Help on custom big bot, bad cylinder, calibration issue or what?

Hi, i need help becouse i cant get rid of the problem i have with my custom bot.
I've writed in section reprappres and upon advice of one member i write here, so sorry for cross posting.

My bot is 500x500x500 custom made,
frame is laser cutted (all 4mm steel plate) and then assembled. Firmware is sailfish (makerbot style electronic)

Z axis is runned by3 leadscrews TR10x2, a 5mm GT2 2MGT drive all the three leadscrews. A nema 23, 1Nm motor drive the z axis. Number 3 20mm linear shaft with bronze bushing keep z axis from moving on the xy plane during movement. Z screws are free to move on the top side (no ball bearings) and i have 2 oblique bearing (X configuration) on the bottom that take all the wheight.
XY movement is corexy with 2 nema 23 motor.A 9mm width GT2 2MGT belt drive the corexy (i about 2.8meters for a single belt, and i use CPC linear rails for the gatry. All motors run 24v with external digital driver. The belt is tight.

I had, with ABS and PLA, the same problem: when in a print i have a "cylindrical" feature, i have dots and zip on all the wall if it is bigger (about 30mm in diameter), and a very bad cylindrical shape if it is small (below 10mm in diameter). I've tried to use or not use random starting point, lowering the acceleration, the speeds, increasing the motors current, decreasing them.

On a big print the problem is less pronounced but is there, especially on the corners (90 degrees corner or also big radius ones)

I use the latest simplify 3d. The print in the photos are all 0.2mm layer height.

I cant go on with this problem. I've checked and z axis move smooth if i move it by hand pulling the belt. i've all lubrificated well. I've print a 300mm radius circla and its out 5/100 of mm in one axis.

I know that its seems a z wobble problem, but the frame is very sturdy and i cant think that 3 10mm leadscrews, free at one end, can bend 3 20mm linear shaft.

The photo 000346 shows the same model, on the left printed at 100 mmsec and on the right at 50 mmsec


I hope someone can help me!

Make sure your X and Y belts are tensioned well and your steps/mm are set properly.

---

greghoge.com

HUGE 3D PRINTER PARTS SALE!!!

I'LL check step mm but belts are very well tensioned. For comparison, i have a rep2x and belt on the rep2x are more loose than on my corexy printer.

thanks for the answer!

Could it be an extruder issue where it spurts out more and less filament at different times?

-os3dp

Is is printing the cylinders smoothly and continuously, or does the print keep stopping and starting?

---

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].

Thank you for the response!
Its seeme to me that the movement is pretty continous and smooth.

Today i've tried to slice that cylinder with makerware desktop. I've modified the json file of the replicator dual standard to meet my steps mm and other things (axis lenght ecc)
I've forgot to mention that i use micron 3dp extruder with 0.35mm nozzle, with 1/4 microstep (so it have about 240 step/mm).

So, with makerware i was astonished becouse the cylinder were good! so good i can see the z wobbling... the only defect, with the z wobbling, is some defect on the external layer caused i think by the starting of the new layer always in the same spot.

So i've tried to modified the simplify3d profile with some info found online and the result is much better than before. Not the best as the print with makerware.

In the photo attached, the cylinder on the left is the old one, the one in the middile is sliced with makerware and the one the right is sliced with simplify3d with the latest profile.

Now i have to trying to eliminate z wobble. If you look at the latest photos you see how sturdy should be the z axis. This is a 15mm aluminium with 20mm linear shaft, 10x2 trapeziodal screw and the black structure is all made by metal.
Z axis speed in simpilfy in set to 15 mm sec.
any hint?

Looking at the IMG_20150507_000407 photo in the previous post, i would like to ask if my z leadscrew support is valid or not.

What you see is an aluminium milled case for 2 oblique bearing 8x22x7 (chinese clone of SKF 708 bearing). They are mounted one on the other in X configuration. One bearing set on the lower steel plate and the upper one on the aluminium case.
Becouse i see a little movement from the bearing during direction inversion (ie jogging z stage up and down on inversion sometimes the leadscrew pull up the bearing of about 0.5-1mm) i put a little PLA printed washer (0.4-0.5mm) between them.

Is correct to use 2 oblique bearings? or was better one radial and one axial stacked?

It is hard to say valid or not. The SKF website shows it in the double stack arrangement, as you have it.

If the bearing is pulled up during a Z reverse that would suggest the sliding bushes are binding perhaps.

The screw is pulling on the nuts on the Z-bed and instead of the bed going down, the bearing is coming up.

Really gravity should hold everything down if it smooth enough to fall.

Are you sure it is Z wobble? It is hard to see. It isn't just bulges of more extrusion on some layers?

A cylinder shouldn't be difficult for any slicer.

Could it be temperature wobbles or something with the filament travel from spool to hotend?

-a

Tonight i will made some hidef photos so you can see better.

It seems to me z wobble becouse the pitch between the "bulges" is 2mm exactly. Its then seems to me that the bulges is more pronounced on one side of the cylinder and less on the other (i assume to divide the cylinder into 2 side of 180° degrees each, for clarity)

BTW i also have problem with temp stability, i cannot find good PID values for sailfish for micron3dp extruder with 24v cartdrige as they are different from the ones on a MK7 or MK8 extruder. I have temp swing of +or- 3-4 degrees

Little Update:
i've destroyd my camera (hit hard concrete floor) when i whas making some photo of the prints

But it was a little lucky becouse when i bend to recover it from the floor i take a look at the belt system that move the 3 leadscrew for the z axis.
The 3 pulleys were moving eccentrically! the one on the motors was not.
So after looking and thinking, i had checked the diameter of the part of the leadscrew where the pulley are inserted and i measured 6mm, where my pulleys are 6.35mm (imperial 1/4 bore).
So this could be the fact at the root of my z problem? At theory, tangential velocity is not mainteined constant on the pulleys now and the 3 leadscrews are not moving with the same velocity.