@dstarke its definitely a Hail Mary, but I have seen stuff that wouldn’t have ever been thought of as a slicer issue but turned out to be. I just think it would be a good idea to either prove it’s a slicer issue (I really don’t think it it is, I think it is mechanical) or (the most likely) eliminate it as a possibility, process of elimination.
Confirmed. I get the same results with PrusaSlicer and Cura…
It’s possible the extrusion rate is slightly too high, in combination with filament that’s slightly too cool. The excess material has to go somewhere, but where it goes is limited by the density and viscosity of the molten plastic. There’s a relatively small parameter region where the excess material first goes to one side, then to the other, meandering back and forth. The machine position will maintain a centerline on the feature, creating a stable meander.
In order for this to happen, the material has to be cohesive enough not to flatten into an ovoid cross-section at each layer but remain more circular. If the material were less cohesive, it could flatten out more, which would happen if it were hotter. I think temperature may account for the pitch increasing as the pattern rises in z-height. The head as a whole will be getting hotter, meaning the filament cools off a bit more slowly because of radiant heat transfer, giving the filament a bit more time to spread out, which would alleviate the problem a bit.
Either of these small process adjustments might fix the problem. Adjusting both would probably be better, since both contribute to ultimate print quality. Whatever happened, you seem to have stumbled upon a fairly small parameter range where this behavior occurs.
Thank you for this detailed explanation @eh9.
Reasons why I do not believe that this is the issue here:
- changing the wall/infill print order did not change anything
- I am already printing a PLA+ filament at 220°C (the manufacturer recommends 190-220°C)
- the effect changes depending on the print bed position
- changing the flow rate from 100% to 88% had no impact
- halving the print speed changed nothing
- setting a line width of 0.4mm or 0.45mm made no change
- the error of around 0.15mm (>35%) is quite high
- there are a lot of areas that are totally unaffected in my prints
- the issue did not occur with small objects
I could also try other temperatures or materials but I highly doubt that this changes anything noticeable.
That is why I would rather like to focus on finding the malfunctioning linear module. Any idea on how to approach this is very welcome.
12 days have passed and I have yet to hear from the Snapmaker support. Plus I am struggling at finding a way to narrow down the linear module in question without reassembling my machine. In the meanwhile I have found out that z-wobbling looks quite similar to this:
Anyone with experience on this? Currently, my best guess is that the linear module for the z axis on the left has had it…
Usually Z wobble is associated with a bent Z leadscrew pushing the toolhead side to side. It’s possible but unlikely I think that your Z modules are that loose. Regardless, if you’re going to do a partial disassembly you might as well do a full disassembly and check all of the carriages for tightness.
I previously asked if your X or Y carriages were loose as they would explain that. I think Z is less likely, but also possible.
It’s easy to test for and “easy” to fix if you’re minimally mechanically inclined. You need to verify there’s a snug tension on the rollers as the carriage is inserted into the module on the rails. If there’s not, adjust the eccentric nuts.
1 Like
Thank you for these insides. If I actually had the time to disassembly the machine I probably had done so already and fixed up the remaining backlash altogether. Unfortunately, that is not the case. That is why I was hoping to find some more hints on this issue to identify the part in question before taking that step.
I don’t know how much time you have to dedicate to it, but it doesn’t take too long, at least a lot less than 12 days (where is snapmaker support anyways).
If you want to go that route we can talk more.
In the meantime, troubleshooting which exact part is the cause is easiest when at least the X axis is removed if you suspect one but not both of the Z towers are experiencing a loose carriage or anything else that would cause Z wobble symptoms.
have you seen @sycto’s video where his bed goes up and down? what do you see? - YouTube, from his thread here: Why does an .2mm initial layer measure .4mm +? - #15 by scyto
That’s the type of wobble to check for in the carriages on all linear modules, a subtle wobble. It can sometimes be seen (in scyto’s case) under powered movement, and sometimes it’s easier to detect by firmly rocking the carriages front back and side side and seeing if there’s any discernible movement.
If it’s not that then we’re still at square one.
There have been so many issues with linear module carriages loosening up though, I’m highly suspicious that’s the problem.
1 Like
also if the Y axis can do what you see to my builld plate video; then the same is potentially true for z and x axis - aka the head would move backwards and forwards…
if a similar issue was on the Z axis i would expect it to translate to minor layer shift every few layers
if a similar issue was on the X axis i would expect it to translate in single layers varying slightly along the X axis of a layer
(unless my mental model is borked, lol)
/edit - oops realize thats what you meant, sorry slow brain day
1 Like
Thank you @scyto. So I am probably dealing with a loose carrier as @brent113 mentioned. It is good to know that I can narrow it down to the Z axis linear modules. Most likely. I was just a bit confused about the actual influence on the print. Switching off z-hop on retraction should had kept it from moving back and forth on the Z axis. But this had no affect. Moreover, the issue occurred here:
but not here:
Anyway, thank you everyone for all these hints. I will try to find some time to fix the (probably) loose carriers and come back with the results. Not sure when this will be, though.
1 Like
What would be the best way to tighten the loose carriers?
a) Start by removing the motor block and than the lead screw (like in this video)?
b) Remove the other end and pull out the carrier there?
c) …?
P.S.: Still no response from the support…
Mine has always had some play. For printing as the nozzle isn’t touching anything that provides upwards pressure it shouldn’t be an issue. Where it can be an issue is calibration of course… is yours moving visibly up and down as it travels left and right? (Attach a small rule and you should be able to film it). (This is a test I plan to do along with seeing if the tool head moves backwards and forwards as it travels left and right, oh and the same test with the z rails as this would mirror the motions I see on y rails where the platform seems to rise and fall).
I have the same issue, a rising spiraled pattern in the direction of the nozzle path. Happens it with every filament (using Hatchbox exclusively).
@snapUser what you are showing is typically an extrusion issue, and probably isn’t related to the topic of this thread. You should create a new thread with a much closer, and more clear shot of the top layer. If you have a shot of the first layer, that would be very helpful as well. Also, have you calibrated the E-steps, and checked to make sure that the values are store in the Controller? Installing firmware can erase your settings.
Printer e-steps was just reclaibrated, and same results. Here’s first layer. I do think it is same issue as OP… upward spiral pattern on faces.
Try using the web form instead of support email, someone else had issue with email and got much faster reply with web form. When I email I always use the zendesk email address instead of support@snapmaker.com.
@snapUser first layer looks great, I also don’t think it’s over extrusion, but it’s also not quite the same issue as OP. Create your own thread and we can help you better there, to prevent this post from becoming cluttered.
@snapUser while this first layer sort of looks nice, it also looks like the head is way too close to the bed. I don’t see a good bead all around, which would mean a slight under extrusion. The thickness of that first layer is much thinner than it should be, and that tiny pinhole in the front right of the square seems to indicate the same thing. Also, check your extruder gear for any debris. It should be shiny metal all the way around with no filament at all ground into it.
more random thoughts
- if you print the same wall thickness and no fill does it do the same?
- if you print spiralized vase mode does it fill the same
- i wonder if you X rail is at an angle - one way to test - measure thickness of that first layer at server points.
- i doubt its this (as SM doesn’t use bang bang), but wild that fluctuating bed temp can affect vertical surfaces
- is it all filaments / models that do this?