Layer Adhesion Problems with ABS-PC and Polycarbonate

I tried UltiMachine Polycarbonate yesterday and ended up having to rebuild my E3D from destroying the thermistor at 310C trying to get layers to bond. I will try the lower layer height and wider extrusion width but I also have the problem that I can not get the stuff to stick to anything. I have tried Mirror + Hairspray, Bluetape + Hairspray, Kapton, and Clean Mirror. The best result was the Mirror + Hairspray but I think there may have been too much hairspray left on so I will maybe try again tonight, this result almost finished the print but I was able to split the parts in half quite easily and the infill was garbage. I will also try to find my gluesticks and try to get something done again but doubtful as my PC was popping and hissing the whole time I was printing.

UHU gluestick on Kapton @100°C heated bed held the Polycarbonate fairly well for me.
Wet filament also contributes to bad layer adhesion IIRC, so your problems might also be caused by that since I have read from others that the Ultimachine PC prints very well at 260°C.

Quote
uGen
UHU gluestick on Kapton @100°C heated bed held the Polycarbonate fairly well for me.
Wet filament also contributes to bad layer adhesion IIRC, so your problems might also be caused by that since I have read from others that the Ultimachine PC prints very well at 260°C.

That is the Purple Elmers stick? I may try drying it in the oven today if I have the time but might be upgrading to 24v today on the ramps.

No, not that I know of. UHU brand glue stick seems to be sort of the German equivalent to Elmer brand. I think you should be able to get UHU in the US, too, but it may be sold under a different name.


Ok, finally got around to photograph and shrink the images a bit.


As you can see, I could still pry the object apart between the layers. This, however, was much harder to do than before. There are also some intra-layer breaks as opposed to complete and clean delamination, which is a good sign for layer bond strength going up.


After initially delaminating the object at the corners with a knife, I pulled the layers apart. Compared to the earlier tests, the object exhibited a lot more tenacity and did not yield that easily. You can also see the white stress marks on the layer interfaces. Another sign that slowly, the material begins to behave. In the original photos, you can see absolutely no stress marks on the layer boundaries.


Here, the knife went diagonally through several layers instead of cutting right through the inter-layer region. Even solid layers (I usually have 3 on top and bottom) separated easily before, but not with this one.


Unfortunately, the next few weeks, I have a ton of stuff to do, so no further testing of misbehaving plastic for now.

Edited 1 time(s). Last edit at 12/19/2013 07:01PM by uGen.

If you are having bed adhesion, abs juice works the best, on a kapton bed as hot as you can get it. infact regular pc will stick to bluetape coated in abs juice even when cold, but it will warp like nothing youve seen before, bringing the tape with it.

[www.youtube.com]

you can see the print came out great it was clear and very strong, I used a thermocouple on my hot end and printed at 265c, however I found a glitch in marlin that if you dont have decimals on all of the steps per mm ie 2560, it drops the decimal points on all of the axis, so I readjusted the steps per mm during the print and it actually moved where it was printing from by almost 2mm. anyways I used a 0.65mm nozzle with 0.8mm wide traces at 30mm/s

Edited 1 time(s). Last edit at 12/21/2013 01:46AM by aduy.

I followed your lead and printed my chinese PC filament with pretty much the same settings that you used (hot, slow and wide extrusion), and indeed, the part holds together much better. I may still need to bump temperature another 5C, but this is very close to where it needs to be. Thanks for sharing your findings, I was about to throw this filament away.

I printed on 90C kapton with ABS juice. It held down well enough to complete the print, but wasn't that hard to remove.

You are welcome!
I found out the wide extrusion part by chance, actually, as I was reading the book "Generative Fertigungsverfahren" (generative production processes), a rather comprehensive book about additive manufacturing for my bachelor thesis. They described each industrial technology, so it just happened that in the part about FDM, this crucial piece of information was revealed to me. And if I didn't read it in this book, I would never have thought that extruding such wide filament tracks at low layer heights with a highly viscous plastic is even possible! All the experience with ABS I had before pointed to a lot of trouble if I was ever daring enough to use these extreme settings, but there we are...

Do you mind sharing some pictures of your finished prints? Oh, and you might also want to try wiping the filament down with some isopropanol or rubbing alcohol. Just in case the surface of the filament is coated with something that reduces stickiness.

Heh, the part is already gone in destructive testing. It held up pretty well, but finally delaminated after a rough encounter with the vise grips. I'll take some pictures of the next print I make.

However, I didn't get the optical clarity that aduy got. I could see through the layers top down, but sideways the walls were pretty opaque. I may have to dry my filament a little more.

If you can look through the layers top-down, it doesn't quite sound like a filament humidity problem to me. But then again, bubbles caused by humidity could have also been displaced by the higher pressure you are applying with the hot, wide and slow method...

Success!!! Finally, I have been able to get PC-ABS to print well!



As you can see, I have already done some destructive testing to see whether my enthusiasm was justified. And indeed it is.
The mangled piece was first tested for layer adhesion by trying to cut along the layers. If layer adhesion is weak, I will at least be able to cut into the object or even split it apart. This was not quite the case. As infill to perimeter overlap distance was not dialed in correctly, I was able to separate perimeter tracks in some places, but after that, I didn't get far. The infill held up quite well as did other corners of the object. No separation whatsoever. The knife made a dent of a fraction of a mm instead.
Next, I tried to compress the end with the hexagonal hole with a grip vice, in which I succeeded. Again, I suspect bad perimeter-infill adhesion to be responsible for lack of mechanical strength.
With a heavy hammer, I then beat on the test piece first striking the top (nothing much happened except of barely noticeable dents where the hammer struck), then the side, which collapsed the part. After having experienced the effects of miscalibrated infill-perimeter overlap, I came to the conclusion that part of the collapse was due to this. Unfortunately, I didn't pay attention to that detail when photographing the object, but the cracks from my first test grew a little bit during this test, but note how there are no other layer separations visible in the photos.

One of the effects of desperately getting PC-ABS to work is that now, perimeter start and end points are barely noticeable due to how I practically ironed layer upon layer.

Still, one thing that is absolutely annoying about this material is that it is rather hard to get to adhere to the heated build platform even at 100°C and ABS juice on Kapton. I will investigate the effects of using new Kapton and a more even application of ABS juice.

I also suspect the filament to be a little bit contaminated with an unknown substance at least on the surface - With the same settings, I have better success printing filament that I have cleaned first. I initially used rubbing alcohol (ethanol), but then, I tried acetone, well knowing that it will dissolve both PC and ABS. However, I wet a paper cloth with a bit of acetone and quickly wiped over the filament to prevent it from being attacked by the solvent too much. Out came a clean filament and blue, powdery residue on the paper cloth. I don't think that wiping this quickly (hundreds of mm per second) with so little acetone - let alone on a paper towel - can melt this much plastic to be noticeable on the paper. I will however conduct more research on this matter. At least after the acetone wipe (which didn't change the filament diameter and surface quality, by the way) and changed settings, I have been printing better than ever before. Next, I guess, I will try "dirty" filament with my new settings for a more conclusive test...

The settings I have found out to work are:
300°C at nozzle
100°C on heated bed
0.1mm layer height
0.8mm extrusion width (yes, iron it on like your nozzle is a steamroller! I guess it helps to have a hot end with a flat bottom like the E3D)
15mm/sec feedrate for perimeter and infill

Just something you should know, be careful running the E3D at 300C. I had mine at 310C and either the thermistor failed or it shorted to the block and caused the cartridge to run wild for about 3 minutes which turned Kapton into crispy black stuff. I now run a modified thermocouple in my E3D and it has done great for many prints now.

Is there some kind of device that can heat the layer previous as it is printing?

Much like when using a Fan to cool PLA,

Use a mini-heatgun to heat the layer below.

@tjb1: Thank you for the warning. Yes, I am aware of the maximum operating temperature of the thermistor. In fact, when heating the nozzle up, the temperature overshoots by 5-8°C momentarily. It didn't seem to affect the thermistor, though.

@ShadowRam: I have a 150W PTC heater designed to be used with small fans. Maybe finally mounting this thing on my printer might help...The X carriage is crammed full already, though and my PSU only supports up to 320W, so that's going to be rather tight.

Just so you know acetone does not dissolve PC, it expands it and makes it cloudy.