Unexplained patterning in our prints

[TheStargazer 1]

All,

So a few months ago we had to replace our temperature sensor on the Ultimaker 2, and took that opportunity to install an Olsson block.

Did a few quick test prints in Colorfabb XT to ensure everything was hooked up right, then left the printer alone for a few months, being too busy to print.

Recently we've returned to perform a few prints but are confounded by a few odd issues we were not having before.

When printing I am getting some odd patterning in the walls of the prints, especially near corners.

Best example I have is printing Coffin's Cube, one of the default prints on the SD card from Ultimaker.

As you can see ~ 4 months ago there was only cosmetic pillowing (while visible to the naked eye, completely smooth to the touch), but in the print I made yesterday it has severe pillowing and an odd pattern in the walls of the print that I've never encountered before.

Both blocks were printed using the default G-Code, Temp = 240, Printbed = 70 C, Fans off for first few layers then on full.

Same exact settings, same filament. Only difference was they were printed about 4-5 months apart.

I've tried newer rolls of the same filament, just in case humidity / water contamination issues were at play (though if they were I would expect to see random problems, not patterned problems) but no change.

Playing around with the settings, if I either increase the top layer height to something ridiculous like 1.2 mm, or increase the temperature to 260 the pillowing and patterns begin to go away....

Any ideas? What could have changed in my printer in the last few months to cause this issue? Some problem with the temperature sensor that we replaced? The olsson block? Something else?

Edited March 15, 2016 by Guest

[SandervG 1,521]

What nozzle have you installed? Are you sure it is the same as the one listed in Cura?

It could be 'plain' under extrusion.

Since you just installed the Olsson Block I don't think your nozzle could be clogged. They should be just new. So an Atomic Method probably won't lead to anything.

Have you ever replaced the PTFE coupler?

[TheStargazer 1]

Hey Sander,

Thanks for your response.

We are using the standard 0.4 mm nozzle and the settings in Cura match. I've tested for nozzle clog even though it didn't seem to match the symptoms (nozzle clog would usually be a random, rather than patterned problem) and filament flows quite easily though the nozzle. No junk came out when I did the Atomic method.

We replaced the coupler at the same time as we installed the Olsson block since we had the print head disassembled anyway.

After work today I plan to double check the printer has the latest driver versions and check that all of the motor mounts are tight.

[SandervG 1,521]

You could also check if the drive bolt on your extruder motor is nice and clean.

What are the teethmarks on your filament like?

A picture could help if you want to add it.

Edited February 8, 2016 by Guest

[TheStargazer 1]

Teeth marks on the filament are nice and light. Not too deep. No grinding visible anywhere on the filament and no particulates of filament in the motor.

I have tried 1 "tight" print where I increased the tightness of the motor by a turn, and one where I loosened it by a turn, and did not see any noticeable difference in the print.

Tried taking a picture but my camera is horrible at close ups and the teeth marks are not visible in the photo.

Edited February 8, 2016 by Guest

[neotko 1,417]

Maybe, said nce you changed to olsson, the metal cap it's touching the block. Check that. Because if it's touching the hotend heat changes and when the fans go on (since you probably are using the 25W heater that comes with the machine) then the hotend might have troubles getting the temp, and then the plastic flow changes, etc.

If it's touching, try to run the fan at 100% on a more paced scale so the heater can keep the temp layer after layer. Also since you use olsson, it might be a good idea to use a new fan cap since the block it's bigger than the old standard one.

[TheStargazer 1]

neotko,

Thanks for the feedback!

When first installed the Olsson block was touching the plate and causing the temperature to drop significantly to the point that we were getting heater errors half way through the print.

We took it apart and cleaned the threads, making sure to screw the block in further the second time and now there is an approximately 1 mm gap between the block and the metal cap.

I've monitored the temperature during a print and the heater does not seem to have any issue currently keeping up with the cooling since we screwed the metal coupling in further. At all times the nozzle is within 1-2 degrees of the set point.

[Remy 13]

Couple more things I can think to check:

is the firmware on the printer up to date (15.04.4)?

Is the printer in the same room/location as it was before? Any new drafts that could be causing temp inconsistency e.g. AC or heating unit kicking on/off in the room, open window etc?

Have you checked the Bowden tube to make sure it's not crimped/damaged anywhere? This can generate friction and cause under extrusion. Is the Bowden tube fully inserted into the print head so that it's touching the PTFE coupler?

Is the feeder motor kicking back (making a 'tok-tok' sound) while printing?

Have you increased the nozzle temp while printing to try and eliminate the issue?

Have you tried swapping to a new nozzle? XT is pretty abrasive stuff so if any was left over in the nozzle, it's possible that it's carbonized and is just hardened in there now. Atomic pulls can look clean but could be that's because the material is no longer getting viscous in there and isn't moving, just blocking up some of the melt chamber. The JET nozzles are designed slightly differently and should be specially treated inside, but I don't know if this makes leftover material harder/easier to clean out. I've seen a few cases with the stock nozzles where leftover material has carbonized so much that Atomic pulls come out fine but under extrusion continues to happen even after several pulls and checking everything else.

[TheStargazer 1]

Couple more things I can think to check:

is the firmware on the printer up to date (15.04.4)?

Is the printer in the same room/location as it was before? Any new drafts that could be causing temp inconsistency e.g. AC or heating unit kicking on/off in the room, open window etc?

Have you checked the Bowden tube to make sure it's not crimped/damaged anywhere? This can generate friction and cause under extrusion. Is the Bowden tube fully inserted into the print head so that it's touching the PTFE coupler?

Is the feeder motor kicking back (making a 'tok-tok' sound) while printing?

Have you increased the nozzle temp while printing to try and eliminate the issue?

Have you tried swapping to a new nozzle? XT is pretty abrasive stuff so if any was left over in the nozzle, it's possible that it's carbonized and is just hardened in there now. Atomic pulls can look clean but could be that's because the material is no longer getting viscous in there and isn't moving, just blocking up some of the melt chamber. The JET nozzles are designed slightly differently and should be specially treated inside, but I don't know if this makes leftover material harder/easier to clean out. I've seen a few cases with the stock nozzles where leftover material has carbonized so much that Atomic pulls come out fine but under extrusion continues to happen even after several pulls and checking everything else.

 

Remy,

Thanks for the reply!

Printer is in the exact same location as it was before, no changes to the room / ventilation.

No feedback sounds from the motor.

Bowden tube is fully inserted and clean of any damage.

Also as mentioned in my original post, the nozzle is brand new.

As mentioned in my original post, increasing the nozzle temperature to 260 will make the problem start to go away, but then I start having a lot of other issues from using a temperature higher than normal for my filament.

Again, I am more concerned with what is causing the problem / why something suddenly changed. Yes there are measures I can take to force the problem to go away, but I'm more concerned with what is causing it, rather than hacking a temporary solution.

[SandervG 1,521]

Can you check the temperature on your hot end?

Is your PT100 still reading the right temperature, or is 260ºC maybe really 230ºC?

Do you have access to a tool that reads temperatures?

[TheStargazer 1]

Can you check the temperature on your hot end?

Is your PT100 still reading the right temperature, or is 260ºC maybe really 230ºC?

Do you have access to a tool that reads temperatures?

 

Yup! Fortunately I have access to some cool tools here at work...

Used an IR camera to get the exact temperatures of the nozzle at both 240 and 260 C set points.

When the printer thinks it is at 240 it is actually at 241.3, when it thinks it is at 260 it is at 269.1.

(The 260 setting actually going to 269 is actually a little concerning if I want to print in ABS but I fortunately don't do that often, for this issue it's irrelevant)

EDIT:

Hmmm Ultimaker didn't like the image files.... Here's the link to the imgur album:

http://imgur.com/a/yqKVR

Edited February 9, 2016 by Guest

[SandervG 1,521]

240/241 sounds right. Depending on the PID values it could always shift around a few degrees.

I'm thinking we are maybe overthinking it.

What material are you using now?

[TheStargazer 1]

240/241 sounds right. Depending on the PID values it could always shift around a few degrees.

I'm thinking we are maybe overthinking it.

What material are you using now?

 

Colorfabb XT is the main filament we use here. I've tested several rolls, both old rolls and new rolls to see if it is a humidity / filament issue and it had no effect.

To me it seems like it is a mechanical issue as it always repeats in the same places of the prints. When it is making certain corners, etc... It is generating these patterns, and also lots of pillowing. But 4-5 months ago it did not...

If it were a filament / clog / feed issue I would expect the problems to occur randomly throughout the print rather than be reproducable.

Today I checked the motors to make sure they were tight against their rubber bands. They were still acceptable, but I tightened them anyway and it had no effect.

Reflashed the firmware for the upteenth time from Cura 15.04.4 but no effect.

[SandervG 1,521]

I have seen a recent case where printing with XT could also wear down the drive bolt. So if you are considering trying out different parts, I would definitely give this a chance too.

@Nicolinux, I have the feeling you have some expertise on this subject as well?

Underextrusion can manifest with horizontal smaller layers, but also vertically.

I noticed, myself included, a few people suggested to check your PTFE coupler.

Maybe I miss it, but I don't see a response on this. Could you confirm?

It can be, due to the spring and pressure it has received over the last 5 months it has deformed a little bit and can cause under extrusion.

Looking forward hearing from you!

[Remy 13]

Looks like the PTFE was replaced when the Olsson block got installed. Could you post a photo of your print head? It's possible the spacing of the parts is off by a bit which can also cause under extrusion.

Have you re-leveled the build plate recently? Starting with the optimal spacing between the nozzle and the build plate can vastly improve quality. If the printer was moved or if the glass build plate was taken out/moved around, it's a good idea to re-level using the wizard on the printer.

Do you have any additional adhesive on the build plate or are you printing onto the clean glass?

Is this issue present in all your models, or just the one you posted a photo of? Have you continued printing and observed the same behavior?

Another thing you can check easily: are you printing with the material still on the spool, or do you cut material off the spool and let it hang freely on the back of the printer? I usually cut material off the spool as it reduces friction and lets the filament feed in much more smoothly. If there's a tangle in the filament, or if it's just wound too tightly/not tightly enough around the spool, this can also cause systemic under extrusion. Maybe give it a shot printing off the spool and see if this helps.

Last but not least: have you tried increasing the bottom/top thickness in Cura? I usually use a value of 1mm to cover up the infill when printing with a lower infill percentage like 10-20%. This maybe doesn't explain the issue but can give you a quick fix to continue printing while troubleshooting.

[TheStargazer 1]

Unfortunately none of these feedback points are useful, as they're all basic 3D printing 101 troubleshooting tips.

1) PTFE is new

2) Print Head photo attached below as requested

3) Plate is releveled after every print.

4) We usually use a very thin layer of water soluble gluestick on the build plate when printing with ABS or ColorFabb-XT. Have not found it necessary for PLA.

5) All models. And it gets worse with bigger. models. Photo attached. The model is supposed to be a rocket and parts, with smooth cylinders.

6) We are not interested in changing the bottom / top thickness, as that would be hiding the problem, we are interested in identifying what has gone wrong with our printer and fixing it, not hiding the problems.

[TheStargazer 1]

An interesting point... We have tried a number of filament types (we have tried older rolls of filament and brand new rolls of filament, with the problem appearing with both) and have seen the problem with PLA and ColorFabb XT, but not with CF-20 for some reason?

As I said, we have monitored the temperature of the hot end with a thermal camera and it is correct.

Edited March 15, 2016 by Guest

[owen 19]

Have you tried lowering the fan speed to 50% or less? I think it's better for XT.

[SandervG 1,521]

Could you take a close up picture of your teethed sleeve on your extruder motor to check if those teeth are still good? XT and XT-CF are pretty hardcore materials and maybe they have had an affect on your teethed sleeve.

Are you sure your nozzle is still good?

Some materials are pretty abrasive and can wear down your nozzle, or leave residu behind.

Besides your PTFE being new, is it also positioned correctly? It doesn't look like an Ultimaker PTFE, so not sure if that gap is big enough, or too big? Is the ID good?

[TheStargazer 1]

Owen - Yes. We're printing at settings that worked beautifully with this printer for ColorFab and PLA a few months ago, but suddenly are causing weird problems. It seems like something is changed / broken.

SandervG - We have tried 3 new nozzles of different diameters. Still have the same problem.

The coupler is a new after market one that came with our Olsson block. It appears to be positioned correctly as far as I can determine.

Again, weirdly is prints fine in CF-20, but suddenly has this problem in PLA and XT....

[TheStargazer 1]

Anyone have any inspiration?

Still having this error.

[SandervG 1,521]

Just had another look at your photo of the printhead, and the part underneath your PTFE doesn't look like the metal part Ultimaker supplies, does it?

Is it from a third party or something?

What other parts have you replaced?

Your rods also have a black residu on them. Are they smooth? Have you lubricated them frequently?

Have you assembled your Ultimaker head according to these guidelines?

https://ultimaker.com/en/resources/253-re-assembly

Perhaps most tips are basic tips suggested, but there is not a lot of information available to think outside that box. Also, often when a user is slightly more experienced the basics are easily overlooked.

Do you have some more photo's of prints?

Do you see for example a change in movement where it goes wrong, or is it an extrusion issue for sure?

[TheStargazer 1]

Only parts that are not from Ultimaker are the PFTE coupler and the Olsson block.

The "black residu" is just tarnishing of the chrome. Bit puzzled by that myself, but they're smooth to the touch and lubricated. I've just been using the lubricant that came with the Ultimaker2. Any suggestions on something better?

Will print a few more test prints on Monday.

[Torgeir 280]

Hi Stargazer,

When I saw your second print, it cached my interest. We’d like to say this is an under extrusion, yes –or no?

If we notice there are some plastics missing in between the treads, yes! But if we study the “nonrandom” pattern –no!

Well, except from the brim that’s looks like there is an under extrusion. The model is not having what we name as a “normal” under extrusion, a complete mess!

Have a closer look at the picture of the model rocket; I’ll guess this is how we see it when looking straight into the printer (object placed in the middle)?

Ok. If we look at the left side of the two stages, they have very few sign of this special pattern that’s visible on the right side of the object. This pattern is repeated in a special sequence and it have to be very precisely in order to make this pattern.

I’m thinking of a mechanical failure that’s become visible due to software behavior/demand, when the printer is doing what it’s normally doing when printing our object.

Ok. What I’ll think is that the bourdon tube has a “little” slack, this can explain this strange behavior. When the extruder head move to the left side of the printer, the bourdon tube may tilt. This tilt or bend action may be just enough to take out the slack in the tube. This is why the print almost looks good on the left side.

So check that the bourdon tube sit rock steady at the extruder and the feeder, this is important.

I’m using two clips at both side, else it will slip if you pull into it.

Just my 5p.

Good luck.

Torgeir.

[TheStargazer 1]

Torgeir,

Thanks for your well thought out response.

I agree it is a very unusual problem, that has us at our wits end.

While we needed to shorten the feeder tube by approximately 1 cm due to some filament that became stuck in it, there seems to be plenty of slack. But to be honest, I've run out of ideas and a new feed tube only costs 20 euro so I'll purchase a new one and see if that helps at all...

Will let you know how it works out.