Cure for The Common Curling: "The Brim"

+1 to the brim.

I found out about this technique about a month ago and it's nearly eliminated close to all my warp + curl. would love it if there was a way to include a small brim when printing multiple objects at once.

richrap Wrote:
-------------------------------------------------------
> What part are you printing, it does not look like
> it should be warping from that photo, I would like
> to give it a try on heated mirror-glass.
> Just checking you are printing in PLA? and onto
> PET tape?

I was printing this part: mf70-cnc-x-mr126.stl from this thing: Proxxon MF70 CNC Conversion Kit

It's not the easiest thing to print, because it's rounded it has a somewhat smaller footprint than the cross-section and the shape seems to produce quite a bit of warping force, especially when you use some extra perimeters because this part needs to be pretty strong.

I was printing it from polycarbonate, on PET tape wiped with dilute ABS juice (since ABS and PC mix) and heated to 140C for the first layer, falling down to 110C after that. I'm sure it could be printed from PLA with no problems at all, but I made my comment because I read some other comments claiming that warping has been "solved" by the brim even for high warping materials.

ttsalo Wrote:
-------------------------------------------------------
> Ok, let's take a reality check here. Yes, "the
> brim" will offer some extra adhesion for the
> object's first layer. SOME. Not infinite. The
> reality has not changed, as soon as the warping
> forces exceed the grip, the object will rip the
> brim off the build plate and warp. Like this:
>
> [forums.reprap.org]
> name=7368045606_c797155efb.jpg
>
> The claims put forth in the first message, like
> absolutely stops all warping and curling are not
> true. Don't take me wrong, I wish they were! And
> even in the pictured object, the brim delayed the
> warping for so long that I ended up with an usable
> object, unlike on my first try without the brim.
> But warping is still here with us and requires
> consideration of what can be printed from what.

Sorry this method only took you from non-useable to useable printed parts j/k

You are right, [probably] nothing is infinite... And you are correct that one helpful method might not necessarily fix all warping on all printers, with all materials, everywhere. (Mr. Polycarbonate Filament)

The Brim sure as heck helps in most cases i.e. it landed you useable print

It is easy to beat the brim with a few not recommendable printer settings

It looks like you will indeed need to make some other adjustments to finish your fight with warp.

Good luck chasing out the last of your warp! I expect a re-post on how you finished the fight!

Edited 4 time(s). Last edit at 06/20/2012 02:04AM by Idolcrasher.

@ttsalo One more thought!

Crank up the loops man! Lay down a 35-50 loop brim and then come back here and tell us it still warped

If you end up having trouble getting a 15+ loop brim to stick to the plate, then I would blame first layer adhesion problems as a root cause of your trouble.

Let us know

It is neat to hear someone else acknowledge that extra perimeters means an increase in warping force. I have found this to be VERY TRUE!

I do not think it common knowledge either!

You are going to have to have a cancel out that warping force with an appropriately sized Brim.



***Break***

Silly side thought.

I wonder if a fairly simple equation could be derived; the factors being the bottom layer surface area, the number of perimeters, perimeter length, and material type. Perhaps some rough calculations could be made on the amount of "warping force" and how much "brim area" is needed to cancel it out.

A little crazy to be worth while; but an interesting thought.

Anything that makes the part stronger increases the warping force.

Even if you manage to hold it down during the build, parts can warp after you remove them if they are left with residual stress.

---

[www.hydraraptor.blogspot.com]

Idolcrasher Wrote:
-------------------------------------------------------
> @ttsalo One more thought!
>
> Crank up the loops man! Lay down a 35-50 loop
> brim and then come back here and tell us it still
> warped

Here's another object, with a 50 loop brim.



Once the warping force exceeds the local adhesion, the brim right next to the object will get ripped off and as the warping continues, it will continue ripping more and more of the brim off. The brim will slow down the warping and keep the object attached to the build plate even after some warping, but it doesn't eliminate warping.

Wow!,

That's a lot of loops! It's really cool you gave it a go! WOW that is a lot of warping force!

That object does not look like it should have that crazy amount LIFT! I don't want to out my foot in my mouth, but I feel like I could print that object in PLA with my heatbed off and get away with minor warp... I would take up that very challenge, but I am moving in to California in a week and everything is packed up.

I can't really give advice on the polycarbonate filament you are using... I am not familiar with its properties at all (where do you buy it?) What kind of properties does polycarbonate filament have?

I wonder what would happen if you made the brim more than one layer thick; perhaps another layer would give it more strength?

I might also recommend laying down small patch (big enough for the print) of blue tape right on top of the bed; Blue Tape can add a ridiculous amount of first layer adhesion.

Wow! Thanks for taking up the challenge man. Post here how you finally beat it so we all can learn!

discocyborg Wrote:
-------------------------------------------------------
> +1 to the brim.
>
> I found out about this technique about a month ago
> and it's nearly eliminated close to all my warp +
> curl. would love it if there was a way to include
> a small brim when printing multiple objects at
> once.

@discocyborg Exactly! That is what I am hoping comes out of this post! I hope that "the brim" is recognized as a useful feature and added to our favorite slicing programs as an option.

@ttsalo So Polycarbonate Filament was added to the RepRap Wiki in March (3 months ago): [reprap.org]

Is it really that new? Or just new to me?

It's used to make bullet proof glass?

Did you get your batch from RichRap's experimental batch?

Look bud, I made no claims about "high warping materials" I was speaking generically about PLA and ABS, aka the common materials. Two forums pages of text and other folks confirm that a brim was helpful to them which I am happy for; but I can't claim to know anything about fighting warping in new, experimental or uncommon materials.

You have sparked some curiosity in me though.

What is polycarbonates "glassing temp" and is your heated bed under it? Does polycarbonate even "glass" or does it act more like a gel (smooth transition from liquid to solid). Should we expect that this material is "high warping?". Is it so bad that we should steer clear?

At least it is now semi confirmed that a brim can "help" a polycarbonate print.

It is possible that If your heated bed is too hot, the bottom layers will stay soft and pull up to the cooler layers which are away from the bed.

Either way, it looks like your first layer adhesion needs a kick in the pants; try the part again with blue tape (then again god knows what POlycarbonate is going to do on blue tape; i bet it will help)

Do you get the same warping problem on the same part or other parts with ABS/PLA with or without a brim?

@TTSALO, "You had my curiosity, but now... You have my attention"

Edited 1 time(s). Last edit at 06/20/2012 10:55AM by Idolcrasher.

Removed repost.

Edited 1 time(s). Last edit at 06/20/2012 10:56AM by Idolcrasher.

The RepRap wiki says Polycarbonate flows @ 300C and calls Polycarbonate a low warping material. So why do you call it a high warping material? Because it warps on you? PLA warps on me when I don't use it properly

The wiki mentions a bed temp of 85 C printing onto Kapton, I am guessings that your +100C bed temp is high.

Edited 1 time(s). Last edit at 06/20/2012 11:16AM by Idolcrasher.

Hello,

I'll probably add a native brim feature to Slic3r. Unlike the skirt workaround, it will put a correct brim around each single object instead of around the whole print. Also, it will be attached to the object continuously instead of computing the convex hull (the first pictures by IdolCrasher show that the convex hull leaves gaps where the part gets convex).

I think in a lot of cases the convex hull might be better than continuous, you should probably test it.
The convex hull works in a similar way to the MickeyMouse ears some use on outside corners to prevent lifting, the continuous "Brim" might actually not work as well.

Hello, a native brim feature was implemented in Slic3r! Next version will be out soon.

Sound Wrote:
-------------------------------------------------------
> Hello, a native brim feature was implemented in
> Slic3r! Next version will be out soon.

That is so awesome!!!! "The Brim" TM is going official on Slic3R!!!! Buwahahahahhaaha!!!!

It is awesome to hear that each object (when printing many things at once) will get a Brim as well! FANTASTIC!

The continuous attachment will be neat as well!

THANKS SOUND!!!!!

Edited 1 time(s). Last edit at 06/23/2012 11:13PM by Idolcrasher.

That's great news - Looking forward to the next Slic3r Release

---

[richrap.blogspot.com]

Thank you very much for sharing this! I stumbled across this thread and started to use "The Brim" and I have no curling! I would much rather use a little bit more plastic for "The Brim" then to have to scrap a half or full print!

Thank you!

CSM Wrote:
-------------------------------------------------------
> Thank you very much for sharing this! I stumbled
> across this thread and started to use "The Brim"
> and I have no curling! I would much rather use a
> little bit more plastic for "The Brim" then to
> have to scrap a half or full print!
>
> Thank you!


No prob CSM!,

I'snt it a great feeling to successfuly pull off a difficult print that had been giving you trouble!?

I knew that I couldn't be the only one having warping troubles; a keen eye on Thingiverse shows warping/lifting problems VERY OFTEN! Folks on Thingiverse just try to use their best prints when taking pictures for posts (rightly so I suppose), but the warping problems are there if you look closely. Everytime I see warped prints on Thingiverse, I PM that person and suggest the Brim. A bunch of folks have written me back to say that it solved their problem; it makes me feel warm and fuzzy inside

The next VERY REAL problem that needs a SUPER SIMPLE fix is extruded plastic lifting upwards while printing an overhang. The solutions I use right now, to beat "Overhang Lift" are far from universal... But as soon as I come up with a fairly universal solution, expect another "beer swinging" forum post, and a drive to get it added to the slicing programs!

Edited 1 time(s). Last edit at 06/23/2012 11:15PM by Idolcrasher.

I think the solution to overhang problems is proper cooling. Just cool the plastic right after you extrude and it shouldnt have any curling problems. Crank up the fan full blast! We need a slicer that turns the fan on for overhangs, that's a feature missing in a lot of programs nowadays.

I have a pentiant for using cheap filament and being bitten by it

Cheap filament "overhang lifts" like it is going out of style!

I have dual g-code controlled fans and still have trouble battling overhang lift with www.repraper.com filament.

I dig the idea of full blast fans during overhang printing! Got any other advice?

So, the fix for warping is a raft? You know skeinforge has this setting, exactly for this purpose...
It's called raft in SF. And it works really well, and is optimized to the size of the part in SF...

I like slic3r, but if you're having trouble with parts, use SF and the right settings. I've had many a part with better results in SF. Though I still go to slic3r first, just because it's pretty/easy.

Not sure that's the proper use of the term raft bud. Every raft I have seen is actually printed layers that the object sits on. In this case, there are no superfluous layers for the object to sit on.

The Brim just got added to Slic3r by its author, Sound. Mostly for being awesome and useful.

I was printing with polycarbonate bought from Orbi-tech. It's not the same stuff that richrap or protoparadigm had, so it may behave differently. I haven't seen anyone else's results from it, so I don't know whether my results were normal or abnormal. Polycarbonate should have same shrinkage factor as ABS, but since it's also a tougher and stronger material, it may produce more warping force for the same shrinkage.

I think the adhesion was pretty good, yet another print try, the 190x120mm Y-carriage, ripped the PET tape clean off my aluminum print platform(!). I replaced the tape with Kapton tape but I'm away from my printers for the next three weeks so I can't run more tests right now.

In any case, the whole warping thing is a major annoyance because there seems to be no repeatable test whose results could be generalized to many objects. Once I get back, I'll try to figure out a test for it. I think that basic warping force should be proportional to the cross-section of the object times the length of the object (times the density, times the shrinkage factor), and the adhesion proportional to footprint (times the Brim Factor, times the adhesion per square mm). I was thinking of an upwards-widening block with a 5*100 mm footprint and the test would be to see how high you can get before the ends lift. If this works, it should allow isolating the different factors that affect the warping.

Just tried Brim in Slic3r and I'm sold, was printing a 160x160 part (7 hours) without any lift. Good job!

Two comments about Slic3r's integration:
1) Brim should be inside as well as outside the part. The part I printed today was a 5mm thick cuboid frame, which could have benefited from being 'Brimmed' inside as well as out.
2) The sequence for perimeters should be changed when Brim is being used (I feel), Brim is performed outside -> in, whilst perimeter is performed inside -> out, it would make more sense to make the perimeter also outside -> in, to be more continuos (possibly makes it harder to peel brim off?).

Anyway thanks Idol Crasher, and as always thank you Sound!

Quote

times the shrinkage factor

Note that the shrinkage factor is how much it shrinks when it cools from the temperature at which it becomes hard to ambient. That is why PLA is so good. It only goes hard at about 55C. PC has a glass transition temp of 150C, which is much higher than ABS, so even if it had the same coefficient of expansion it would warp a lot more.

---

[www.hydraraptor.blogspot.com]

A "BRIM" of six loops on a 78 mm x 6.5 mm disc definitely cured the warp problem for one particular object I have been stumbling with for quite some time. Working with PLA, I expect about 2% shrinkage as the material cools. If, as in the case of a disc such as the one I am working on, the center of the object has more thermal mass than the edges (which are going to cool faster than the center), warping is to be expected. I noted that after the ruined pieces were allowed to cool to room temperature, they flattened out.

I suspect different geometries will require different BRIM configurations.

Could moisture absorbtion affect PC warping like this?

jason.fisher Wrote:
-------------------------------------------------------
> Could moisture absorbtion affect PC warping like
> this?


Probably. I read that PC sucks up moister like it is going out of style.

In case you don't use slicer, there is an openscad module that can add something similar (the author calls it "print anchors") to any stl : [www.thingiverse.com]

---

Most of my technical comments should be correct, but is THIS one ?
Anyway, as a rule of thumb, always double check what people write.