Layers not the same width in a part that should be smooth

Hi everybody!

I have a problem that I can't solve. My prints are otherwise smooth but the layers differ in width. The thing is the center of the outlines of all the layers apperas to be at the same line but for some reason I have massive differences in the the diameter (both inner and outer) of the layers. Here is a pic of a test piece where I got it SUPER visible Sometimes the effect is almost gone, and then there are cases like this.

I use an external PID (Rex C100) for the temperature controlling and it is tuned in quite small variations of the temperature (less than 3°C). Material is ABS from China.

Is this from the slicer program or what the heck..? I use Cura.

Thanks for any suggestions!

Edited 1 time(s). Last edit at 07/27/2015 04:47AM by PropellerHat.

What type of printer do you have, what is the Z steps/mm, and what layer height are you using? On some printers, it helps if the layer height is an exact number of full steps of the Z axis.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

I have a printer of my own design. The Z-rods are M8 metric ones, on each corner of the bed driven by a single stepper (3A nema 23) with belt and pulleys having some gearing ratio (20T to 36T). The steps per mm on Z is 513. I have tried several layer heights, lately 0.2 and 0.3mm. The ribbing (just learned the correct term) seems to differ with different layer thicknesses.

Edited 1 time(s). Last edit at 07/27/2015 05:37AM by PropellerHat.

513 is quite a low value for the Z steps/mm on a Cartesian printer. Assuming you are using 16x microstepping, try layer heights that are multiples of 0.0312mm which is your full-step size.

Another possible cause of layer banding is cycling of the heated bed temperature, if it is fluctuating by a significant amount.

Yet another possible cause is inconsistent movement of the Z gears, for example if the teeth do not mesh very well. So try a layer height that corresponds to rotation by a whole number of teeth.

Edited 1 time(s). Last edit at 07/27/2015 05:42AM by dc42.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Do you use auto-bed-levelling?

If you have play in any axis, it can cause oddities in the other axes. e.g. play in the Z axis can cause unevenness in the X and/or Y axes if your slicer reverses directions between layers.

No I don't have autoleveling. The axes seem to be tight. This was my first guess when starting to hunt the reason. When closely studied the test pieces show the problem really to be wider lines on other layers than on other. If it would be misplaced layers the lines should just shift a little to some direction, but they seem to be centered right. This has to be a case of too much plastic on some heights of the piece whether it is due to uneven Z-moves or uneven extrusion. The beautiful sine wave shape of the problem (uneveness of the vertival walls) makes me think it's Cura in the end but whyyyyy...All the other slicers are just too horrible to use. Also I don't hear the extruder motor altering in speed at any moment, so a wavy extrusion is sort of out of the possible reasons..

Edited 2 time(s). Last edit at 07/27/2015 05:51AM by PropellerHat.

Ok it really seems to happen exactly on the pitch of Z-rods. And since I have 4 rods, in the middle of the bed the nature of ribbing is not so easy to see. This test piece is printed in one corner of the bed, so the problem is easy to see. Totally looks like my M8 solution is poor in the end.

Suggestions:

1. Does your print come out at exactly the correct height? If it is under-height, it could be that your motor is missing steps depending on where in the cycle it starts. The solution would be more motor current, or lower Z acceleration and/or jerk rate.

2. What is the mechanical arrangement of your motor, belts and pulleys?

3. With the power off, is the Z motor easy to turn by hand? It's easy to over-constrain a rising bed, especially when using 4 screws instead of 3, which will make the system stiff due to misalignment of the screw threads and the nuts.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

The print height is what I expect. I rebuild one corner and got a lot better result when printing at the same corner of the bed, though still a bit of ribbing. The rebuild was mainly to open up the threads of the rod with a thead tool. The surface of the rod was coarse and after rethreading it is shiny and nice. There appears to be a lot of play in the nut. It wobbles easily on the rod. It might cause the movement to be wavey instead of linear since the rod-nut contact does not stay the same. Probably the final solution is to carefully attach the belt pulleys to the Z-rods so they turn co centrically and then making a special nut with longer helix than in a normal M8 nut to keep it alignet to the rod for real.

Thanks for the help guys!

Is this a case of your layer height not corresponding well to your Z steps? Check out the "Optimal layer height for your Z axis" section of the calculator

Hello again. I have now been printing enough to admit my so called solution to the problem did not work. The ribbing occurs still the same. The ribs turn out to be exactly the pitch of my M8 Z-rod what ever layer hight I use, so the problem has to come from the rods. I am willing to update the Z-axle to some purpose made rods like Acme screws but since the rise is a lot bigger in everything else than M-standard rods I will have to change the drive system into some geared steppers and this calls for tinkering (and paying).

I have 513 steps per mm on the Z so steps going even with the layer thickness is not the cause of my problem (you could not see even an error of 10 steps I assume).

If somebody knows how to make a M8 rod work flawlessly as a Z-screw, please let me know

Could you post some photos of your setup - I haven't quite got it visualised yet. Do you have smooth rods as well as the four threaded rods?