Can't get PETG to stick, looking for ideas

Hi,

I'm desperately looking for some tips since I can't get PETG to work on my printer. I have a heavily modified Anet A6 with E3D direct drive and autolevel probe. I know my heatbed is not entirely even, not even with the glass attached to it but with PLA and the autoleveling, my first layer is consistently spot on and the prints come out great so I believe the bed leveling is compensating enough.

I plan to build a Hypercube evolution and would like to print the parts for it in PETG. That's where the problem starts. I just can't get the first layer to stick. I've spent many hours trying to follow advice I found on the net but it just won't work.

Here's what I tried:

Print surfaces:
- glas
- glas with gluestick
- glas with plain paper
- glas with plain paper and gluestick (works great with PLA)
- glas with hairspray
- glas with lemon juice
- glas with blue painters tape
- aluminium with build tak

Temperatures from 235°C - 260°C

First layer speeds from 3600mm/min down to 200mm/min

In the beginning I always had a clump of PETG around my nozzle but nothing on the bed. Since I read that blue tape supposedly works well with PETG, I did a systematic test with this surface.

Basically I started with calibration for PLA as I'm used to, paper test and z-probe offset adjusted properly. I then went ahead and increased the distance of the nozzle from the bed in 0.05 increments (M851 Z* ; M500) each with different nozzle temperatures between 235°C and 260°C until the clumping around the nozzle went away. At this point the PETG stayed on the bed but it didn't really stick. Corners become round because when the direction changes, the PETG corner gets pulled to a curve, also straight lines curve vertically creating little peaks and valleys. That's when I started to go down with the speed until I reached 200mm/min (2000mm/min at 10% speed for first layer) still doing each speed setting at different nozzle temps. The speed decrease improved the situation ever so slightly so I figured maybe I need to lower the nozzle again a bit. I once again did that in increments of 0.05 but it didn't get any better (again trying different temperatures). Eventually the nozzle clumping started again. Most of the tests were done @ 60°C bed temperature however every once in a while I also tried 70° and 80°. Also all these tests have been done slightly underextruding as seems to be recommended for PETG.

Does anyone have any further Ideas what I could try to get this stuff to print?

Thanks in advance,

Richard

UPDATE: Attached a picture of a failed print. I noticed that the straight line, S3D prints to prime the nozzle was actually quite ok. It seemed, a lot more material was extruded therefore I tried a few step wise increases of the extrusion multiplier. Image is from the last experiment and might be overextruded. Before things looked basically the same, the lines were just much thinner. Same crooked lines and clump of filament where the solid part was being filled.

Edited 1 time(s). Last edit at 08/08/2017 09:27AM by rstcologne.

Decided to document my journey, maybe it helps someone with similar trouble, getting to grips with PETG.

I was now able to get my very first PETG print completed. I realized before that the priming strip S3D puts out at the beginning actually came out relatively well. So I tried to replicate this. With 0.4 nozzle and a 0.2 layer height I can now achieve sufficient adhesion for a complete first layer when I set first layer height and width to height 150% and width 120%. Also I had to increase "infill perimeter overlap" slightly to 20% to close gaps. I print the first layer at 40% of 2000mm/min so basically in slow motion.

Update: This is true for relatively simple shapes. More complex shapes with finer structures still show the same problems, so still some more tweaking to be done.

Edited 1 time(s). Last edit at 08/08/2017 04:35PM by rstcologne.

I could print PETG on a cold bed with genuine blue tape. But occasionally I saw all the symptoms you are noticing too.
What helped me was slicing with Cura and print a brim around the part. ( Cura has a different method to print the brim, it's better than other slic3rs I tried )

In some occasions I had to add "mouse ears" to corners to prevent them rolling off the bed.

BTW, I made the corners of a Kossel XL from PETG and in the first few month, I had to tighten the screws again and again. It seems PETG is a touch stuff, but creeps under pressure. ( more than PLA )

@o_lampe, thanks for the tips. In the meantime, I think I have solved my biggest issues with PETG. So far I tried eSUN green transparent and orange transparent PETG with lots of filament wasted and very limited success (one 2x2x2 test cube actually finished and a part of a new bowden extruder which immediately cracked due to bad layer adhesion. Today a few spools of quite affordable Nunus PETG arrived. I started with a non-transparent light blue.

My setup has also changed. I now have an e3d v6 clone with bowden instead of direct drive on my A6 (Extruder parts printed in PLA since I couldn't get the PETG to work properly). Also I put lamination sheet on my printbed. Works great with PLA by the way and also PETG as it seems.

After calibrating extrusion for the new extruder I basically fired up the settings I stopped my experiments with and tried a print. It stuck right away, however the material was a bit loose on the bed and also way to underextruded. Also I had the impression that the material wasn't liquid enough. So I went down with my z-height by 0.1, increased temp by 5° pushed up the extrusion multplier a notch, felt lucky and upped the slow motion speed I was using before to 50mm/s.

And... Wow, what a change. The machine printed what looks like a perfect first layer. Good adhesion, no gaps, no clumping around the nozzle. Need to check the size of the piece once it's done. It's now printing away happily and so far the outcome looks really great. The new filament also comes out of the nozzle much cleaner. The eSun was somewhat bubbly at all temperatures. By now I assume it might be wet even though it came right out of the sealed bag. The non transparent blue PETG doesn't do that at all.

I'm sure my machine modifications had an impact as well. But I strongly believe the filament made the biggest impact. Here's what works now:
Nozzle temp: 255°C
Bed temp: 73°C
Speed 50mm/s (slowed to 40% on the first layer but not even sure if I need to go down that low)
Z adusted to roughly .5mm higher compared with my PLA settings
First Layer at height 150% and width 120%

I'll do some temp calibration tests next and also work on the first layer to try and get it finer.

I keep replying myself but maybe this thread helps someone in a similar situation as myself at some point.


UPDATE: Print came out ok. Far from perfect, I think it was slightly overextruded and also it's quite stringy but I haven't changed my retraction settings since switching from direct to bowden so I guess it's somewhat expected. At least now I have something to work with to further tune the settings.

Edited 1 time(s). Last edit at 08/09/2017 06:01PM by rstcologne.

I read somewhere, you can recycle PET up to 5 times before it is useless. But for 3D printing only new PET should be used. Maybe eSun hasn't read that yet?

BTW: you are the first person I know, who's using %-values also for first layer height. Most people prefer fixed values for that.
It puzzles me, because the %-value for 1st. layer width is based on layer height...so what is the 1st. layer height value based on in your case?

OK, just checked it in my current slic3r version: %-value of layer height is based on default layer height. ( which is 0.35 )
So using 150% means 0.525mm 1st layer height. That's way to much for a 0.4mm nozzle.

Edited 2 time(s). Last edit at 08/10/2017 03:41AM by o_lampe.

I'll try and dry the spool in the oven. Haven't given up on the eSun filament just yet. I'm really curious to see the result.

I use the percent values since S3D seems to want percentages for the settings I mentioned. At least by default. Maybe this can be changed. I suppose in the end it doesn't really matter, the math to get to absolute values is not that difficult :-)

Ok, results are in. Both my eSun PETG spools spent 5 hours in the oven at 70°C.

While they were in the sauna, I spent some time optimizing my settings with the blue Nunus PETG. Basically I could bring down Z again to my PLA settings, increased overlap between infill and perimeters to reduce gaps, reduced first layer height and width to 100%, tuned my retraction settings etc. Quite a few calibration cubes later I think I'm now more or less at my PLA settings with higher temperatures and some adjustments to extrusion to tune for the filament. Also I successfully printed the X carriage for the Hypercube evolution and the result is great.

Right now, a 2cm hollow testcube is printing with the orange eSun after the oven treatment using the same settings I tuned for the blue PETG. First observation, it's far less bubbly. It stuck to the bed right away, no curling and clumping around the nozzle. The print is 3/4 done, very consistent, the transparent filament looks great. I'm actually considering printing the hypercube part again since it looks much nicer than the blue and was the color I had initially planned. :-)

It seems the main reason for the trouble I had was humid PETG right out of the sealed bag. If your PETG just doesn't want to comply, try drying it in a convection oven. Worked wonders for me. Do use an external thermometer for the oven though since it turned out that at low settings like this, at least my oven was far of. I had to dial in 55° on the knob to get it to 70°. 70° on the knob was actually more like 90°C. Also it overshot quite a bit when heating up initially. You don't want to fuse the filament to an unusable lump while trying to dry it.

I think this will conclude what turned out to become a bit like a PETG printing initiation log and of course the drying wasn't really my idea but maybe it will help someone in a similar situation.

Cheers

You can't be entirely sure, it was the oven that cured the eSun spools. All we know is: the bubbles are gone.
You'd have to print a part with the "old" settings again to outrule their influence.
IMHO, the 1st layer height had an impact too.

Edited 1 time(s). Last edit at 08/12/2017 01:47AM by o_lampe.