Ultimaker S5 strange phenomenon

[chuckmcgee 7]

It's the included glass bed.  With this being the new S5, I haven't had it but a couple of weeks - so not sure the glass has any pits or imperfections.  The prints are no more stuck than they normally would be.  When the bed is cooled, they pop right off.

[model_dude 1]

On 8/1/2018 at 2:25 PM, jakepoz said:

I'm curious: is that an aluminum bed in those pictures, or the standard glass one. Could that be contributing to more heat transfer to the part itself and therefore the folding of those layers?

 

@model_dude I saw your thread here: 

Is this what you mean by lower-wall cave in? 

 

It kind of looks like it could be, but if it is it's a pretty severe case of it! I haven't seen anything that severe - it usually looks more like this:

 

 

Out of interest, what was the temperature of the bed set at @chuckmcgee?

Edited August 2, 2018 by model_dude
broken link

[chuckmcgee 7]

2 hours ago, model_dude said:

Out of interest, what was the temperature of the bed set at @chuckmcgee?

 

60 degrees on the bed.

[SandervG 1,521]

57 minutes ago, chuckmcgee said:

 

60 degrees on the bed.

 

When I check out your Cura profile you uploaded some time back, I see your heated bed is actually set to 70ºC. Excessive heat can also lead to walls caving in on the bottom of the print. I would recommend to bring it back to 60ºC. 

 

Some of the other superficial artifacts you see might be caused by cooling. Personally I don't have much experience with Pro PLA or esun material, but it looks like as if the infill shows through the outer walls. It could be that the walls shrink some due to cooling, and that is why infill starts showing. It could explain why a higher infill shows it less, the space between the infill is smaller, which gives it less room to shrink.

 

For your model, I think you could also use a layer height of 200 microns. It may help your overhangs, and the rest of the walls which just go straight up should also not be a problem.

 

Do you have any Ultimaker PLA lying around by chance that you could try?

 

Good luck!

 

[chuckmcgee 7]

19 minutes ago, SandervG said:

 

When I check out your Cura profile you uploaded some time back, I see your heated bed is actually set to 70ºC. Excessive heat can also lead to walls caving in on the bottom of the print. I would recommend to bring it back to 60ºC. 

 

That could be.  I never changed default and assumed it was 60.  I agree that 70 may be a little too hot and will definitely pay attention on the next print!

 

19 minutes ago, SandervG said:

Some of the other superficial artifacts you see might be caused by cooling. Personally I don't have much experience with Pro PLA or esun material, but it looks like as if the infill shows through the outer walls. It could be that the walls shrink some due to cooling, and that is why infill starts showing. It could explain why a higher infill shows it less, the space between the infill is smaller, which gives it less room to shrink.

 

For your model, I think you could also use a layer height of 200 microns. It may help your overhangs, and the rest of the walls which just go straight up should also not be a problem.

 

I will definitely try that.  I wouldn't mind the print-time savings for sure.

 

19 minutes ago, SandervG said:

 

Do you have any Ultimaker PLA lying around by chance that you could try?

 

Only the tough PLA material.  I have the black that came with it and bought an extra roll of black and red.  I will buy a role of non-tough to have on hand and for testing.

 

19 minutes ago, SandervG said:

 

Good luck!

 

 

You and the rest of this community are the best!  Thank you so much for continuing to consider ways to help me.

 

[kmanstudios 1,120]

Sometime in the middle of the night, the print failed (machine error, not print error) and will try to print again. But this time, I will try basic setting that have always worked for me if that is ok.

[chuckmcgee 7]

Sounds like a plan.  If it works for you, maybe you can send me the project file and try here.  I just ordered a few rolls of Ultimaker PLA

[chuckmcgee 7]

1 hour ago, SandervG said:

Some of the other superficial artifacts you see might be caused by cooling. Personally I don't have much experience with Pro PLA or esun material, but it looks like as if the infill shows through the outer walls. It could be that the walls shrink some due to cooling, and that is why infill starts showing. It could explain why a higher infill shows it less, the space between the infill is smaller, which gives it less room to shrink.

 

@SandervG I meant to ask...for this point, would you recommend changing the fan/cooling settings in Cura or do I just increase infill for more surface integrity? 

 

In the case of this model, I believe I didn't design it well.  I don't need 5mm thick walls, which invites the infill between the walls at a lower level.  Besides improving future designs, I'm just trying to better understand my path forward on this model - the shrinkage and contraction on the current model happened with and without the chamfer.

[kmanstudios 1,120]

Here are two version of the files on settings that work for me on my crazy models. And, I routinely use the Matterhackers Pro-PLa due to expense. One at 0.15mm and 0.2mm layer heights. I agree that due to the design, the 0.2 should suffice.

 

E-sun is something I am not familiar with because I have seen examples where the diameter is not consistent enough and has caused underextrusions or worse.

 

UMS5_988243375_UMS5_controlbox-V2K.curaproject.3mf

UMS5_988243375_UMS5_controlbox-V2K-02mm.curaproject.3mf

 

I am going to squeeze in the 0.2mm layer height a bit later for a while just to get it past the point of your issues. It is also about half the printing time.

 

Having looked at the model, I do not see it as a leading cause. You may want to redesign aspects due to your preferences but it should not be a contributing factor to print issues.

 

Edit: Corrected size info from 0.1mm to 0.15mm layer.....fat fingered the typing...sorry ?

 

Edited August 1, 2018 by kmanstudios

[chuckmcgee 7]

Thanks @kmanstudios.  I will try those out on Thursday when my current print is done.

[SandervG 1,521]

15 minutes ago, chuckmcgee said:

@SandervG I meant to ask...for this point, would you recommend changing the fan/cooling settings in Cura or do I just increase infill for more surface integrity? 

 

Hard to say because I'm not that familiar with those materials. It feels like you're mostly haunted by material properties and therefore I would say that fan/cooling/heating would be the fastest settings to experiment with. But usually there are multiple roads leading to Rome ?

[chuckmcgee 7]

Got it - thanks!  A few things to try.  I can confirm that the current print (which is exhibiting the same issues, less profoundly @ 20% infill) has a bed temp of 60 degrees thanks to @kmanstudios suggestion to change filament to Ultimaker Yellow PLA (regardless that it isn't UM brand).  

 

These are the current basics for that material in the current project:

[kmanstudios 1,120]

9 minutes ago, SandervG said:

It feels like you're mostly haunted by material properties and therefore I would say that fan/cooling/heating would be the fastest settings to experiment with.

I am familiar with the materials in question and honestly, I have never had a problem...kooky thing is that I actually use the UM PLA settings and only change things like brims options, support options and almost exclusively use slicing tolerance set to exclusive.

 

We shall see because I am going to be setting the print on soon just to see. I will be using the 0.2mm file I posted.

[chuckmcgee 7]

yeah...the 0.2 wall may have been an easy fix here - besides a higher infill.  We shall see!

[kmanstudios 1,120]

Infill is just the default here.....15%

[kmanstudios 1,120]

OK....pulled the print off the machine and here is what I got:

 

I used old PLA and that was at the end of the spool which meant it was not as pliable as it was when new or UM or Colorfabb materials. It started out ok, but as it got further into the end of the spooling, it was really fighting the curvature of the filament baked in by the tight winding and age. So a bit of underextrusion. Do not let that get in the way of what we were testing......had this been a real job, I would have used new materials as they do not give me this issue.

 

In this pic, you can see the layers where it started out ok, but then went into underextrusion. OK layers on the left, underextrusion on the right. I did have to change materials though after the first few mm. That is why the dark grey to light grey.

 

Regardless of the underextrusion, I did get straight walls and no dips. This pic has a line parallel to the side of the print to make more clear. You can see it above. but I thought this may help. You can really see the underextrusion on this one.

 

New, or even better materials will make a difference from this result. Keep in mind that  I used the MH PLA Pro when I am doing things that are going to be heavily post processed. Colorfabb or UM materials for the raw print. For the really down and dirty prototyping for myself to work out stuff, I have gone as low as the MH build materials just because I do not want to waste good material on what is essentially a throwaway.

[kmanstudios 1,120]

I thought I would add something to make the issue with underextrusion clear as it can happen with any filament at any time.

 

When you get to the end of a roll, the winding is much tighter. If the filament is not pliable enough, this tighter winding will produce friction inside the bowden tube as well as the nozzle and make for difficulties of getting a smooth flow.

 

Even if the winding is not too tight and the filament is old and lost its pliability, this will happen as well.

 

Now, I used old filament since I was just doing a test. Not to knock the importance for you, but using my newer/better filaments for actual final prints unless I am going to super post process it with heating or sanding or filling or whatever.

 

And, since we were testing the actual layer 'slidng' (for lack of a better term), I was not looking for final print ready results. I did hope that it would not be so underextruded, but I wanted to get through the print.

 

Hope that is ok and I also hope the added info helps you and others ?

[chuckmcgee 7]

Wow @kmanstudios.  Thank you for going to that level.  I think the next logical thing for me to try is to load your project and print it on my print.

 

And I would do that - if I didn't run into another issue this morning.  One of the things I noticed about the print that was running until early this morning is that I never configured the X/Y offset after changing from the BB core to the AA core in extruder 2.  As a result, I noticed (and no one else will) an ever so slight shift when the color change on my model changed from one color on Extruder 1 (red) to the rest of the height of the box using Extruder 2 (white).

 

As soon as my print finished this morning, I knew I would be printing my final piece, which I planned to use PVA as support material for.  In loading up the new materials (and core), I decided to now do an X/Y offset calibration...except it won't work.  I have two videos loading that I will share.  Every attempt comes with one of two errors (and this is after a couple power cycles of the printer and reloading of material and cores):

1. Nozzle offset probe failed.  Please check the nozzles and bed and try again
2. Difference between detected height of both print cores exceeds realistic values

These errors come either after the BB core is extended and the bed begins to rise to the nozzle...only to stop way short and then throw this error...or if will successfully complete BB (after retrying from getting error #1) and then stop when the bed is rising to the AA core.  It stops way short and throws that second error.

 

Worth noting, since I have had the printer:

• I get error #1 at the start of every print, regardless of which cores are loaded.  I have to acknowledge the error and choose to "retry" - which it does successfully
• Often times, when I choose to change the core in slot two of the print head, the changed core is not recognized.  The wizard tells me to open the door...(which I successfully confirm), then to remove the core (which it recognizes), then to insert the new core - which is where it hangs...it never detects the replacement core.  I can hit X on the wizard, but that prompts me to cancel...which I eventually have to do after putting the new core in.  With the new core in and back to the main screen, the status shows no core in slot 2.  A reboot of the printer shows both cores in place.

I bought the printer from Matterhackers and have been sharing these findings with their tech support...I think they are going to turn me over to Ultimaker support at this point. 

 

Here are the videos:

 

 

 

Edited August 2, 2018 by chuckmcgee

[chuckmcgee 7]

Oh...and I am among the others that lose the video feed of the printer midway through the print.  Resetting Cura Connect and/or restarting the printer are my only options.  There is another thread in this forum with this topic.  Their experience is the same I am having with the S5.  I just figured this was a firmware issue and it would eventually get sorted out.

 

With these issues - I may get "uninvited" from this forum ?

 

I also saved the logs and have those to share if someone could get something useful from them.  Just let me know.

Edited August 2, 2018 by chuckmcgee

[kmanstudios 1,120]

47 minutes ago, chuckmcgee said:

With these issues - I may get "uninvited" from this forum ?

Never!! It is important to share negative issues as this is how things get fixed. I see it the way I see myself being corrected....an opportunity to 'fix' my internal database. Proper information is proper information..... :)

 

52 minutes ago, chuckmcgee said:

Difference between detected height of both print cores exceeds realistic values 

Make sure all schmutz is removed from the ends of the nozzles. And make sure there is no material at the probe location. Feel it. I have found very, very slight bumps of material not visible to the eye that way. The probing is sensitive. Also, make sure the nozzles are fully reseated. Also make sure your pads are not creating an issue by blocking it somehow. That pad is very sensitive to how it is placed. It can even pop the bracket off if not in place properly.

 

54 minutes ago, chuckmcgee said:

Often times, when I choose to change the core in slot two of the print head, the changed core is not recognized.

Just as a whim, try to wipe the contacts clean with iosopropyl alcohol. There hay be something there that is impeding contact.

 

55 minutes ago, chuckmcgee said:

These errors come either after the BB core is extended and the bed begins to rise to the nozzle...only to stop way short and then throw this error

Make sure it is cleaned completely of junk under the buildplate. It does not take much.

 

56 minutes ago, chuckmcgee said:

One of the things I noticed about the print that was running until early this morning is that I never configured the X/Y offset after changing from the BB core to the AA core in extruder 2. 

Always important. Especially if using PVA as that will put that slight shift if material in use and the OVA into each other and rough the hell out of your surface.

 

59 minutes ago, chuckmcgee said:

I bought the printer from Matterhackers and have been sharing these findings with their tech support...I think they are going to turn me over to Ultimaker support at this point. 

Having purchased form them (great guys BTW, you are under FBRC8's support purview. I am tagging @fbrc8-erin to make sure she sees this. She is awesome in knowledge and explainability.

 

1 hour ago, chuckmcgee said:

Thank you for going to that level.

No worries :) Others have done the same for me and it is how we learn. And, tests are worth a million ideas of theory :)

[chuckmcgee 7]

All great responses...thanks kman.  Only thing I will confirm is that the bed was super clean on leveling.  I did see the artifacts on a couple previous attempts, but today it was clean.  If you see the YouTube videos...you will see the bed doesn't even get close to the nozzle before failing with the error.  It looks like it is about an inch away before giving up.

 

Will try the alcohol though...both to consume (the consumable kind) and to clean the leads on the other ?

 

Oh...and I have interacted with @fbrc8-erin quite a bit.  She has been amazingly helpful with my UM3E.

[kmanstudios 1,120]

@fbrc8-erin is awesome indeed. A great asset to the community :)

[SandervG 1,521]

2 hours ago, chuckmcgee said:

• Nozzle offset probe failed.  Please check the nozzles and bed and try again
• Difference between detected height of both print cores exceeds realistic values

 

That is unfortunate to hear, we're sorry. Have you tried inserting that print core in slot 1? Do you get the same message, and we can suspect the print core of malfunctioning, or does it work fine in slot 1 and your slot 2 might not reading correctly?

[chuckmcgee 7]

15 minutes ago, SandervG said:

 

That is unfortunate to hear, we're sorry. Have you tried inserting that print core in slot 1? Do you get the same message, and we can suspect the print core of malfunctioning, or does it work fine in slot 1 and your slot 2 might not reading correctly?

 

Hi @SandervG  I get the same issues with both the AA and BB cores in that slot.  Should I try to swap the AA from 1 to 2 and see what happens?

[fbrc8-erin 272]

Thanks @chuckmcgee & @kmanstudios for your sweet words!

 

Chuck,

 

Since I know you've got a UM3X and the S5, by any chance do the Cores you currently have installed on the S5 have the red rings around them? The red rings won't fit right with the S5 nozzle cover, so if you've swapped the Cores out for your UM3X Cores, the ring could cause an issue. 

 

Is your lifting switch lifting correctly at the start of the print? If not, you may need to run the lifting switch calibration (basically the same process as on your UM3X). If the wrong nozzle is down when you start the active leveling, it will error out. 

Edited August 2, 2018 by fbrc8-erin