Failure Analysis - Dualextrusion PLA/Scaffold 24h,

Show off your prints, mods, etc.
HMBR
Posts: 14
Joined: Wed Sep 06, 2017 6:10 am

Failure Analysis - Dualextrusion PLA/Scaffold 24h,

Postby HMBR » Mon Aug 06, 2018 7:22 am

Hi Guys,

I tried to print a large dual extrusion (DE) model over night but sadly it failed again.
Materials used were:
- PLA, red - DasFilament
- Scaffold - Spool.Works (E3D)

I don't really know, why it failed though, because i used settings, that usually work fine when not printing a DE model. Maybe you see something in the pictures, that I don't.

What you see in the pictures:

0 - Nozzels after failed print. Looks dirty, but not like anything is loose and material got squeezed out at places its not suppoed to
1 - How I found it, when I shut it of. It was like 7hrs (out of 24h) to go.
2 - Broken top taken off. Platform adhesion was not the problem. Some slight problems with the support material and a pretty ugly 1-line wipe wall ;-)
3 - Top view of the upper broken off part.
4 - Bottom view of the upper broken off part
Attachments
0.jpg
1.jpg
2.jpg
3.jpg
4.jpg
Last edited by HMBR on Mon Aug 06, 2018 7:36 am, edited 2 times in total.

HMBR
Posts: 14
Joined: Wed Sep 06, 2017 6:10 am

Re: Failure Analysis - Dualextrusion PLA/Scaffold 24h,

Postby HMBR » Mon Aug 06, 2018 7:27 am

5 - Top view of bottom part (taken off from platform and removed wipe wall)
6 - Bottom view of bottom part

And a screenshots of my settings. If you want, I can send you my settings file and the stl file.

Thanks for your interpretation and help, guys!
Attachments
5.jpg
6.jpg
Settings-Support.png
Settings-Ooze.png
Settings-Layer.png

Jetguy
Posts: 2649
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Failure Analysis - Dualextrusion PLA/Scaffold 24h,

Postby Jetguy » Mon Aug 06, 2018 2:27 pm

Simple answer is you had some form of a nozzle strike against a curled up bit of the layer, and that caused an XY axis skip, and then because the layer no longer lined up to the previous layer you have even less adhesion, more warping and filament going astray, more catches on the nozzle and you start to birds nest.

Let's take some lessons learned about this print before even the failure happened.
#1 Support material is not a magic fix. Surface finish on even a supported surface "looks like" it was supported compared to a wall that was not supported. It's a crutch at best to attempt to make the "unprintable" print, if doing so with acceptable issues.
#2 You had overhangs that were not supported that look bad before the failure. So for all the effort, all the settings, this was still going to be a less than perfect print before the failure ever happened.

Again, let's take this picture of the base.
6.jpg

This is upside down so we are looking at the face that touched the bed. There are slopes and overhangs that were failing and had no support, but please listen, adding support is not likely to fix them either and adds that much more print time and material to the job further increasing the risk of a major failure.


In short, dual extrusion and multi material support is not the magic answer. Between the cost, complexity, and fact even on it's best day, a supported structure still leaved evidence of support compared to a smooth non-supported wall structure, again, is it worth it? Not to mention, the hours that any support structure adds to the total print time- just that much more time for a failure to happen 1/2 way through, not only wasting all that filament and time- but now you go for attempt 2.

By far, I know you want to print single piece but if that causes you to need support material, the far better option is cut the STL into pieces, and print and orient the pieces such that overhangs are eliminated or reduced, and no support material of any kind is required. Yes, you have seams of gluing the large parts together, but this is far more cost effective than multiple attempts to print, the cost of support material, your time in dealing with failures and multiple attempts, and in the end gives you a better surface finish and often a better part.

Here is an excellent video showing a model that if you printed with support and depending on how you oriented it would just never turn out well. However, if you cut the model, print with no support, orient individual cuts to limit overhang angle and put parts in such a way the layers are better oriented, the "unprintable" becomes printable. https://www.youtube.com/watch?v=YEzr-Yne4pA
Example model cut vs support.jpg

Jetguy
Posts: 2649
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Failure Analysis - Dualextrusion PLA/Scaffold 24h,

Postby Jetguy » Mon Aug 06, 2018 2:36 pm

Again, I can hear users groaning, "You mean I bought this dual extruder printer and it's worthless????".
No, I'm not saying dual extrusion doesn't have it's place, but right here is a classic example of a print where a user designs a organic rounded flowing shape, and then attempts to use support, and it doesn't come out as you want and results in failure and frustration.

The 3D printer is a tool, and as such it has limitations and best uses cases and worst use cases.
Understanding that, and using the tool in the best way gives you the best results.

Again, support material is a crutch. There is just no argument otherwise. Support material can assist with making a completely impossible to print object partially printable- maybe, on it's best day and with luck. Depending on how unprintable your object was to start with is a factor, and how much support is required VS desired comes into the mix. If your print "requires" support material, and that print is over say 2-3 hours long without support, then is the time to examine does support material make any sense at all, VS cutting the STL model into pieces and orienting the pieces to not need support. The larger the model- again, this an equation of time and material so a big print that requires support halfway up the model and is at risk for failure, you get a very poor ratio of cost, time, and risk of failure VS success.

I know full well you expected and wanted some magical profile or setting change that just fixed your ills, but that's just not how this works.

HMBR
Posts: 14
Joined: Wed Sep 06, 2017 6:10 am

Re: Failure Analysis - Dualextrusion PLA/Scaffold 24h,

Postby HMBR » Tue Aug 07, 2018 10:53 am

Jetguy, thanks for your long reply and all the explanation.

I absolutly agree, that cutting a model and printing it in multiple parts, is in many cases the more time and cost effective method but it also, at least in my opinion, takes away the -let's call it- "magic" of additive manufacturing. Also, breaking it down into multiple parts will probably still leave you with some supported and thus critical areas (as shown in the posted video).

Personally - ignoring time and money - I prefer using supports and dual extrusion from the start rather than glueing the parts together afterwards. This part now is a design object that could have been cut and oriented to ensure a optimal printing process but i.e when I build parts that carry mechanical loads, I can not always orient the part the way the printer would work best, but have to orient it regarding the later applied flux of force. Keyword here is layer delamination etc.

So what I'm looking for is to understand, why did the layer at that one point curl up so it collided with one of the nozzels, provoking an axis skip and how can I prevent that in the future from happening. Another guess I had is, that there was underextrusion in thoses lower layers (around layer 140-150), that recoverd again, but then collaped at a later point. Again, why does that happen? That usually doesnt happen with single extrusion (supported/non supported) right?

So again, I agree with your statement, that the printer is a tool with its ups and downs, like any other production machine and should be used accordingly. But rather than acceping the fact, that I can't print gemometically complex models in one piece - whats actually additive manufacturings USP against all other production processes is - I would like to get a deeper knowledge of the machine and thus get better, more reliable and reproducible prints.

So, looking at my temerature settings, is the pinting temperature (pla/scaffold) to high? The cooling to low? Are there any settings that allow for temperature controls on outer shells with high overhang, to reduce temperature jams in these areas? Any other solutions to this problem?

Jetguy
Posts: 2649
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Failure Analysis - Dualextrusion PLA/Scaffold 24h,

Postby Jetguy » Tue Aug 07, 2018 11:52 am

#1 I think this is an N2 or N2 plus based on your pictures, but have no idea what other modifications you have applied other than the fan ducts and coming system. There are a number of other areas of known improvement on this printer- specifically the filament feeder section. You said you suspect under extruding- well right there- the stock as supplied dual extruder giant heavy metal block assembly with less than the ideal drive gears on the motors and fixed pinch idler distance, let alone the additional factor that motor heat could come into play.
#2 Unsure if your air duct mods were effective or improved compared to stock and if they contributed.
#3 I don't see a shot of your cooling tab so I don't see where you posted the temperatures used?

I'm at a point where I'm past stock and never ever going back.
My minimum standard is:
#1 Replace the drive gears on the feeders with the known and very specific sharp tooth gears. Then carefully adjust the stock feeder motor to v-groove bearing distance fo optimal filament grip. viewtopic.php?f=2&t=1805&p=34303&hilit=sharp+tooth+gear#p34303 and viewtopic.php?f=4&t=6846&p=27374&hilit=shim#p27374
#2 Potentially lower the stepper driver current setting to reduce the feeder motor heating problem. viewtopic.php?f=4&t=1887&p=15699&hilit=motor+heat#p15517
#3 Insulate your hot ends.

My recommended standard is: (My posts in this 8 page topic details my current setup) viewtopic.php?f=4&t=4841&hilit=berd&start=100#p28586
Bondtech feeder upgrade
E3D V6 adapter
E3D titanium thermal barrier tube
E3D copper heater block
E3D 30w premium heater cartridge (24V)
E3D thermocouple probe
E3D copper nozzle
E3D silicone sock
Properly installed Berd air cooler system
Capricorn PTFE tubing for both internal feed path inside the extruder assembly to the E3D adapted hot end and the feed guide tubes from the spool.

I'm also done using dual extruder and trying to get and maintain 2 nozzles in the N series perfectly the same level and then also adjust the XY offsets. Single extruder machine only and either use same material support (PLA with PLA support) or use the Mosaic Palette+ if you want to go down that road of needing dual or multiple filaments.

Jetguy
Posts: 2649
Joined: Tue Mar 22, 2016 1:40 am
Location: In a van, down by the river

Re: Failure Analysis - Dualextrusion PLA/Scaffold 24h,

Postby Jetguy » Tue Aug 07, 2018 1:08 pm

I think some of what you are seeing is your hotends are not insulated and your modified cooling setup is not idea under dual extrusion when both hotends are hot and radiating heat along with a bunch of hot air being blown around by the fans.
Which is why I'm saying what I'm saying- you didn't "need" dual extrusion water soluble support material for this print. If you needed support, fine, enable same color material support and print completely in single nozzle.

2.jpg

Going back to this picture, the wipe wall was not printing well (I've never been in love with Raise3D implementation of wipe wall in Ideamaker) and It's possible contamination into the layer contributed with some level of fail at the layer. Next, I see your fairly dense support material, but one of the overhang areas in the rear left corner appears the support material never even touched it where it was sharply overhanging. Now maybe that missing because of the layer fail. Again, for me, I'm looking at the result and for the support material used and the time to that point, this is not getting even a remotely decent result.
Reviewing your settings:
Settings-Support.jpg
Settings-support material.jpg
Attachments
Poor support performance.jpg


Return to “User's Gallery”

Who is online

Users browsing this forum: No registered users and 0 guests