retraction settings to what you have and I’m giving it another go whirl. Probably be a bit. I’m not comfortable leaving it running while I’m at work. But I’m interested to see how it turns out!
Out of curiosity, what wall settings and initial layer settings do you have? I’m guessing the smaller the wall settings the finer the detail but thinner the wall? I’m ultimately trying to see if I can print some miniatures with this, so I want to see how fine of detail it can go.
there are a number of different “wall” settings in cura, there is one for wall line width, that should generally be left at the same size as the nozzle size (0.4mm standard) and that is how thick a single line is. then there are a bunch of settings to determine how many individual wall lines you want on your print. other than that wall line width, none have a major effect on detail, but do have a major impact on strength and print time.if you want the best detail and only plan on printing small prints, i would recommend buying a smaller nozzle so that you can turn that down. know when you do that you may have to re-tune the printer to get good quality prints. if you want to learn more about how the setting in cura affect your print i strongly reccomend getting the “settings guid” plugin. by going to the market place and downloading it, it should be at the top of the list but you can search for it in case its not. here is a snip:
this plugin will help explain all the settings in cura and can do so far better then i can.
you can get some really good detail out of most 3d printers, the SM 2.0 included. but it will take quite a bit of effort, and tinkering. not trying to scare you but its just the way it is. in the future if you have the opportunity to get a SLA printer, that would be best for miniatures as they can produce much more detailed prints then any FDM/FFF printer. there has been some discussion on making a SLA add-on for the SM, so you can hope for that too (but i’m not sure i would hold my breath).
last of all, my biggest recommendation is just keep printing, the more you make and play the more you’ll get how the whole system works and the better prints you will get. i look forward to hearing how the next tie fighter comes out, and as always if you run into any more problems we are here to help.
I remember trying to mess with Cura. I couldn’t get it to connect to my printer though. I could make the settings that my printer had, though I’m not sure how to send prints to it, unless you just had to save the gcode file and move it to your printer via USB, though I’m not sure if that’s possible.
I’m absolutely fine with tinkering and putting effort into the 3d printer. I want to get good at it. I want to understand it. When I decided to get a 3d printer, I had no idea that there were different types of printers that would be good for different types of things. Honestly, if I could get something like the MakersMuse’s tolerance and clearance file to print and actually work I’d be happy with that. But the stringing may still be an issue. Speaking of…
The TIE Fighter finished.
The details is really quite nice. With the exception of a few gaps on the wing panel, and the holes on the top of the cockpit and engine it’s really damn smooth. I think making my walls a bit thicker may help with the engine and cockpit, so I’ll try that next. Part of the wing came off the bed too, but I think I need to do a raft as opposed to a skirt and maybe that’ll help. I’m still encountering stringing issues though and the ‘zits’, though I think they are one and the same since they don’t look like blobs really.
There’s a temperature test tower I read about: https://blog.ktz.me/printing-temp-tower-to-get-better-3d-prints/ May end up running that if I can figure out the gcode.
So, a few questions for clarity here. Am I right about Cura in how to set it up with the SnapMaker and transfer the files to print? Would it be worthwhile to do the temp tower? Lastly, I’m wondering if some of the defects like the holes in the top of the cockpit the engine or the small gaps in the wing panels are a result of the PLA being out in the open air an absorbing moisture. I read that you can pop it in the over to dry it, but I’m not too keen on cooking plastic, is that an issue some people have encountered?
As always, thank you Atom. Look forward to hearing from you.
Glad to hear your prints are improving. You are correct, to use cura you need to save the gcode to a flash drive, then plug the flash drive into the printer and print from it. In my opinion it is actually easier then printing via wifi with Luban.
Temp towers can be useful, I don’t use them personally, but that’s mostly out of laziness. Generally you only need one per filament (all SM black filament should need the same temp, but other black pla or other color SM filament might work better at a different temp.)
The holes and gaps are not likely the result of the filament (though the stringing and zits may be) its more likely a result of k value, or retraction (my earlier settings where just recommendations that work for me others have set retraction as low as 1mm and as high as 5, speeds range from 20 to 60). They could also be a cooling issue. Sometimes when layers get small then they print too quickly and the previously printed layer does not have time to cool before the next layer is printed. There is a setting in cura to fix this problem.
I forget have you already put thermal paste in the thermistor hole? It will help keep the hot end at the set temp and could help with the zits, holes, and stringing.
Temp. Tower, section 4. FINETUNE YOUR SOFTWARE SETTINGS TEST-SECTION
I would suggest to calibrate your k-factor (linear advance), the most people in the forum use a value between 0,08 and 0,1 (if you do the test from the following link, you have to edit the gcode, the "K"´s have to be small “k”, because of firmware!!)
Atom, I’m working on the retraction right now. I set the distance to 0 and speed to 10, then I plan to up the distance until I hit a sweet spot, then work on the speed at that sweet spot in roughly the same manner. It’s funny though…my second print (distance 1, speed 10) actually seemed to work really really well. There’s one wisp, no zits or strings whatsoever. I’m continuing to print higher until I see it degrade though.
As for the k value. I set mine to 0.08. I couldn’t get the gcode on the marlin website to print properly. It seems like the x coordinates go into the negative, and apart from manually changing that (which I’m still hesitant to do) I can’t seem to get it to work. It prints a single line on the corner edge of the machine and hardly moves. So I just took the recommended value for now.
I have not put any paste in any sort of hole. I didn’t realize that was a thing! Would the thermal paste you put on CPUs work? I have that laying around somewhere. I’m going to have to take a look at my machines once it’s done printing to figure out where that hole is, but I’ll definitely give it a go whirl.
Thank you for the guide! I did not know about the k value on the website issue. If I can get it working, I’ll keep in mind to manually change the casing on that.
Do you know why whenever I generate the code it goes into the negative positioning value in gcode? From what I’ve seen, Snapmaker doesn’t go into negative coordinates on the machine, yet this gcode is telling it to go into negative x values. The result is the machine prints on the corner edge and doesn’t advance from the edge. Besides the origin checkbox, I’m not seeing any options to mess with the coordinates. Is this something I’ll have to mess with manually too? Or am I not seeing an option on this form?
You have to enter the size of your build plate, after generating you could preview your code in Luban or any other slicer to double check.
Yes thats exactly the kind of paste, here is a thread pointing out what your looking for:
Whats wrong with this picture - printre hot ends
Here is a mini I did in about 2 hours, poorly painted. Thought you might like to see what you can expect, pluss SW for the win
That is pretty damn good detail! Looks really small too, do you have the measurements for it? Did you need supports too or was it fine without it? Lastly was that in Luban or Cura? I highly doubt the slicer software would change in regards to all these measurements right? (1 mm retraction and 10 mm/s would be the same in Luban and Cura since their just values given to the machine)
I’m excited to start printing another TIE Fighter after all this poking around. I’ve redone the extruder calibration. I took off the top part of my enclosure to get a more accurate measurement of the filament. I’m still doing the retraction tests…but I’m almost done. After this print I’m going to change the layer height, then the speed and see if that somehow messes with my retraction settings. After that I’ll poke around with the k-value gcode and see if I can get it to measure my bed correctly and see how that looks.
But thank you! It’s good to see a benchmark I can hold my machine too.
The little Droid is 29mm tall by 35mm long. It did require supports, I sued tree supports. I used cura to generate the gcode and printed it via USB thumb drive.
You can find the profile I used and more pictures (prepainted) here:
Suggestions for removing supports without destroying the model
The basic settings tend to be the same, but cura has many more settings to allow for fine tuning (such as minimum layer time that was needed for this print. Luban does not offer that setting). Its also important to remember that different slicers will use settings in different ways. Even the same slicer can have inconsistencies, with this guy it kept alternating between two different support layouts, even though I didn’t change any settings.
And I belive it can do better with more tweeking and some minor modification to the printer.
So I went and printed close to…a dozen? Maybe a dozen and a half of those print towers. I found that for a low resolution (0.24 layer height) print a retraction of 1 and a speed of 10 gives me no strings, no wisps, and the final print points are nice and sharp. I used that for a medium resolution (0.16 layer height) and a high resolution (0.08 layer height) thinking that the retraction settings would be the same.
They definitely aren’t.
In fact, I couldn’t get any retraction distance or speed setting to work. I even tried messing with the flow to try and get a better resolution, but I still had strings.
The thing is, the prints at low resolution looked really sharp. So on a whim I decided to print another TIE Fighter, but using that low resolution. This is the result:
The print lines are as sharp as I have ever seen them. No stringing or wisping at all. Even the laser cannons on the TIE Fighter are crisp and well defined. There are some issues though: There’s a very tiny bit of an elephant’s foot going on, but not too noticeable. The sides of the wings have layers that seemed to have separated a bit and there’s definitely steps on the top of the print you can make out, which I assume is because of the lower resolution.
So the questions I have are thus:
Is there another setting besides retraction that would help for medium and high resolution prints to help with strings or would I need a different slicer for more retraction options?
Let me know if I’m spot on with this as I want to see if my assumptions are right, but the layers separating can be alleviated by increasing the temperature of the plastic, right? Or even making sure the enclosure is actually closed (I kept it open for all the testing I’ve done) so the plastic temperature doesn’t cool too quickly?
I may do this one more time, but with a higher k-value.
I guess next step is playing with Cura!
I strongly recommend playing with cura, its just better than luban. As far as layer separation, raising the print temp is the usual fix, you could also reduce the cooling fan speed, but the SM does not have great part cooling to begin with so I don’t recommend that.
Oddly enough printing at to low a temp can also cause stringing. So maybe turn the heat up just a little (5 degrees?) And see if that helps both issues…
Either way they are looking way better then when you started and thats great!