Switching to ABS - need any tips or tricks

This week i've starting to use ABS now that I have a E3D v5 extruder.

What suggestions do you guys have for changing slicing settings for ABS versus PLA? I use both Slic3r and Skeinforge. At the moment i'm using my PLA config files and switching out the temps for bed and extruder ( 115c and 230c respectively). And yes, using acetone slurry for the bed.

Do you use the fan duct for cooling printed ABS plastic? If so, what layer do you turn it on?

-----------------------------------------

For reference my current printer setup:

Sturdy90
3mm mirror glass for bed
EZstruder direct drive setup
1.75mm filament fron Ultimachine.com
E3D v5 1.75 extruder
0.4mm extruder tip
0.2mm layer height as default

Quote
lions3
This week i've starting to use ABS now that I have a E3D v5 extruder.

I have printed ABS with the standard J-Head using the same settings as for PLA apart from two things:
Temperatures : Bed needs to be around 130C and extrusion in the order of 245C. (Bed is as high as my 12V PSU will take it)
Fan: Leave it off for all layers

Alan

Yes 115 seems a bit low for printing on ABS slurry but if its working for you let us know!

I would add the following:

- May need to recalibrate your E steps per mm (as the martial has a different hardness the hobbed bolt will dig in to the filament to a different extent which results in a larger or smaller effective diameter of the hobed bolt and a different steps per mm)
- Find a way to keep drafts off the printer (especially if you wont to pint larger items). The problem is you kind of need to have better ventilation of your room to stop your self being (slightly) poisoned by the ABS and no drafts on the printer to stop warp... Some kind of chamber with a small extractor fan to the outside of your house is the ideal. But even a big old cardboard box you can put right over the printer would be a good start as then you can have the window open.

thanks for the feedback!

Quote

Bed needs to be around 130C

130c for the bed? Wow. A co-worker had a solidoodle and I remember he was running at 110. So I just assuming I would probably need to run something similar.

Quote

May need to recalibrate your E steps per mm

Thanks for reminding me. I completely forgot to update it the firmware for E steps.

Quote

The problem is you kind of need to have better ventilation of your room to stop your self being (slightly) poisoned by the ABS

As to the fumes. How long do they linger? Do they settle if I put the printer in a box?

110 - 120 first layer works well for me when printing on PET tape and glue. I think you will need to be closer to 130 for the 1st layer with ABS slurry (could be difficult to get to 130 with a 12V power supply but as close as you can get may be ok). You should be able to drop lower than that for the 2nd layer. Try it on some small prints 1st and see what you can get away with. You wont to use the lowest bed temp you can but without your parts coming unstuck.

> How long do they linger?

A few hours in a closed room. Not long if you can open the windows in the room after you print.

> Do they settle if I put the printer in a box?

I think they do settle in side the box... But that may just be my imagination.

The big advantage of the box is that you can open the windows in the room while keeping the printer its self toasty warm. Maby this is not a big deal if you dont have to be in the same room as the printer.

Also you may not be annoyed by the fumes... It seems to vary from person to person. Its worth reading up on the safety aspects of ABS and then you can make your own choices.

Edited 1 time(s). Last edit at 05/28/2014 07:07PM by ElectroWomble.

I just printed my first ABS part today, after a year of PLA . I am sharing my details and lets hope you find them useful!
After spending some time in the forums I read about a lot of problems for ABS (mainly bending?). I got really confused about whether to use tape, and what type of tape, or ABS juice, but it's toxic etc etc. At the end I decided to go ahead and print as I was printing with PLA, that is Mendel90 without any modifications, directly on the glass bed (without any cardboard insulation yet).

For my first test I just went into the gcode of a previously sliced part, which was sliced with my PLA settings, and just changed the temperatures for extruder (230C) and bed (110C). And I pressed print! I was really excited when I saw the results. It turned out an extremely good print for me!

I am sure I am going to face problems when I try to print bigger parts but it was a first step in ABS.

Nice print. You may get problems with bigger prints... or you may not! Keeping things stuck to the bed seems to be the most mysterious part of this hole 3D printing thing. Definitely worth experimenting.

Did you do anything to clean the glass?
Also what brand of ABS did you print with?

Hi, thanks. I am currently printing the second ABS part just to make sure it wasn't luck the first time. It seems to be sticking so far...

My glass was a bit dusty so I just removed the dust with a tissue. Nothing else. Also I bought the ABS from here.

Unfortunately before I finish writing this post, the second experiment failed :/.

Maybe its because I didn't clean the table at all this time. Further experiments will follow

Put your printer inside a large box with the 3 sides taped up so they stay together, put something on the top (I use a expanded polystyrene base from a fridge delivery). I also put a blanket on top.

If you're slicing something then start the heater bed temperature early, it takes a while to get to 100+ degrees. You can also put something on the bed to insulate it a bit so it's quicker.
If you're really in a hurry use a hot air gun to speed it up but move it around the bed or it will crack!

Here is my start g=code for ABS:

; START CODE
M107 ; turn fan off (just in case)
M83 ; use relative distances for extrusion
M140 S90 ; set bed temp
M104 S50 ;MIN EXTRUDER TEMP
G28 ; home
G1 X0 Y-50 F3000 ; Go to the middle of the right side
G1 Z5 F5000; down a lot
G1 X-95 Y-50 F3000 ; Go to the middle of the right side
G1 Z0 ; right to the bed
M190 S90 ; wait for bedtemp
M140 S100 ; Set bed temp higher
M104 S190 ; set extruder temp (since this is quicker than the bed)
M109 S240; wait for 240 on extruder
M190 S110 ; wait for bedtemp
G92 E0
G1Z0.05 ; raise a little to the bed
G1 Y0 E5 F200 ; make a thick line to prime extruder
G1 E-0.5 F1200 ;retract a little
G1 X-96.5 Y83.2 Z2.4 F2000 ; move to bull clip
G1 Y90.2 F200 ; wipe extuder nozzle
G1 X-82 Z0.2 F2000 ; move over to the side
; END CODE

It sets the bed temperature before anything else since that takes ages, moves down to the bed to wait, does a priming line and a wipe on the bull clip....
Don't heat your nozzle until you're around 90 degrees or all the ABS will dribble out before you're hot enough...

Depending on the size of the part a big brim is usually a good guard against corner lifting, I usually use 4mm.
ABS juice every print....

Edited 1 time(s). Last edit at 05/29/2014 08:20AM by Traveltrousers.

Has anyone successfully printed all the way to the edges using a stock M90 without an enclosure?

I've been printing solely ABS for over a year now using green PET tape on IKEA SORLI 3mm glass mirror tiles but only for items that stay at least ~40mm away from the edges. Any attempts at printing all the way to the edges has resulted in significant lifting and unusable objects. I am happy with every other aspect of the print.

I have been using E245 and B135 but notice that the corners of the glass still only get to 80-90degC whilst the middle is a more balmy 115-120degC

It currently takes 12minutes to hit 135degC with a cardboard 'blanket' during the bed heating phase so I'm not sure I can crank up the bed much higher using the stock PSU and PCB bed. Will an enclosure (and hopefully a more even temperature across the glass) eliminate this last annoying defect?

I tried slurry before switching to PET but results were less successful - though this was possibly because the acetone was leaching some compound out of my white "Plenty" rolls. I am willing to switch back (I have tons of acetone now!) if I can find the magic combination of settings/preparations to guarantee a better full-size print!

The problems with the edge are not just that the bed its self is colder there but also there is no layer of warm air sat on top of the bed at that point (as it tends to get shunted off by the movement of the bed.) nophead is as ever ahead of us hear [hydraraptor.blogspot.co.uk]

I have had some success with printing long thin things by putting them diagonally across the bed as this makes the ends not quite so close to the edge of the bed. Rounding the corners of the part helps too as dose using a brim or skirt.

That's interesting. I even remember reading that post during an earlier hunt for information but I was looking for something else at the time so it didn't occur to me that it's another symptom of the same problem. Unfortunately this time I have an item that is 1mm away from the edge of my glass so I have no room for baffles.

I wonder if the cushion of warmer air is what helps it stick or the presence of the baffles to help hold down the tape. If it's the cushion of air then I should be able to achieve the same effect simply by making short corner baffles that clip between the heated plate and my wad of cardboard.

Sounds easy enough to prove, so I'll try that later with some cardboard baffles. The objects I've tried tend to lift less at the back corners than at the front, so that would lend credence to the theory that the warmer air (which can escape less easily at the back) helps here.

Yes please tell us how it goes!

Hmm. I ran the same gcode through for comparison but this time with cardboard baffles around the corners, with no noticeable improvement.

Whilst my cardboard baffles are flush with the square corners of the bed they are probably still too far away from the job, unlike the printed ones that Chris uses.

Interestingly, my object consists of a solid layer of about 4mm thick, with some strengthening ribs running on top to act as reinforcement, which adds a further 4mm to the total height. The object looked stuck flat until it got to the ribs (which don't cover the entire surface). It's as if the different contraction rates of the ribs compared to the underlying plate caused the edges to be pulled upwards.

Oh well. Enclosure time...

Hmm I was also thinking of building an enclosure but for sound reduction. Just for clarification, enclosure is good for ABS because it will keep the temperature difference between the bed and air smaller? And therefore the ABS wont bend at the edges?

An enclosure reduces the warping effect. It only eliminates it completely if it is heated to just below the point where the plastic hardens. That is about 70C-80C for ABS which is clearly impractical for a machine made from ABS parts.

Having said that if it stops the part detaching from the bed then the warping is much less than if it does of course. So any heat you do add allows larger parts to be made.

You could eliminate warping for PLA entirely though with a chamber at 50C but you have to make provision for keeping the electronics and hot end cool.

The machine I showed printing ABS right into the corners had a 6mm thick aluminium bed so the heat is much more consistent across it. The downside is it is a lot heavier, so the machine was slower.

---

[www.hydraraptor.blogspot.com]

Hi Chris and thank you for the reply.

I never had any problems with PLA, if I remember correctly. The warping problem only appeared when I started printing with ABS.

So you wouldn't recommend a non-heated enclosure at all? Wouldn't it at least increase the air temperature to some degree and therefore it will reduce the force of warping? This could be enough to prevent you part from lifting from the table and eventually warping.

Why does the aluminum plate matter so much? If I understand correctly, wrapping occurs because the temperature around the part is low therefore the part cools down and contracts. How come uniform temperature along the bed is enough to keep the part from warping?

Thanks,
Demetris

You can't really have an unheated chamber as the bed heats it so the temperature it reaches depends on how well it is insulated. I control an exhaust fan to cap it at 45C and one of my machines reaches that with the heat from the bed, motors and extruder but they are motors that get much hotter than the kit ones. So yes, simply enclosing the machine reduces warping but there is always a finite size of object that you can make in ABS before it either pulls off the bed or rips itself apart. Also it may warp after you remove it from the bed due to built in stress.

As QuackingPlums pointed out above the Prusa bed varies in from 80C to 120C from the edge to the middle but ideally you want it all at ~110C.

---

[www.hydraraptor.blogspot.com]

Thank you again for the answer, I understand what you are saying. I think I will proceed building an enclosure and see how it goes.

I understand that we want the bed to be uniformly heated to 110C. But even if it is heated uniformly how does it prevent warping since the air directly above is what causes the part to cool down. What I mean is the the air will cause shrinking anyway, either you have uniform bed temperature or not. What am I missing here?

Thanks,
Demetris

Edited 1 time(s). Last edit at 06/03/2014 11:09AM by demetris.

Hmm.Now I have even more questions! Have any of these been answered already?:

Does a heated chamber at, say 45C reduce the temperature gradient across the bed? i.e. would it be easier to hit 110C at the corners if the ambient temperature within the enclosure was higher?

If I could somehow ensure the bed stayed above Tg for the first layer, would the temperature gradient for subsequent layers be less of an issue? Is there any reason why the firmware restricts the bed temperature to a maximum of 135C?

Is there a thermally conductive material we could use (such as thin alu sheet with suitable thermal transfer paste/adhesive that isn't electrically conductive) to spread the heat on the Prusa bed more evenly?

Is there a better alternative to the Prusa bed that will run off the 12V ATX supply? I know people have mentioned CF mats and other alternatives but none that sounded like a satisfactory drop-in replacement IIRC.

Is the problem strictly related to bed-temperature, air temperature, proximity to other parts/baffles or a combination? I'd be happy to reserve the outermost 5mm of the bed for printing a baffle (and thus reduce my maximum printable area to, say 190mm x 190mm) if this would be sure to eliminate all lifting issues for example. Likewise I would invest the time and effort into building a PID-controlled heated chamber if there is a "magic" temperature that would cure this issue.


One thing I noticed when I was trying to work out optimum temperatures is that dropping the bed temperature to <120C for subsequent layers resulted in lifting even for small (50mm x 50mm) parts. This was measured temperature at the edges of my parts on top of the glass, not by the thermistor.
If for whatever reason the Tg is higher on my ABS then I doubt I'd stand a chance of ever getting the corners of my bed that hot.

You can also try adding a 1mm x say 20mm tab to each corner which you would have to cut off afterwards...

You could also try pointing some 100w incandecant bulbs at the print....

Just put it in a large carboard box and cover it... you'll soon figure out if this is the optimal solution.

I've been watching this thread for a while now and wondered whether anybody had read Richard's blog here: [mynext3dprinter.tumblr.com]


Alan

It was actually Richard's blog that steered me away from buying a commercial printer and towards the reprap community.

I had read his post about his heated chamber when he first put it up but have been waiting ever since to see evidence that this results in better prints at the edges of the bed - at the moment with just green PET tape and a draught-free room in my house I am already happy with small to medium sized prints.

I haven't really gone for the whole torture test thing but I do print a lot of smaller objects with intricate sections that require engineering tolerances - most of which are fine so long as I stay within my "lift-free" zone. I have a little project now that would really benefit from being able to print right into the corners, and I'm willing to put the effort into achieving this, provided it's actually feasible.

The great thing about this hobby is that no matter how much I think I know about thermoplastics, there is always more to learn! The downside is that there are just so many variables. If I could eliminate some before I start then at least it leaves a more manageable task ahead.

Quote
QuackingPlums
Hmm.Now I have even more questions! Have any of these been answered already?:

Does a heated chamber at, say 45C reduce the temperature gradient across the bed? i.e. would it be easier to hit 110C at the corners if the ambient temperature within the enclosure was higher?

Yes it will do as the thermistor will keep the middle at the same temperature but there will be less heat lost from the edges so the gradient will be less.

Quote
QuackingPlums

If I could somehow ensure the bed stayed above Tg for the first layer, would the temperature gradient for subsequent layers be less of an issue? Is there any reason why the firmware restricts the bed temperature to a maximum of 135C?

You can have the first layer way above Tg to increase adhesion but after it is laid down you want it just below Tg so that it is solid to resist the contracting layers above.

I don't think the firmware limits the bed to 135C. The resistance of the copper tracks increases dramatically with temperature so the temperature stops rising when the power has fallen to equal the heat lost by convection. That is about 135C and the current drops from 10A to about 7A.

Beds heated with resistors or silicone heaters have constant power so the temperature continues to rise. They need a thermal cut-out for safety.

Quote
QuackingPlums
Is there a thermally conductive material we could use (such as thin alu sheet with suitable thermal transfer paste/adhesive that isn't electrically conductive) to spread the heat on the Prusa bed more evenly?

Yes a thin AL sheet will help, but even 6mm AL has some drop at the edges, perhaps 10C.

Quote
QuackingPlums
Is there a better alternative to the Prusa bed that will run off the 12V ATX supply? I know people have mentioned CF mats and other alternatives but none that sounded like a satisfactory drop-in replacement IIRC.

There is this: [reprap.me] but I can't vouch for it. I think the resistance is too high to get up to 135C @ 12V.

Quote
QuackingPlums
Is the problem strictly related to bed-temperature, air temperature, proximity to other parts/baffles or a combination? I'd be happy to reserve the outermost 5mm of the bed for printing a baffle (and thus reduce my maximum printable area to, say 190mm x 190mm) if this would be sure to eliminate all lifting issues for example. Likewise I would invest the time and effort into building a PID-controlled heated chamber if there is a "magic" temperature that would cure this issue.

All of those have an effect. To completely cure warping you need the whole object kept just below Tg and cooled to that as quickly as possible. That means a heated chamber and a fan that blows air at the same temperature. That is what I believe Stratasis machines do and they can print massive ABS parts. I don't think they heat the bed but I may be wrong. There would be no need as it would be at the same temperature as the chamber anyway.

I don't think there is any other complete solution and a machine made from ABS can never print arbitrary large objects from ABS. With a chamber at 45C and 6mm aluminium bed I can print low things right to the edge and I can print fairly big ABS objects in the middle but some things that fit in the build volume would still lift at the corners or de-laminate higher up. I suspect I would be able to print arbitrary large PLA objects but I have never tried. If I needed to print a large boxy object I would choose PLA.

There is contrary line of thought that says cool each layer to room temp as quickly as possible but I don't subscribe to that. It might work for PLA but I doubt it for ABS.

Quote
QuackingPlums
One thing I noticed when I was trying to work out optimum temperatures is that dropping the bed temperature to <120C for subsequent layers resulted in lifting even for small (50mm x 50mm) parts. This was measured temperature at the edges of my parts on top of the glass, not by the thermistor.
If for whatever reason the Tg is higher on my ABS then I doubt I'd stand a chance of ever getting the corners of my bed that hot.

That seems odd as I can print all the parts in the kit and some are bigger than that. However I have rounded a lot of the corners because sharp one tend to lift.

---

[www.hydraraptor.blogspot.com]

Just wanted to share that i've completed 2 long prints yesterday and today using UHU glue sticks instead of acetone slurry. Hopefully adhesion will continue to be good because the setup process is much easier.

Thats interesting to hear Lions3, never tried UHU with ABS.... what's your first layer height?

I'm running 0.3mm for first layer.

Ultimaker 2 ships with a glue stick and I've heard multiple times on the IRC channel that a few people have been using glue sticks. Yesterday was my first time actually trying the process. So far, so good.

Hi Lions3,
Just chipping in with a little on my experience of printing with ABS. I had no success until I added a hood, but I was operating in a cool environment at the time (my garage!). This was all with my older machine, a Prusa Mendel. I moved the electronics to outside the hood, and the heated bed brought up the temperature in the box enough to help. I've more detail on an older blog post here, if you wish to read more detail.
I haven't tried ABS in my Mendel90 yet, but plan to build some kind of surround or hood when I do.
Regards,
NumberSix

---

[numbersixreprap.blogspot.com]

NumberSix, thanks for the comment.

Yeah, my next step is to make an enclosure. I'm thinking of measuring my Sturdy90 and just getting a hardware store to cut some acrylic for me. Then print some connectors to attach it to my MDF Frame. Shouldn't be too bad. Just enclose the front & top and then the center back area. Leaving the 2 recessed areas (melzi/power and the cooling fan) still open.

Thats ruffly what I have done. It works well and by the time the Heated Bed is at 120 the chamber is at 43 - 45 degrease. Once the print starts I run a 60mm fan 12V (same as the one on the fan duct) but I run it from the 5V rails to slow it down a bit. That seems to keep it around 43-45 depending on how warm the room is... I plan to add temperature control at some point. The fan sends the fumes down an old vacuum cleaner pipe (that I wambled ) which i sling out of the window when I print. Works well!