Printing ABS with UM²

[lordmao 0]

Hi everyone,

I am trying to print a part in ABS, but the shrinking is driving me crazy. Neither heated bed nor glue will prevent it.

But I have noticed that it always happens in the same corner of the building plate, the back left corner (the closest to the extruder origin position).

 

I have tried to turn the part around and it always start from the same corner.

I have to abort the print before the fan carter hit the shrinking ABS...

 

The heated bed is set to 80°C and I'm printing at 250°C 20mm/s 0.1mm layers. I have tried other settings, without improving results.

 

Is there a way to avoid, or at least limit the shrink? Without a heated chamber that I don't have?

Do you have any idea what could be the cause behind the shrinking always starting from the same corner of the building plate, no matter the orientation of the part?

 

Thanks for your advices.

 

[lordmao 0]

I have to add that the first 5 to 10 layers are perfectly fine, but it start lifting after that.

And from then, the effect only increases, until touching the carter around the fan.

 

[colorfabb 3]

I have to add that the first 5 to 10 layers are perfectly fine, but it start lifting after that.

And from then, the effect only increases, until touching the carter around the fan.

 

You could try printing without the active cooling enabled, or at least turned down by a lot. This works fine for me on a ultimaker original with heated bed at 100C.

 

[lordmao 0]

Thanks for the tip, I will try that and I will let you know if it worked.

 

[gr5 2,167]

Most people print ABS with the bed at 110C. The glass temperature is around 105C so if you go over that then the material is still relatively solid but you can bend it - it is soft. Soft enough that it won't lift off the bed.

Make sure the bed is below 90C before removing your part!!

Similarly the glass temp for PLA is around 50C to 60C depending on additives.

It's fun to take a thin sheet of PLA or ABS - the dimensions of a thin coin. And place it on the bed and set the bed to just below or above the glass temp and you can let it heat up for 5 minutes and then try to quickly pick it up and bend it or use pliers.

I made 20 of these "coins" and I should leave them on the bed and then I can let it cool 1C at a time and try to find the glass transition.

 

[gr5 2,167]

Why is rear left corner the worst? I think it has something to do with the 3rd fan - the rear fan. And the nozzle placement. The nozzle isn't centered in the print head.

Also maybe something to do with being far from the center of the print bed where things are warmer.

 

[Dim3nsioneer 557]

Apart from what has been written so far in this post, I see that the print has warped while the brim still stays flat on the bed. Thus, it looks as if the connection between brim and print is not good enough. I also think I see the lines of the brim having some space in between them at the right back corner.

If this is the case (it is hard to see in the picture if the brim really still has a connection to the bed or has just fallen down again) then there might be additional issues like underextrusion during the brim / first layer, too large distance of the print head from the bed during first layer, an unevenness of the bed (not very likely with a glass bed, I have to admit).

Anyway it might be an idea to check the distance between nozzle and bed at different spots and especially to compare the distance in that corner to the others.

 

[lordmao 0]

Yes the brim was still flat on the bed, and yes there was some space between the lines of the brim.

I won't go for under-extrusion as every other portion of the print was just fine and the speed very slow. But I will check the bed level to see if it might come from there as well.

Readjusting bed level, heating bed at 110°C and turning completely off the fans, these should help me print that out.

Thanks

 

[simon 1]

I haven't received my UM2 yet, and this is precisely the sort of concern I have about it. As the printer is supplied with PLA, not many new users have tried ABS yet it seems. I use an Afinia H series and only print ABS. The heated bed is set to 105 automatically, although I think it only reaches about 90-95. I always have the fan set to divert, so the part gets no fan cooling. I've printed literally 100's of parts like this, no problem. I don't think you can eliminate warping with ABS, certainly not without a heated chamber. But hopefully you can manage it to the point where it works well enough. The Afinia uses an intermediate platform between the heated plate and the part. They call it a 'perf' board (short for perforated I guess) and it looks like one of those blank PCB's you make projects on, but without the copper. The benefit of this is that the first layer of the raft actually squirts into the surface and bonds itself there. I run the printer in a metal cupboard to keep a good ambient temp (about 25 ish) and prevent drafts. After printing I remove the board and then with a sharp scraper, slice the part off. Then you peel off the raft. Works perfectly every time. The perf board provides just a bit of compliance (movement/flexibily) between the heated plate and the part. If there's no give in the system and the plate is very stiff, the stresses that build up in the part seem like they are strong enough break the bond between the part and glass. I'm thinking that I may have to modify my UM2 to replicate this proven method of printing ABS. I might even consider removing the side fans because I won't be using PLA, which would reduce head weight as well, but time will tell. If you make any progress with ABS, good or bad, please continue to share your results. I will when I get mine.

 

[porter 8]

hi all

have a makerbot rep 2x still waiting on my Ultimaker2, and use it only for ABS prints. The printer has a 90/95% success rate. i use the following setting for ABS:

print bed at 115c,

print temp 230c,

use raft and level the bed on the tight side of the tolerance.

this printer is supplied with Kapton tape which helps improve adhesion and reduce warping, would recommend it.

hope this is of some help.

 

[lordmao 0]

Okay... Now I probably have tried to be too accurate while leveling the building plate. The first layer just won't print, the nozzle is so close to the building plate that what I have can't really be called a layer. It's thiner than anything I thought was possible, and the feeder was crying a bit.

So I just have to wait for the build plate to cool (soooooo long to heat up and sooooo long to cool afterward) and restart the leveling process again.

 

[Dim3nsioneer 557]

I guess, you know about the sheet-of-paper-leveling? :smile:

If you have an equalized leveling which is just a bit too close or too wide, there are some ways you can adjust it with the software by tweaking the start.gcode (if this is possible for the UM2; it is for the UM1). But just do this when you feel comfortable with it...

 

[lordmao 0]

Yes I am using a sheet of paper to fine-tune the leveling :wink:

But I will not try tweaking the gcode for the moment. I will wait to see if I can solve this "normally", as it's the first time I'm having this issue.

 

[gr5 2,167]

Why do you have to wait for the build plate to cool? If it is a failed print then you don't have to worry about damaging the print and you can just scrape it up with a putty knife while still hot. And if it's a "good" print you only have to let the build plate cool to 85C or so before carefully prying it up.

 

[lordmao 0]

I do agree with you, even if I already have burnt myself a few times for being too hurry :mrgreen:

So now I prefer waiting a bit more. But it's not that long to cool, only a minute or so. The realy long phase is the heating of the bed.

 

[lordmao 0]

There is still one hour to go, but I think that I can safely announce that it will do.

Only a bit of lifting off.

I am printing at 250°C, on a 115°C bed, fan speed 10% (fans off seems to decrease the quality too much), 0.1mm layers and 20mm/s.

That way I'm also avoiding stringing and I keep quality high.

But I had to start the print with the bed temperature set to 80°C and only after the beginning, rise the temperature to 115°C. Otherwise the ABS is melting in the nozzle for too long before being extruded - it takes the bed something like 15min to reach 110°C - and it is stuck in there. The feeder is then digging/cutting the filament. So I had to change material, cut the "bitten" portion and reload. The first material to come out of the nozzle was then yellow/brown (who said overcooked?)...

So starting the print with the bed still warming up seems to do the trick.

 

[simon 1]

I've been wondering if frosted glass would be an improvement for ABS. It's still a flat standard but the increased surface area might provide some extra bonding between the part and glass, which might help to prevent any lifting. I'll try this on my UM2 if and when I get it.

 

[joergen 2]

I've been wondering if frosted glass would be an improvement for ABS. It's still a flat standard but the increased surface area might provide some extra bonding between the part and glass, which might help to prevent any lifting. I'll try this on my UM2 if and when I get it.

 

frosted glass has a beautiful surface, but less adhesion than regular flat glass. it is not so much the surface area, but maintaining the critical temperature for adhesion. in other words, as soon as you have a tiny bit of airflow getting under the plastic, you loose adhesion locally, and then you have a cascading lift-off event. the acid etching of the frosted glass does help with the airflow, while flat glass has a better air seal, and hence gives better adhesion

 

[simon 1]

Well that's saved me wasting a little time and money on an experiment ! thanks :-)