ABS - partial insufficient strength of printed objects

Share your successful printing settings with others here. If you have any question about the parameters or printing skills of some filament, please find here.
Tinkerer
Posts: 138
Joined: Mon Oct 09, 2017 8:30 pm

ABS - partial insufficient strength of printed objects

Postby Tinkerer » Sat Nov 11, 2017 7:18 pm

Today I printed a case for a raspberry Pi. Not something to complicated I thought.
The print went fine so far but when I worked the print of the print surface, I was surprised to notice
that the print did have not the needed strength - at least partially.
IMG_3881.jpg


As you can see in the picture it seems to be that certain parts of the print are really weak!
Which is strange since the rest of the print does give a quite sturdy impression.

Does anybody know possible reasons for that?

Wet ABS?
Problems with the extrusion? (Bondtech)
Temperature problems (like not keeping temp up?)
Or simply bad quality of the ABS material used?

Jetguy
Posts: 1841
Joined: Tue Mar 22, 2016 1:40 am

Re: ABS - partial insufficient strength of printed objects

Postby Jetguy » Sat Nov 11, 2017 9:20 pm

Of all materials, ABS presents a higher challenge, especially compared to PLA.
Why?
ABS was designed for working in an injection mold- basically the polar opposite of 3D printing.
Can you print with ABS? Yes, quite few folks use it daily and much of the 3D consumer/hobby printing started on ABS.

Why is ABS harder to print with?
The exact same reason I brought up injection molding, ABS has a high shrinkage ratio. The goal of it was to squirt into the mold, then shrink as it cooled so it can be pulled out of the mold.
Understand that you squirt down hot molten ABS plastic and in that state, it is expanded thermally.
As is cools towards room temp, there is a non-linear rate of shrinkage. Basically, there is a temperature that is still very hot that ABS stops being liquid and starts being solid. As you cool beyond that, you get an ever increasing shrinkage as you near room temp.
A 3D printer prints layer by layer. You squirt down a hot layer, it cools and solidifies, at the same time shrinking in linear length. You squirt of hot layer and it partially melts and tries to bond to the previous now hard and cooling layer. When this new layer cools and shrinks, this creates a massive shearing force on the layer to layer bond.

What do we do to combat this to make 3D printing layer by layer work better with ABS?
#1 we use higher heated bed temps (100-120C max) and this keeps the first layer against the bed warm and thus does not fully shrink that last percentage because it's held at an elevated temp. That said, the layers higher up as we build more are not as effectively keeping the heat retention of the bed. This is where the sealed up printer "chamber" is allowed to hold in the heat so the ambient temp and thus the air around the part is maintained higher than room temp. This is where we ensure the doors are closed and the lid is on the printer. Note- that some folks and I agree, the lid seals too well and you need to find a point where you gap or vent the lid to keep the temp hot- but not so hot as you cause more problems like stepper overheating. See this thread for some discussion and further links in replies viewtopic.php?f=5&t=163&p=19301&hilit=lid+vent#p19301 and also here directly on user John who does a ton of ABS printing viewtopic.php?f=4&t=1709&p=16520&hilit=lid+vent#p16520

#2 We block any cooling airflow directed at the printed layer to allow it to cool slower. This helps layer bonding while also reducing differential shrinking between the current layer and the previous layer. On the Raise 3D series there are printed covers that go over the fan ducts. This is because the fans are needed to cool the heatsinks on the throat tubes and must run. That said, the duct splits the air between cooling the heatsink and blowing at the now printed layer and the covers block the duct that blows at the layer.
Great thread on the topic viewtopic.php?f=2&t=1632&p=13743&hilit=duct+cover#p13743
The original link and info by user Scienteer is in a Google group and you may not be able to see it without joining that group https://groups.google.com/d/msg/3dprint ... AfyyHoAQAJ


#3 We use custom blends of ABS material that are not the same used for injection molding so that we get a lesser total rate and even more cured shrinkage rate. These are again specific low warp plastics made just for 3D printing VS bulk plastic borrowed from the 3D injection molding industry. Again, this thread covers and shows the exact differences material brands can make. Topic viewtopic.php?f=4&t=2572

#4 We understand these limitations to choose the right plastic, the right machine setup for the very print we are trying to print. In particular, objects with long straight sides are going to experience the worst effects of material layer by layer shrinkage just because shrinkage is linear. If you have a 10mm long wall and it shrinks 3% of 10mm, that's not huge. If you have a 200mm long straight wall and it shrinks 3%, 3% of 200mm is not an insignificant distance.


So again, basic 3D printing theory.
There are several critical temps for any plastic:
#1 is called the glass transition temp and it is when the plastic stops being rigid and is now soft and flexible, but also not liquid melted. You have to cool just below this temp for the part to be 100% solid, but cooling much lower and that rate of shrinkage (every degree cooler results in ever more shrinkage).
#2 Actual melting temp- the point the plastic is liquid enough to flow through the nozzle.
#3 Maximum temperature before the plastic degrades- this is the upper limit you can set extruder temp to before you begin burning and damaging the plastic inside the nozzle. ABS turns a distinct brown in this state and can begin to leave deposits inside the extruder and jam the nozzle.

So, your printing temperature settings:
Extrusion temp will be higher than melting point but LESS than maximum damage point of a given filament. The idea is you need to be higher than melting temp because when it leaves the nozzle and is melting into the previously colder solid layer, it must transfer some heat to the previous layer. That said, high temps mean it takes LONGER for the filament to become solid. This is where smaller objects will deform from this extra heat as the continuous layers and adding heat to a small area it never solidifies and deforms under gravity.

Heated bed temp controls ONLY the layers near the bed. However, since chamber heat is not a feature we have, you need to ensure the ambient temp inside the printer is above room temp but actually lower than heated bed temp (easy, because the heated bed generally will not be able to heat all the air in the volume of the printer to actual heated bed temp. Again, if you start reading material data sheets for the very plastic you are trying to print- glass temp is specified. Heated bed temp may be very near this and we want ambient slightly less.

Again, the why behind all this:
The idea is that by keeping ALL printed layers near the just solid state and an elevated temperature near glass transition point temp, the plastic layers do not yet experience full shrinkage. This way, once all layers are printed, the entire part and all layers cool down to room temp uniformly at the same time. This reduces the inter layer stress and warping and layer cracking.

Jetguy
Posts: 1841
Joined: Tue Mar 22, 2016 1:40 am

Re: ABS - partial insufficient strength of printed objects

Postby Jetguy » Sat Nov 11, 2017 9:37 pm

Duct cover pictures. Posted here viewtopic.php?f=2&t=1632&p=13743&hilit=duct+cover#p13743 but copied for ease of use.
1.jpg
1.jpg (17.98 KiB) Viewed 270 times

3.jpg
3.jpg (26.35 KiB) Viewed 270 times

2.jpg
2.jpg (25.51 KiB) Viewed 270 times

Jetguy
Posts: 1841
Joined: Tue Mar 22, 2016 1:40 am

Re: ABS - partial insufficient strength of printed objects

Postby Jetguy » Sat Nov 11, 2017 9:51 pm

Also note that vapor acetone treatment is a known way to strengthen ABS parts.
viewtopic.php?f=5&p=21043#p21042

This is slightly dangerous in that you are exposing the plastic to acetone vapor which is flammable. It's also going to be more effective on a better ABS part (one that does not have huge inter layer stress built up from improper printing conditions). Acetone is a solvent to ABS and widely available. The other issue is that too much vapor can damage the part making is soft and it then can deform or stays soft forever if the vapor gets into the infill area and cannot escape. https://www.youtube.com/watch?v=h2lm6FuaAWI
https://makezine.com/2014/09/24/smoothi ... one-vapor/
https://www.matterhackers.com/articles/ ... -polishing

Tinkerer
Posts: 138
Joined: Mon Oct 09, 2017 8:30 pm

Re: ABS - partial insufficient strength of printed objects

Postby Tinkerer » Sun Nov 12, 2017 10:44 am

Jetguy,
thanks a lot for the very exhaustive explanations and informations.

So what I read out of your text:

There is no particular wrong setting in my print but it just can happen when printing ABS?

What makes me wonder is, that there are layers below and above the weak layers,
which are quite strong and okay.
So I assume something went wrong just in one layer?

I may should have been posting the basic settings used:

- Fan covers installed (=cooling off)
- Printbed = 105°C
- Nozzle = 250°C

User avatar
Julia Truchsess
Posts: 209
Joined: Mon Mar 21, 2016 1:30 pm
Location: Connecticut
Contact:

Re: ABS - partial insufficient strength of printed objects

Postby Julia Truchsess » Sun Nov 12, 2017 2:40 pm

There are a few things you can do to strengthen parts:

• Use thicker layers. Thicker layers carry more heat energy from the extruder to the print, increasing layer bonding.
• Use more shells. Shells contribute more to part strength than infill.
• Increase the amount that infill overlaps the inner shell(s). I forget the exact name of the parameter, but it increases the "connection" between infill and shell.

As Jetguy said, while ABS has some nice characteristics such as ease of support removal and nice surface finish, it's not the strongest material out there in terms of layer bonding and overall "solidity".

Vice Chief
Posts: 43
Joined: Wed Aug 23, 2017 5:04 am
Contact:

Re: ABS - partial insufficient strength of printed objects

Postby Vice Chief » Sun Nov 12, 2017 9:20 pm

Hey tinkerer, I don't think this is normal at all. My machines print ABS very nicely.

Looking closely at your picture, I see a number of odd defects that don't look like they were part of the original model. I marked them in red in the attached image. Did you design the model?

To me, it looks like that layer of the print may have been a little cold, or (despite the covers) there was still fan-air cooling the print as you went along. It's also possible that your filament is bad. 250 might be a little hot for the extruder - I tend to extrude ABS around 245C.

[img]
tinkerer.jpg
[/img]
Attachments
tinkerer.jpg
Last edited by Vice Chief on Sun Nov 12, 2017 10:02 pm, edited 1 time in total.

User avatar
Mecha_Monster
Posts: 140
Joined: Wed May 11, 2016 2:07 am
Location: Mexico
Contact:

Re: ABS - partial insufficient strength of printed objects

Postby Mecha_Monster » Sun Nov 12, 2017 9:29 pm

I agree with Vice Chief.

When I print with ABS, the temperature is the most important variable for print strength/adhesion. When I print too cold (or the covers slip and some air escape) I experience delamination and brittle prints.
For natural ABS, usually I print at 235 degrees. but when I use colored ABS I go higher, for ex, white is 242 and black 245. The colorant added to the ABS modifies its melting point.
Also, the same happens when printing too hot. The print turns out brittle.

Jetguy
Posts: 1841
Joined: Tue Mar 22, 2016 1:40 am

Re: ABS - partial insufficient strength of printed objects

Postby Jetguy » Sun Nov 12, 2017 10:38 pm

While I fully agree that more shells can make some aspects and certain parts stronger, the vertical walls simply have a fixed number of shells that fit. While Ideamaker has improved and changed versions there are still potential versions and situations where the thin walls are NOT filled with infill. Again, what can happen is if the number of extra shells does not fit, the infill does not come back and fill in this space because the slicer thinks the shells claimed it earlier in the slicing process. This is supposed to be fixed with thin wall behavior settings and extrusion width settings but there can be issues and that's why previewing gcode on any print that has thin walls is a good idea.

Point being, the break appears to show near hollow walls with limited infill. Something you should check on preview of the gcode before ever printing.

Somewhat seen in this thread is how you can adjust and see the difference viewtopic.php?f=8&t=4321&p=20661&hilit=shells#p20661
standard_fine_2.JPG

VS
standard_fine.JPG


Also here viewtopic.php?f=6&t=906&p=20948&hilit=thin+walls#p20948
scr111.png

Tinkerer
Posts: 138
Joined: Mon Oct 09, 2017 8:30 pm

Re: ABS - partial insufficient strength of printed objects

Postby Tinkerer » Mon Nov 13, 2017 11:50 am

Vice Chief wrote:Hey tinkerer, I don't think this is normal at all. My machines print ABS very nicely.

Looking closely at your picture, I see a number of odd defects that don't look like they were part of the original model. I marked them in red in the attached image. Did you design the model?

To me, it looks like that layer of the print may have been a little cold, or (despite the covers) there was still fan-air cooling the print as you went along. It's also possible that your filament is bad. 250 might be a little hot for the extruder - I tend to extrude ABS around 245C.


Hi ViceChief,

the defects you marked are not as designed, right.
These are exactly points where I try to find out why these defects happen...
I also thought whether these defects can be related to variations in the temperature of the extruder head.
Simply do not have an idea yet how to control/check that during the print -
I can only rely on the display of the N2...

The part has been designed w. Fusion 360 and converted to STL from there. Looks fine in IdeaMaker/preview.

Regarding the missing infill issue Jetguy mentioned:
Yes, I already saw that there are unfilled areas in "thin" walls.
Will try Cura to see whether that works better.

Regarding temperature:
I started printing ABS at 225°C and did have only unusable parts until I came to 250°C.
While these parts still turn out to have the weaknesses described earlier.
255°C was the max I tried yet. Worked the same way as 250 but did smell a bit more -
so I guess 250°C is the max safe temp with this type of ABS (it's called "ABS Pro", brand "Renkforce" (which is some sort of rebranding)).

The fan covers are always installed and the printer is always closed when I print ABS.

Jetguy
Posts: 1841
Joined: Tue Mar 22, 2016 1:40 am

Re: ABS - partial insufficient strength of printed objects

Postby Jetguy » Mon Nov 13, 2017 1:38 pm

Those defects may be under extrusion (the amount of plastic commanded in the gcode is not the amount of plastic your extruder was able to push out during the print).

Since this specific layer this happened at is ALSO the first layer where a lot more retractions happen as the nozzle moves between one area of printing to another- that might be a contributing factor. When you retract filament, you then pull the hot melted tip end of the filament into potentially the part of the extruder that is not hot and that plastic tip can bind causing a slight delay between the time the extruder tries to extrude more plastic and the time the plastic begins coming out the nozzle.

Just the facts here:
Printing ABS requires the lid and doors closed, higher temps and thus higher ambient temps.
At the same time, this can absolutely overheat the air around the extruder feeder motors and feeder assembly and make it so hot to soften filament in the cold section of the extruder. This cannot be allowed to happen and can cause feed failure and failed prints.
You upgraded the extruder feeder to Bondtech- but are you 100% sure you also set the proper stepper driver current- BEYOND just following the instructions and ensured you are not overheating the motor and thus the extruder feeder assembly?
Did you also check, validate, verify, and even potentially just take the extra step to apply blue threadlocker to the setscrew for the Bondtech feeder gear?
Did you take your hand a measure the motor temp on the extruder during the print?
Did you observe the print and notice the extruder feeder ever skip steps? This is a skill and a sound you should be listening for during any print- but especially when you try something new or modify your printer. I'm not talking about an XY skip, the extruder axis too can skip steps with either an extreme jam or stepper driver not properly adjusted on the Bondtech.


Return to “Printing Settings”

Who is online

Users browsing this forum: No registered users and 1 guest