because i pulled it off while it was warm 
thats excellent, i have 1 mm diff between lowest and highest
Still bigger then the layer height.
The levelling you are getting of +/-0.2mm from average is just fine. That is doing exactly what the levelling is supposed to be doing - correcting for the deviation in the bed.
I’ve done this math before but don’t really care to do it at the moment - but calculate the flatness of the bed as deviation from a planar fit. The entire “flat” surface is slightly tilted to the side, and the deviation from that is actually even smaller than +/-0.2mm.
Overall, the bed is pretty flat. The comparison you draw to layer height is not super useful unless you are planning on printing with bed levelling turned off. If you desire to get the flattest bed possible then you would need to shim the carriage where it mounts to the linear modules to correct for the tilted plane.
That is alarming to me and could single handedly be explaining all of your issues, assuming the “tiny bit” you say is in the ballpark of .1-.2mm
That needs to be corrected ASAP.
Is the drop because there is so much tension on the card it’s lifting the head up forcefully? Or is there that much slop in the X axis that the entire toolhead is able to be lifted just from sliding a piece of paper under it?
If the head is loose then the force of the extruded filament will be lifting the head up instead of appropriately squishing the filament.
1 Like
not relevant based on my first layer testing
i agree that alarmed me too
1 Like
The module doesn’t appear to be loose. Appears to be the card “holding” up the head, then when it’s removed, it drops down. When I looked at the head earlier, it seems to be tight. What would I check to verify that all is tight?
It’s been so long since I last used a calibration card I can’t remember if that is normal.
I’d say if you firmly move the head up and down by hand, maybe applying around 5 pounds of force, there should be no perceptible movement or ‘ticking’ sounds.
If there is a ticking sound it’s likely caused by play in the internal bearings.
No movement or ticking in the head, x axis or module. Guess that’s good.
I’m thinking that since the up/down on the handheld screen is showing a minimum of 0.10, that the correct adjustment should be a fraction of that. When I calibrate it using the touch screen 5X5 matrix, it shows 0.05, but doesn’t when using the gcode G1029 directly from a terminal emulator or, I’m assuming from Luban, but I haven’t tried Luban yet.
Just did an 11X11 level, and it’s pretty much identical to the 10X10 level. The prints sure are coming off the bed nice though.
When you’re doing that part of setting the Z height you don’t have to use the touchscreen. I’ve always jogged using Luban via USB in a custom increment of 0.02mm. At least no bigger than 0.05mm. I think 0.10 is way too coarse for this type of adjustment.
Glad to hear the prints are a coming off the bed nice, sounds like that’s a good first layer.
First print came off acceptable, crossing fingers for the next print.
It’s not normal, it means he’s got one of two possibilities going on.
-
He has z offset way too close (most likely as it was also evident in his earlier bed test squares print)
-
the carriage slider is loose in the x axis or possibly z axis.
1 Like
This post made me revisit bed levelling, since as I wrote I suffer from a bumpy bed and have the effects of uneven first layer (which I yet ignored since they did not bother me too much). So I followed these guidelines to run a “manual” 11x11 grid calibration. I ran it twice to see how stable data is, and I was positively impressed how consistent the measurements are (see images below - when I flip between the two graphs, they are nearly identical but for a slight Z offset). After that, I did the touchscreen 5x5 grid and I can immediatly see that it is too coarse to take care of my bumps. And I guess that’s what I see in my first layers. Plan for the future:
a) do the next print with an 11x11 calibration (@gwfami perhaps you should try this too)
b) get washers to put between carrier grid and heated bed to smooth out bumpiness
And here are the results of above calibration runs:
Check for nozzle wear for both the height and orifice diameter. Certain abrasive filaments can wear the brass nozzle. The height of the print head back post can be adjusted, which can get too close to the plate and drag on the printed filament if the nozzle height gets too worn. Also check your bed to frame attachment. (It is possible to install the bed attachment frame upside down, which will warp the bed flatness. I know because I did it.-)
Carriage slider is nice and tight, as are all of the other parts with one exception.
Is it normal for the black block on the bottom of the print module to move? I took the cover off to see if there were any bolts on that side, there werent.
Normal
Moves up when pressed.
@gwfami something is loose for sure, mine doesn’t budge. It doesn’t have any relation to the nozzle itself, but it might have an effect on the sensor, I can’t recall if the sensor would be loose if that is. I’ve had a head torn apart for a while now and took out the sensor, I just don’t recall everything it’s fastened to.
@Artezio I assume that you have the new printhead then, as @gwfami I just checked with mine, and I have the same “loseness”. I’ve disassembled my printhead several times and remember that there is just nothing to hold the airguide firmly in place, but it is not an issue, this part is not under any mechanical stress.
1 Like
@Hauke yeah you’re right, I do have the new print head, I’ve never had the 2.0 head so I don’t know how it’s put together. I never would’ve thought that they just had it sitting in there though, that’s pretty odd design choice.
If you take the print head off, you will see a slot with a set-bolt that allows the back post, which does not move to be adjusted vertically.
Well, with the first printhead they have done more than one questionable design choice… Sucking in warm air for print cooling, the wobbly air guide, having the stepper motor axle of the feeder made the gear wheel so that when it is worn you need to change the whole stepper… plain stupid.
Just out of curiosity: Did they change the feeder gear? Is it now exchangeable when worn out? Or is it still fixed part of the stepper motor?
@Hauke ill have to look at the feeder gear, I never thought about it. It sounds like the 2.0 head was released in prototype form. The dual extruder module is also very noticeably different than what the prototype of it was when looking at the pictures, so hopefully it doesn’t have the same kind of prototype problems the 2.0 head had.
1 Like