Please help identify reason for bad quality
May 24, 2015 06:17PM
If possible, can you guys help me identify where my prints are going south?

Pics atrached.

Prusa i3 w/auto bed leveling/ Repetier 1.0.6 / Cura
Print settings:
Print speed 20mm
Perimeter speed 25mm
Quality (Height) 0.2mm

Retraction speed 20mm
Retraction distance 0.8. (Possibly the issue, I have a Bowden setup w/Greg wade)

Extruder temp Have tried 185 & 175 both same results
Bed temp 60

I have perfect calibration on all axis, using a digital caliper, as well as the 100mm extrusion. You can see in the cylinder that it actually was printing perfect for a bit, but started to get nasty towards the top.

Any advice you guys can give would be greatly apprectaited. Thanks.

Edited 1 time(s). Last edit at 05/24/2015 06:18PM by usslindstrom.
open | download - 20150524_163633_resize_20150524_164241.jpg (29.3 KB)
open | download - 20150524_163932_resize_20150524_164241.jpg (32.9 KB)
open | download - 20150524_164019_resize_20150524_164241.jpg (32.4 KB)
Re: Please help identify reason for bad quality
May 24, 2015 08:30PM
I have a prusa i3 also. I've had crappy prints (that looked like yours) until I decided to completely tighten everything up.

Replaced my horrible ebay metal bearings with these


put set screws in the X axis motor mount/idler where the Z threaded rods go through, rebuilt my Y axis and attached the hole thing to a 5/8 inch piece of mdf

now its solid and i get good layering.

I have other issues with print quality that are purely me not setting the numbers right in Repetier host smiling smiley
Re: Please help identify reason for bad quality
May 25, 2015 02:54AM

Retraction speed 20mm
Retraction distance 0.8.

I have a direct drive and retract 4.5mm at 60mm/s ( higher speed and the right lenght are the key)
With a bowden setup, you might need even more length
I also lift z-axis after retract 0.075mm.

This will help with a few of your problems.
My prints got better, with belt tensioners installed. Better bearings for x &y are on the list, too.
Re: Please help identify reason for bad quality
May 25, 2015 03:07AM
What type of hot end? My previous one was a Ubis, and it worked fine with retract of 4.5 My Chinese E3D V6 needs it to be just 0.5mm

I guess what I'm saying is to not make big changes from what is working. Also, only change one thing at a time.

I'd try reducing the extrusion amount by 10% at a time until things start getting worse. Choose the best of these and set the extrusion amount to that. Then try increasing the retraction distance by 1mm at a time until things start getting worse. Choose another parameter (nozzle, temperature, retraction speed, bed temperature, X & Y travel speed and acceleration, ...), rinse, repeat.
Re: Please help identify reason for bad quality
May 25, 2015 03:46AM
...what Frank said!
And don´t print huge parts, before you are satisfied with the test-cube results.
To test retraction behaviour, you better print two cubes with a bit distance between them.

Edited 1 time(s). Last edit at 05/25/2015 03:49AM by o_lampe.
Re: Please help identify reason for bad quality
May 26, 2015 09:19AM
Thanks guys. Few things to try here. It'll take a bit to get through them...

On a side note, as I'm running through the gambit of things to do, another quality item that popped up, when printing a 20mm calibration cube, is that the 4 corners always have spikes. While I'm adjusting the other parts, is there something else I should look into as well?
Re: Please help identify reason for bad quality
May 26, 2015 09:21AM
Oh, and you guys asked about which hot end...

On the i3, I've gone through the mk7 that came with it, to a Pico (awesome, but accidentally broke it when tightening, and now I'm on a j head.
Re: Please help identify reason for bad quality
May 26, 2015 11:54AM
Wrong retract can cause spikes, but usually only on one edge.
Printing to fast can cause all edges to look like the extruder "overshoots"

I´ve reduced the "outer shell" speed to 30mm/s. The rest is 50-60mm. Full speed ( without extruding) is 100mm.
Re: Please help identify reason for bad quality
May 30, 2015 04:48PM
Thanks guys... getting better little by little. Just short of grabbing those ebay bearings, I've gone ahead and tightened everything up dramatically.

It's made some significant progress. One item I must have previously overlooked, is the base of my y axis was floating on one of the corners. You could push it down, or pull it up, and the whole bed would change dimensions from the nozzle. Everything has been cranked down and secured. For the z threaded rods, there was a bit of play with them as well, until I added a spacer that someone gracefully uploaded to thingiverse.

Everything is pretty solid, unfortunately, I'm still having the corner spikes on 90 degree angles though. Everything is turning out awesome, but it's like it's extruding too much in all of the corners. The extruder doesn't pause at all, so I'm extremely confused how or why it's happening. Pics attached. Other forum searches finds a few people who have experienced this phenom too, but nothing in the way of what had fixed it for them from the ones I could find.

Thanks for all the help guys, this community is great.

Edited 2 time(s). Last edit at 05/30/2015 05:00PM by usslindstrom.
open | download - 20150530_082807_resize_20150530_152025.jpg (83.8 KB)
open | download - 20150530_083657_resize_20150530_152024.jpg (90.4 KB)
Re: Please help identify reason for bad quality
May 30, 2015 05:01PM
^ Editud fur spellun
Re: Please help identify reason for bad quality
May 31, 2015 04:22AM
Is it a cube or a chinese pagode-temple? spinning smiley sticking its tongue out

I´ve had the same issues with my printer. I installed belt-tensioners and reduced "outer Shell" speed to 30mm/sec.
Much better now.
Re: Please help identify reason for bad quality
May 31, 2015 05:24AM

Try to reduce your flow rate now. Your sending too much material, reduce by steps of 2 %. Until no more chinese pagoda smiling smiley

Edited 1 time(s). Last edit at 05/31/2015 05:26AM by Laopa.
Re: Please help identify reason for bad quality
May 31, 2015 02:20PM
Haha. Nice catch on the pagoda comments.

I stepped down 2% at a time, until believe it or not, I hit 60% flow rate. The spikes all but vanished finally, but the lack of plastic made the surfaces pretty bad. *attached. Tons of gaps on the top especially.

So, it's probably a combination of flow rate and something else. For The Outer perimiter, I've Been using 20mm/s.

Side point, this prusa kit came originally with a direct drive mk7, mounted directly to the x carriage, but would jam a bit too much for my liking. I converted it to a Greg wade, and Bowden tube setup. Jams have completely dissapeared, but the quality definately leaves a lot to be desired.

I also attached a pic of one of the original calibration cubes that I printed after I assembled this unit. 100% perfect. What else could I be misding?
open | download - 20150531_124257_resize_20150531_124728.jpg (240.5 KB)
open | download - img20150531_125318.jpg (11.5 KB)
Re: Please help identify reason for bad quality
June 01, 2015 01:25AM

Look better.

Before doing cubes and cubes, you must do a final flow rate calibration.
use This help, this is one of the best (my point of view). when you can success with that,
we will check other problems.


Also check if your belt do not touch the U guides in X and Y, or you will have to smooth the U.

And then do a temperature calibration.

Different material = different temp.

CARE if you use PLA, adjust temperature for PLA!

Send us your Soft (Repetier or else ) configuration for speed.
Re: Please help identify reason for bad quality
June 01, 2015 08:26AM
Thanks for the words of encouragement... but unfortunately I have to disagree with you that it's getting better.

The previous pic didn't do it justice. I've uploaded some new ones. In order to make the spikes virtually dissapear, I brought the flow rate all the way down to 60%... this completely demolishes the rest of the cube. Plus, if you look closely, you can see the points are still present, so I should continue the path of reducing the flow rate still. Doing that would still make it worse IMO.
open | download - 20150601_065432_resize_20150601_065518.jpg (318.5 KB)
open | download - 20150601_065356_resize_20150601_065518.jpg (304.1 KB)
Re: Please help identify reason for bad quality
June 01, 2015 09:04AM

When you adjust a printer, everyone do the same mistakes and try to push as much material as they can, but to fine tune your machine it better to start by the opposite.

If you down your infill at 60% it mean you miss something in the extruder calibration, better check the 100 mm again. After you can readjust +/-5%.

After place all your speed at 30mm/s
The first layer down a littte at 25mm/s

And do the flow rate calibration. Then the temperature.

If not do that you will turn arround.

How many layers for the top?
What infill %?
What thickness is your first layer?
What are your wall thickness?
When you start, the 2 first loops bigger than your print, is the thickness constant?
can you show us a picture?
Re: Please help identify reason for bad quality
June 01, 2015 01:41PM
If you are using prontrface watch carefully as it changes direction around a corner. I switched to repetier host because prontrface would actually back up!!! While going around a corner causing spikes. Annoying. Now i never see that and have perfect corners with repetier host.
Re: Please help identify reason for bad quality
June 02, 2015 12:56PM
Haha... unfortunately, it's Repetier host already. smiling smiley

I went back to the drawing board based off your suggestions Laopa. My calibration numbers are spot on. 100mm travel in all axis directions literally makes 100mm, even in extrusion.

So, the only thing it could be left over is speed I started thinking. First, I read up on other systems like the utilimaker that have Bowden setups, and how they can supposedly hit 300mm movement speed. So, for testing, I tripped my jerk settings, and upped the print speed to 150. Corner blobs vanished, but I was probably going to have to go crazy on temp, since the extruder probably couldn't melt as fast as I was pushing, so the faces of the cube started to get pretty ugly.

Armed with that, I decided to go the other way. I put the flow rate back to 100% and dumbed the speed down to 5mm/s on everything. The cube took about an hour and a half, but it's finally flawless. *attached.

I guess the hard part is going to be finding the happy medium between crazy long prints vs quality. I'll gladly take suggestions. smiling smiley

*THANKS TO EVERYONE IN THIS THREAD! You guys definately make the community a nice place to seek information out.

Edited 1 time(s). Last edit at 06/02/2015 12:57PM by usslindstrom.
open | download - img20150602_111847.jpg (89.4 KB)
Re: Please help identify reason for bad quality
June 04, 2015 09:30AM

Your print looks much better sure but you still have some big problems.
You increase flow rate to 100% and down speed at 5mm/s.
When everyone can run in basic 25mm/s.
So for me it mean you will leave your plastic material heat inside the nozle far more too long.
Now as you say here. So much flow rate for so slow speed. Did you check your nozle temperature?
Start from cold using small thermometer and read, increase by 20 deg and measure and so on.
It would be very interested to see the result. If you are out of range, mean you want 230 deg, but in fact get only 180 deg, yes you can melt plastic, yes you can print, yes but at 5mm/s.

Please give us your temp result.
Sorry, only registered users may post in this forum.

Click here to login