Uneven layers
-
@antlestxp said in Uneven layers:
8bit with octoprint vs maestro on the same printer. I will admit 2 different brands of abs and re-sliced to account for the different fw flavor.
-
Did you slice using the exact same layer height in both?
-
Have you run heater tuning on the bed heater? If you haven't, then it will run in bang-bang mode, which often contributes to layer banding.
-
Are you using the same Z motor current in both cases?
-
What stepper drivers does your 8-bit board use? If they don't support stealthChop, or it doesn't use them in stealthChop mode, then it's possible that the default low-speed stealthChop mode of the drivers on the Maestro could be reducing the Z axis precision. So try configuring the Z drivers not to use stealthChop (see the M569 command).
HTH David
Duet WiFi hardware designer and firmware engineer
Please do not ask me for Duet support via PM or email, use the forum
http://www.escher3d.com, https://miscsolutions.wordpress.com -
-
@dc42 In my case it's as follow:
- Yes exactly the same slicer settings in all cases.
- No bed heater installed yet.
- N.A
- Is this something that I should try? I'm using a Duet Wifi with an Duex2.
Also I just tried with diffrent microstepping and with/without interpolotion. Same results.
Tried with flexible couplings and with stiff couplings on the steppers as well, same result.I'm comming back to that this could be a firmware or config issue, but I'm out of ideas =/
-
what do you mean n/a for question 3?
-
@veti Dave was replying to @antlestxp as he had changed from an 8 bit controller to the duet maestro. I have only used the duetwifi in all examples
-
- yes same slicer settings with the exception of relative vs absolute extrusion.
- i used bang-bang on 8bit and PID on the maestro
- same current on both (the maestro runs so much cooler, pretty happy about that)
- Trinamic 2130 (had a love hate relationship with them)
I will get some single perimeter prints going with and without stealthchop and see what happens.
-
@haggan90 I don't think it is a mechanical problem. we are using 2 different printers and see the same issue. My peaks are exactly 4mm apart. Looking at a print a while back with the bad bearing I see that the bulges looked like they were the same distance but they actually weren't. Also where they started were different every print. Yesterday I printed several different models of different shapes and sizes and they all start at the same spot. I even removed and reinstalled the z leads and it looks like they are still in the same position.
-
@antlestxp said in Uneven layers:
My peaks are exactly 4mm apart.
I presume you mean the horizontal bands have peaks 4mm apart in the Z direction.
- What is the pitch of the Z leadscrew on your printer?
- Do you have 1 Z motor or 2? If 2, how are they connected to the Duet, and what is the model number of the motor(s)?
-
The bulges match the pitch. The lead screws are 2mm pitch 2 start. There are 2 screws and 2 motors connected one to each connector on the duet. I will have to dig up the model of the motors but I know for sure they are 1.8 degree 40Ncm.
-
16th steps no interpolation. spiral shell at .2mm
-
awe man... i hate to admit my issue was mechanical after all... I must have knocked one of the z leads off center doing the board install. I stripped the z axis down, clean, lubed and the next print was flawless. hopefully @Haggan90 your problem is as simple as an alignment issue as well.
-
@antlestxp Thanks so much for coming on and letting us know the cause!
-
@antlestxp I'm super glad it worked out for you!
I've abandoned my lead screws and aiming for a belt driven Z axis instead.. doesent matter what I do, I still get this imperfections. -
@haggan90 I'm gonna try a belt driven screw before I revert belt only. The 2mm pitch screw will prevent the bed moving, a belt only would need a worm gear to hold the bed in place in case of power loss. Maybe something to consider, can print most of the parts although I'm going all metal, with a bearing block to provide both radial and axial support for the screw, but that means I also have to redesign the x-axis (then again wanting to use openbuilds style anti backlash nuts (flat/square type) so would have had to do that anyway)
-
@haggan90 said in Uneven layers:
I've abandoned my lead screws and aiming for a belt driven Z axis instead.. doesent matter what I do, I still get this imperfections.
It might not be the lead screw of the belt. it could be friction because of the linear bearings.
your SCSXUU might be to short to give a stable bed try some SCSXLUU. -
@bearer that's awsome! I habe a system with a 3:1 gear ratio in mind with a 10mm belt. As I use 3 individual motors for the Z axis I don't think the bed will "fall" down during a powerloss, maybe it will slowly glide down
-
@veti if I use the longer bearings there is a chance of binding when I level the bed, as I use 3 motors for leveling. I've measured that the bed moves a maximum of +-0.02mm in X/Y during movement in Z.
-
@haggan90 I would think that you'd want the flex point to be between the rigid bearing and the bed plate rather than the bearing itself.
-
@phaedrux it's not possible in my design atm. Best case scenario would be if I used linear bearings for the z axis as well, then i could have all 3 points move more independent.
-
@haggan90 Perhaps consider using oldham style couplings to mount your screws to the bed to allow for some self alignment.
https://www.thingiverse.com/thing:2907060
