Raised corner problem

Hi everyone,

I've been having some problems with print quality - specifically, my corners turn up as the print progresses :-( See attached image. (How do I put it inline?)

The STL model has square corners, and the cube in the picture was sliced by Skeinforge (although I used to use Slic3r, and that wasn't any better; I switched to Skeinforge to see if that would fix this). It seems like a bit too much plastic is being extruded at the corners, and/or the printer is cutting them. The problem gets progressively worse as the print height increases, so that after a while the head is pushing the corner out of the way to lay down the next layer. So it gets messier and messier. The corners get progressively less square as we go up, although that could be the previous layer pulling it in, or something.

I don't think it's upwards warping; it happens even when the corners do not warp. See the var visible corner in the picture.

Someone suggested it was looseness somewhere in the axes, and the direction change was the problem. I've tightened everything I can find, although the fit of my 0.35mm J-head to my Wade's block is not perfect. If I slow right down, the problem doesn't happen. (I have a minimum layer time of 15s; if I just print a shell, it therefore goes very slowly. The print for that is very good.)

Is there a Marlin firmware setting I can set that would make it slow down when changing direction?

Any ideas? I want to make sure it's not something else before I go to the expense of obtaining a new Wade's block and J-head.

Gerv

I had this issue, but to a lesser extent. After much playing with calibration prints etc.. It has almost gone on every layer although my prints stil aren't perfect yet . I doubt its a hardware issue.

I'm not 100% sure which setting fixed it, but re calibrating my extrusion multiplier and also tweaking my z-lift were around the time the problem minimized/disappeared.

BTW: still using slic3r as skeinforge is still a bit steep a learning curve at the moment, and cura didn't really work for me. I really like the repetier/slic3r mix, I just wish a few more of the bugs were ironed out.

Edited 1 time(s). Last edit at 04/03/2013 05:04AM by twisted.

It looks like you are not getting good enough adhesion on the bottom layer, the problem just gets worse on the object if there is no change in shape as you hieght progresses. Have you tried blooming the bottom layer, slic3r, cura and skienforge all support this feature. In Skienforge just use a few loops on the bottom layer at zero distance. It does loook like you are pushing a little too much plastic out so worth reading this triffid calibration guide

Oops, didn't notice the lifting for some reason. If using PLA, I found a quick wipe of the glass with acetone before printing normally fixed sticking issues with small prints, or a mix of PVA glue and water brushed on the bed for larger prints.

My £0.02 ....
Gerv - I am not too sure what you mean by " I don't think it's upwards warping..." as prints cannot warp downwards. Two things which you could try:
1. Increasing the heated bed temperature so that the lower layers stay that little bit closer to the glass transition temperature and do not contract so much as they cool
2. Slow down the first layer speed although if you have a print with a large bottom layer surface area, you increase the potential for a blocked JHead. This can occur if the extrusion rate is not sufficiently high to prevent the PLA getting to a "soft" temperature higher up in the JHead towards the grub screw.

Alan

EDIT
Two replies sneaked in whilst I was composing mine - Hexitex, What do you mean by "Have you tried blooming the bottom layer.."?. Done a search for "blooming" and not come up with anything other than your post.

Edited 2 time(s). Last edit at 04/03/2013 05:41AM by Alzibiff.

I did go through the calibration guide (and that caused me to replace my hobbed bolt, which wasn't gripping properly). I'm fairly sure now that if I tell pronterface to extrude 10mm of filament, it sucks in 10mm. I haven't done Triffid's E Steps Fine Tuning yet; but I don't think that would cause a problem of this magnitude, would it?

The bottom layer adhesion is pretty good - at least, getting the prints off is hard! I did have more bloom on the bottom layer with a slightly lower first Z-height, but I changed it because the squished bottom layer looked ugly. I had this problem both before and after that change, so I don't think that's it.

Is there a "bloom the bottom layer" setting, or do you just mean lower the starting Z-height a little?

Gerv

twisted: the problem occurs even on corners which don't lift. I'm using glass cleaner to clean my glass at the moment; it seems to work better than acetone, is cheaper and smells a lot better! As I said, I have to blow hair-dry my prints at the end before they will come off the bed so I don't think I have an adhesion problem in general.

Alzibiff: the lower layers of the print end up larger than the higher layers, if anything, so I'm not sure how contracting-lower-layers can be the problem. Can you explain more?

I've already slowed down the first layer speed a lot. I have no problems there; the first layer goes down beautifully flat. Whatever problem it is increases linearly in magnitude as the print grows. As you can see from the picture, the higher you get, the more the corner is bent upwards.

Gerv

It looks like too much plastic to me.
Calibrate with 100mm.

---

Bob Morrison
Wörth am Rhein, Germany
"Luke, use the source!"
BLOG - PHOTOS - Thingiverse

Yes 10mm is too short for accurate calibration.

The curling inwards looks like it isn't cooling enough or the bed is too hot. Do you use a fan?

Try printing 4 at a time and see if they come out better.

The poor corner at the front looks like retraction parameters are not optimised.

---

[www.hydraraptor.blogspot.com]

I agree with Bob. Too much plastic.
If you are still using Repetier Host, just head for the Flowrate slider under Manual Control and try reprinting at 95%, 90% or even 85%.
You'll quickly see the difference that reducing the excess plastic will make.

Thanks, everyone :-)

I'll recalibrate with a longer length of plastic.

The bed is at 70 degrees, or at least that's what I set in Pronterface. I don't use a fan during the build, but I did use one (a hairdryer) to get the object off the bed because I'm impatient. Is that likely to cause a problem?

Is there a HOWTO somewhere on optimising retraction parameters?

Gerv

Had this Problem too. The corners are the on in the back, right? There is not enough cooling in the back. Try printing slower. Or use the Fan-settings in slic3r, even if you don't have fan, just set minimum time for one layer to ~20s.

I get the problem in all 4 corners, although different amounts in each. I did try printing slower - I did one print at 1/3 speed for the entire thing. That did improve the problem but didn't eliminate it entirely.

Gerv

Just a quick note, I think 'blooming' is referring to the skirt settings. If you have a few mm skirt at 0mm from the object, it creates a one layer thick flat skirt/brim around the print to increase surface area and therefore increase bond to the bed.

Have you checked your filament diameter and input it correctly? How did adjusting the flow rate work out? Do you have a glass sheet on the bed too?

I have the same issue on my MendelMax, even after calibrating my extruder with 100mm of 3mm PLA filament. Did you happen to find a solution?

Edited 1 time(s). Last edit at 05/20/2013 01:11AM by Zangetsu57.

Our current suspicion is that the tip of the extruder is wobbling, either due to it not being secured well enough to the Wade's block or perhaps because the X bars are flexing.