Calibration/symmetry advice please

I'm getting reasonable prints now, but, as you can (I hope) see in the image something is wrong.
It's a circle, but not as we know it!

I've got rid of the gaps caused by loose belts, and the extruding seems to be correct, but there is a consistent divergence from a correct circular shape (esp around the 11 oclock position)

I'd be grateful for any suggestions.

Thanks.

Edited 1 time(s). Last edit at 05/11/2012 04:53AM by jman.

Ensure that your x and y steps per mm in firmware are correct. Basically, you command a move on an axis (say 100mm), then measure what it actually moved. The new steps/mm value is: (Current Steps/mm)*[(Commanded Distance)/(Actual Distance)]. Repeat as needed to get it just right. There are also several helpful videos on youtube that show the process in detail.

Thanks for the reply.

I guess I shouldn't have said 'calibration'. While there was a slight oval distortion, the problem I hoped would show in the photo is that there are irregularities around the arc of the circle which are not aligned with the axes, or even at 45degrees.

That is, it's not a smooth, regular circle, and so the thickness of the ring varies from 6.6mm to 7.2mm.
It's a 'lumpy' circle!

I have fine-tuned the calibration, and the thing is (on a sort of average) 33mm across like it's supposed to be, but it's not round.

So I dunno... but was wondering it it might be some sort of friction effect along the axes.

Edited 1 time(s). Last edit at 05/11/2012 04:54AM by jman.

Are you using Sprinter or Marlin?
Marlin produces far smoother circles and rounded edges.

Try re-lubing the smooth rods.

---

My Reprap blog

jds-reprap.blogspot.com

my circles looked similar when running on sprinter. Now they look great on marlin

---

Ryan
Quality Engineer & Hobbyist
thingsandtrains.blogspot.com

Thanks Ryan and Joe.

I'll give both those approaches a try on the weekend!

In my case it was backlash:

[forums.reprap.org]

-Rob A>

Just to clear up, the problem turned out to be a slipping small gear, and belt fouling.
I've made sure the motors all have flats on the shaft, and changed to metal gears and T2.5 belt. The latter may not have been necessary, but I thought I'd give it a try.
So far so good.