Bad Z-Quality after heated bed upgrade on UM1

[mtbr 0]

Hey guys,

I've installed the UM heated bed upgrade to my UM original which works, except this problem:

http://umforum.ultimaker.com/index.php?/topic/2995-heated-bed-for-the-ultimaker-1/?p=73471

At the moment I'm experiencing a really bad quality in z-direction:

http://umforum.ultimaker.com/index.php?/gallery/album/1414-z-quality/

The walls are running straight into x- or y-direction.

Some of the x-walls are ok, some are really bad.

Also some of the y-walls are ok and some are really bad.

My settings were:

faberdashery filament (which worked really well before)

200 °C nozzle temp

0.1 mm layer hight

40 mm/s print speed

2.85 mm filament diameter

100 % flow

also the nozzle size is ok

Any ideas where I should start to search?

Edit:

the hight of the walls is 20 mm

 

[anon4321 16]

Is the new platform rigid? Is the glass secure? Something is appears to be shifting during your print.

 

[Dim3nsioneer 558]

The problem is not completely unknown with this heated bed: http://umforum.ultimaker.com/index.php?/topic/6907-horizontal-banding-on-um2/

Things you may check:

- Does the z stage run up and down smoothly?

- Does the z nut have some significant play?

- Does the fan blow directly onto the nozzle?

- What do the walls look like when printed without fan?

- How stable is the nozzle temperature (some colors, including white, are more sensitive to the temperature).

- Compare the layers: do some show effects of under- or over-extrusion?

 

[Daid 306]

Other possibility:

Z motor driver overheating?

 

[mtbr 0]

Is the new platform rigid? Is the glass secure? Something is appears to be shifting during your print.

 

as rigid as it was designed to be - in my opinion it is a bit too flexible, yes and yes

@Dim3nsioneer:

The problem is not completely unknown with this heated bed: http://umforum.ultimaker.com/index.php?/topic/6907-horizontal-banding-on-um2/

Very interesting link. Was this a beta version of the heated bed kit? The first post was from 2013. My banding looks quite similar. My feeder works perfectly as far as I can judge it (the imprint in the filament looks perfect). I've noticed that the new stage on the one hand has no play and runs absolutely smoothly, but on the other hand it is not very rigid in z-direction. The aluminium table seems to be quite flexible in z-direction (my taste). But I'm sure that it has been tested and should work well as it is. Anyway, i will try to reinforce it and see if it helps.

Things you may check:

- Does the z stage run up and down smoothly?

yes, absolutely

- Does the z nut have some significant play?

not at all

- Does the fan blow directly onto the nozzle?

no, and I have not changed the fan or the fan duct.

- What do the walls look like when printed without fan?

I have to check, but I would be surprised if it would help as my cooling setup worked well before the upgrade.

- How stable is the nozzle temperature (some colors, including white, are more sensitive to the temperature).

If the displayed temperature on the ulti controller is correct, it is absolutely stable

- Compare the layers: do some show effects of under- or over-extrusion?

I can't see under-extrusion but some parts look (to me) like over-extrusion and there is much stringing although retraction was on (original retraction settings of cura). Really strange are also the knobs that come out of the walls - I've added pics 4 and 5 to my gallery. I haven't experienced that before.

http://ultimaker.ipbhost.com/uploads/gallery/album_1414/gallery_17653_1414_3461474.jpg

http://ultimaker.ipbhost.com/uploads/gallery/album_1414/gallery_17653_1414_3401690.jpg

 

Other possibility:

Z motor driver overheating?

 

mhm interesting idea. I would wonder why, as my Z-stages runs really smooth. For the sake of completeness, my board version is 1.5.6.

Daid, do you have an Idea about my bed-temp issue?:

http://umforum.ultimaker.com/index.php?/topic/3755-horizontal-banding-elimination-thread/

 

[mtbr 0]

I've added two more pictures:

http://ultimaker.ipbhost.com/uploads/gallery/album_1414/gallery_17653_1414_953817.jpg

http://ultimaker.ipbhost.com/uploads/gallery/album_1414/gallery_17653_1414_1253553.jpg

there you can see a riffled structure. maybe a sign of vibration?

 

[Dim3nsioneer 558]

I have the banding problem on my UM2 which is the origin of the HBK. I don't know exactly what you mean with the first post being written in 2013. If you mean the link in the first post of my banding thread, it's a thread on the general nature of horizontal banding which was also a problem sometimes on the UMOs.

 

[mtbr 0]

I have the banding problem on my UM2 which is the origin of the HBK. I don't know exactly what you mean with the first post being written in 2013. If you mean the link in the first post of my banding thread, it's a thread on the general nature of horizontal banding which was also a problem sometimes on the UMOs.

 

Sorry, I messed the links up, instead of your thread I read the previous thread started by Nick Foley here:

http://umforum.ultimaker.com/index.php?/topic/3755-horizontal-banding-elimination-thread/

Now I've also read you thread, your issue looks a bit similar to mine.

My theorie is, that with the (maybe worst case) settings that I've used the bed was induced to an elastic z-movement, thus resulting in the artefacts of the riffled structure (post #6). I've checked the stiffness of the bed by putting a mass of about 1 kg on the front of the platform, which effected the platform to elastically lower about 1mm. Thus, 50 g would force the platform to lower about 0,05 mm (half the layer size I used) - if applied on the front of the platform. I can imagine, that when wiping the melted filament off while printing, a changing z-forze (due to a changing upper surface of the previous layer) is applied on the platform, thus resulting in a z-movement of the platform. This may be an explanation for the riffled structure, but I don't know if it explains the other z-artifacts.

Unfortunately I'm quite busy the next days, so I can't test it right away.

Any ideas where the knobs (post #5) could come from?? Maybe this is the key?

 

[Dim3nsioneer 558]

Imagine you put the weight on top of the wooden z stage, it would have gone down even more. If you put weight on the z stage, you should do it in between the z bars:

This ist significantly easier with the HBK z stage (just put it below the cover). If you really see a difference in the print quality you may want to check if all the screws of the z stage are tightened reasonably.

The blobs are artefacts from the strong stringing you have. Try to reduce the printing temperature and speed. The z hop feature with two times the layer height might also improve the situation. Make sure the nozzle is clean on the outside. And last but not least check your model in the layer view of Cura if it has some weird slicing results at the spots in question. If it has, it's a defect of the model.

 

[mtbr 0]

Imagine you put the weight on top of the wooden z stage, it would have gone down even more. If you put weight on the z stage, you should do it in between the z bars:

 

Hi, the reason why I've put the weihgt on the stage, was not to tweak my results, but to measure an elastic vertical deformation of the platform due to bending of the platform. Thus, when I would have put the weight between the rods, there wouldn't have been any vertical deformation, as the platform is rock solid in that area.

 

The blobs are artefacts from the strong stringing you have.

 

I've added a new picture, where I've marked the blobs/knobs:

http://ultimaker.ipbhost.com/uploads/gallery/album_1414/gallery_17653_1414_211132.jpg

in the middle of the wall there should be no reason for stringing, just a straight line of filament?!

 

[Dim3nsioneer 558]

I've added a new picture, where I've marked the blobs/knobs:

http://ultimaker.ipbhost.com/uploads/gallery/album_1414/gallery_17653_1414_211132.jpg

in the middle of the wall there should be no reason for stringing, just a straight line of filament?!

 

It's surprising where Cura sometimes decides to stop a line, travel somewhere else, print there and then travel back and continue the line... especially with infill... it can be in the middle of a wall.

 

[mtbr 0]

It's surprising where Cura sometimes decides to stop a line, travel somewhere else, print there and then travel back and continue the line... especially with infill... it can be in the middle of a wall.

 

I've had another look at the toolpath in the layer view of cura. The walls (or at least the outer shells of them) are printed as straight lines as I've expected, but the walls are crossed by the printhead several times when it moves between other walls. I guess that material is wiped off at the wall when the nozzle crosses the wall. So this might be the explanation for the blobs/knobs.

The other issues must wait some days until I can start testing. Thanks for your help up till now !

 

[derpeter 0]

Hi mtbr

we have exact the same problems. Sometimes it looks like the printer is missing a Z move. If the object grows the nozzle starts moving inside the already printed PLA. I also observe some beding on the edges which is the oposite i expected from the heated bed.

I'm wondering if the manual is wrong and we need to select ultimaker+ instead of ultimaker.

 

[gr5 2,224]

Sometimes when Marlin is updated the values in eeprom are not updated. Marlin looks at the old version and the new version and only resets them all if it is a major version change. So you might want to do a "reset to factory defaults" which will reset all the motion settings (vmax, jerk, steps/mm, etc).

 

[mtbr 0]

Hy guys, thanks for the news. I didn't have time to do some further testing. At the moment my z-stage is disassembled, and I'm in contact with the support to solve my temp problem:

http://umforum.ultimaker.com/index.php?/topic/2995-heated-bed-for-the-ultimaker-1/?p=73471