2nd print on UM3E, what is wrong ?

[alex jaspers 8]

Hi there,

 

I have printed a senseo cup holder as shown in attached pictures.

Using Ultimakers original  PLA and PVA and slicing in Cura 3.1 at 150um and 40% infill, rest std. settings as shown in picture.

UM3E firmware: 4.1.4.20171204

 

The result is somewhat a surprise disappointment 

- war-page of both rear bottom corners

- different layer thickness visible at several locations

- some (openings)

 

* Why so much warpage in both rear corners (see one detail of rear left corner) , d wall = 2 mm and I closed front of UM3.. and it is PLA .. ??

* The detail picture  of layer build up of right side shows a zone (2) with layer thickness of ca. 300 um , in zone 3 it is the actual 150um as set in Cura

* at a height of ca 70mm there is a colour band visible, indicating density change period... tension change on feeding filament or ??

   > the total print was done during night time.

*  Is the filament guiding of material 1 (outer spool) ok ?, It is done as in manual but to me seems a possible unstable setup ..?? (filament stropping ...)

 

Any advise would be great.

I did not tinker any setup but  the part design is not that complicated or has wall thickness variations that can be troublesome.

 

Thks

Alex 

 

Gcode file header: 

 

;START_OF_HEADER
;HEADER_VERSION:0.1
;FLAVOR:Griffin
;GENERATOR.NAME:Cura_SteamEngine
;GENERATOR.VERSION:3.1.0
;GENERATOR.BUILD_DATE:2017-12-05
;TARGET_MACHINE.NAME:Ultimaker 3 Extended
;EXTRUDER_TRAIN.0.INITIAL_TEMPERATURE:205
;EXTRUDER_TRAIN.0.MATERIAL.VOLUME_USED:77407
;EXTRUDER_TRAIN.0.MATERIAL.GUID:506c9f0d-e3aa-4bd4-b2d2-23e2425b1aa9
;EXTRUDER_TRAIN.0.NOZZLE.DIAMETER:0.4
;EXTRUDER_TRAIN.0.NOZZLE.NAME:AA 0.4
;EXTRUDER_TRAIN.1.INITIAL_TEMPERATURE:100
;EXTRUDER_TRAIN.1.MATERIAL.VOLUME_USED:10871
;EXTRUDER_TRAIN.1.MATERIAL.GUID:fe15ed8a-33c3-4f57-a2a7-b4b78a38c3cb
;EXTRUDER_TRAIN.1.NOZZLE.DIAMETER:0.4
;EXTRUDER_TRAIN.1.NOZZLE.NAME:BB 0.4
;BUILD_PLATE.INITIAL_TEMPERATURE:60
;PRINT.TIME:57742
;PRINT.SIZE.MIN.X:9
;PRINT.SIZE.MIN.Y:6
;PRINT.SIZE.MIN.Z:0.27
;PRINT.SIZE.MAX.X:213
;PRINT.SIZE.MAX.Y:207
;PRINT.SIZE.MAX.Z:116.97
;END_OF_HEADER
;Generated with Cura_SteamEngine 3.1.0

 

 

 

   

 

[SandervG 1,521]

Hi Alex, 

The difference between zone 1, 2 and 3 are manifestations of the active leveling I think. 

PLA shouldn't usually warp that bad, so I am inclined to think that the adhesion to the bed was insufficient. Did you use any adhesives such as the glue stick?

 

I usually interpret the color band difference as a change in temperature. Perhaps something with the fans?

 

The set up looks alright, what you want to prevent is the filament from reel one, getting on reel two, and block that reel. 

Usually, it should/will be a bit tighter during printing. A large retraction might retract it like your photo (but it looks like a lot), but during regular printing it should go back to a tidier form around the reel again. 

[alex jaspers 8]

Hello Sander

No, I did not use a glue stick or lacquer, hmm that can indeed be the issue in case of the high side walls

But this irregular layer thickness is strange...

 

[neotko 1,417]

PLA doesn’t need the door. The excessive heat can make the layers don’t cool correctly and what looks like warp could be a pla overheating issue. Basically if it’s too hot it doesn’t stay on the crystal temp.

 

the layer issue does look like active leveling failing. I would do a perfect manual level and try again without the door. 

 

Glue cant hurt indeed but it can be very difficult to remove the print without removing the glass. Remember to cool it properly or the glass will damage when removing the print if you use pla and glue stick. 

[SandervG 1,521]

1 hour ago, alex jaspers said:

But this irregular layer thickness is strange...

 

You are referring to the zones 1, 2 and 3? 

Have you tried to make a print without active leveling turned on, to see how that influences your print? You could print just the first 5 cm and see how that goes. That should give you the information if active leveling is involved or not. I also forwarded your post to our experts and awaiting their input too. 

[gr5 2,170]

So many issues...

 

Well the number one thing to do is get rid of the cover like neo says.  Unlike most materials, PLA softens around 52C and you want it to get down at least to 60C before you print the next layer on top.  So 100% fans is desired as soon as the first layer is done or within a few layers anyway.

 

One of the photos shows a bottom corner "warping".  What's happening is when it is printing about 5 to 10mm from the surface these upper layers are cooling and lifting very hard on the corners - pulling the upper layers inwards.  This is EASILY fixed and best fixed by having the part stick like hell to the glass.  I recommend a very thin, invisible layer of pva glue.  One way is to use hair spray but then you have to remove the glass from the printer so you don't get glue in your printer.  Another way is to use the supplied glue stick, spread it around a little with lots of gaps showing bare glass.  Then wet a tissue and spread this around thoroughly (and also a side benefit - removing most of the glue).  Or my favorite method - mix elmer's wood glue with 10 to 20 parts water (it's not an exact thing) - mix this up in a jar - shake well - then paint this watery glue onto the glass with a paint brush.  heat the bed to 60C and it will dry roughly in the time it takes to get to 60C.

 

And also make sure you are squishing the bottom layer into the glass well. Consider maybe turning the 3 leveling screws 1/3 rotation CCW to move the glass up and closer to the nozzle.

 

Those are the easiest two things to do...

[gr5 2,170]

The different layer thickness might be because of auto-leveling.  I don't use that and don't recommend it if you want your prints to be as "perfect" as possible.  The way it works is it compensates for a tilt to the glass and so it prints at an angle the bottom layer and then slowly adjusts until things are level over the course of some height - I think maybe 10mm.  This means the bottom of your part is going to be crooked and some sides will have the layers closer together and other's farther apart (for that bottom area - again I think maybe 10mm).

[alex jaspers 8]

Dear all,

Thank you for clear feedback.

Biggest mistake was indeed to close the cover, im case of PLA and it's rather low Tg no fast cryt. of printed layers could be reached.

It made bottom part very irregular as a result.

 

Regarding  autoleveling. I see.. This technology has it's limitations due to visual impacting when correcting for tilted surface.

I will do some more tests and measure the actual layer thickness changing for better understanding it's visual impact/limitations.

 

[SandervG 1,521]

Hi @alex jaspers , hope you are well. Did you do any further printing and investigating to the artifacts you saw on the above print?

I have shared your thread with our product experts for their input and I would be happy to share their feedback with you. 

 

A lot has been said already, but there were a few additions.

 

What you have labeled as zone 1, 2 and 3; zone 1 can indeed be influenced by active bed leveling, although no longer in zone 2. It is only potentially visible in the first 10mm of a print. 

If your print would be overly tilted, it could seem like on the lower side of the bed, that the layers are slightly thicker, where it compensates for the offset. 

 

 

The impression was that the layers were not of irregular height but minimally misaligned. It can be slightly seen on zone 2 in this photo. Why? Could be due to 2 reasons. (we suspect). 

Option 1, up until zone 2, it seems like the print was mostly a big square. Then a big hole appears on the front. It could be that due to suboptimal jerk/acceleration settings the lines get slightly misaligned. 

 

Option 2 could be that when the hole appears in your print on the front, this is where PVA was first introduced. This also comes with additional cooling. When it is just PLA, layers are cooled more equally. It would make sense why the bottom layers look more aligned, and why they also look more aligned above the top line.

 

Option 3, which may be a long shot, was that your switch bay might be calibrated a bit tight, which might push your print head slightly off-set with friction. Not likely, but perhaps worth a shot to look into. 

 

Finally, I would also like to propose that would print your model as well and we could compare results. This could potentially confirm or rule out any hardware. 

Feel free to send me a direct message with your gcode attached. 

 

If you have any further questions, feel free to let us know! 

 

 

 

 

 

[alex jaspers 8]

Hello Sander

thank you for your feedback.

I am very late on reply due to extreme bussy traveling irt job.

I have done a few more prints but other parts that where needed for business.

I will come back on this specific issue as soon I can free up some time.

thkx

alex