PA6-GF impossible to print

Bambu Filament and Accesories Filament
1

I have already read all the other posts about this filament here in the forum, but no one else seems to have the same problems as me.
I keep reading that PA6-GF absorbs a lot of water, which makes the filament very difficult to print with, but the filament was dried in the AMS HT for 24 hours straight out of the packaging and printed from there.

The problem I have is bed adhesion.
The filament shrinks extremely quickly and bends upwards, so that it detaches from the plate after just a few layers and the print has to be interrupted.
I have tried various settings.
I set the bed temperature to 100-120 → makes no difference.
I set the nozzle temperature from 265 (standard in the profile) to 295 → makes no difference.
I tried nozzles 0.4 and 0.6 → no difference.
The chamber temperature is constant 65 degrees.
I am using Smooth Pei with 3dlac as the plate.

It. just. won’t. stick.

All other filaments have produced flawless print quality with my H2D so far. I haven’t had any problems with any other filaments. On the contrary, they look perfect. But PA6-GF is simply impossible.

It’s not like I’m close here. I’m miles away from a usable print.
I am trying something simple like this: https://makerworld.com/de/models/146046-filament-clip-for-bambu-spools

What am I doing wrong?
These problems aren’t caused by moisture, are they?
Did I get a bad batch?
I’m desperate. I think I’ll just throw the filament in the trash as it is.

2

Nowadays I print my nylon on the Glacier plate, it has good grip without any adhesive. Prior to purchasing that plate I would use nano polymer adhesive with PEI plate.

1 Like
3

I have tested with 3dlack too and it brings nothing.
Use the Gluestick! Thats works fine with default profile.

4

Ditto on the glacier, it just worked, all defaults, 0 prep, easy removal, no cleanups, 0.4 nozzle. (light blue PA6-GF)

2 Likes
5

I dont own an H2D, nor the AMS HT; yet PA6-GF isn’t supposed to behave as that. Besides the moisture love, it isn’t prone to many issues, especially with a heated chamber and a specialised dryer.

Dont be. This can and will happen occasionally, and even in the possibility of being a bad batch, you need to do some tests to be certain.

That’s great. What was the moisture content at the end of the drying period?
I assume you’re referring to BL PA6-GF; if not, please indicate the filament in use.

The print files are small, multiple, multicoloured and the profile set for PLA-CF. How did you set up your print for PA6-GF?

I advise making a simple single print, following the BL manual, and everything else by default.
It can be a benchy, or a cube, whatever is enough representative, but not large enough to avoid filament waste and other problems.
Ensure that the moisture content is adequate (as you are doing), and that your smooth plate is immaculately clean, and with the glue recommended.

If that print wraps, fails, take the same pictures, or make a video… to help understand what is happening.

1 Like
6

Moisture content is at 6% now.
Print files are small yes. I printed it single color/filament and set everything to default.
I followed your advice printing a benchy:
time-lapse video: Watch video_2025-08-24_15-50-39 | Streamable

IMG_6547
IMG_65471920Ă—2560 504 KB

IMG_6548
IMG_65481920Ă—2560 256 KB

At first, it looks good. After a while, the bottom bends (because it gets too cold?) and, as you can see, it then becomes round.

7

Thanks.

From the print timelapse and images, you seem to have a consistent first layer, no moisture singes, consistent layers, yet not sure if not printing too fast. It indeed seems a bed adhesion issue.

Some ideas:

  • Redo the bed sheet cleaning, using mild water and natural dish soap.

  • Which glue are you using?

  • Even if I dont see the issue, it will not prejudice drying slightly longer; * From the timelapse, it isn’t possible to understand the print speed, but you may benefit from slower print.

  • Temperatures → are the chamber and plate temperatures constant over the print?

1 Like
8

So another 24h of drying in AMS HT … down to 5%.
Cleaned my bed again using isopropanol → spraying 3dlac → do not have something else (glue or something) …
First print was default settings → default speeds, like 100 to 300mm/s depending on layers … now i reduced everything higher than 50 down to 50:
Seems like almost the same to me: Watch video_2025-08-25_23-08-10 | Streamable

IMG_6554
IMG_65541920Ă—2560 423 KB

I guess the floor will eventually get too cold and bend so much that it comes loose from the bed. I can hardly imagine that another glue would help much. I guess the force is just too strong. You would have to somehow prevent the material from bending too much, which is probably caused by cooling, right?

Yes temperatures are absolutely constant. As i said … other filaments on this printer (PLA, PETG, ASA, PC FR) are clean as ■■■■.

9

I’d say just give this a try, BIQU Glacier build plate, or other PA capable glue-free plates recommended in this thread.

imo bed adhesion is a basic problem that should just work once adequate temperatures have been applied and H2D is more than enough to deliver the needed other conditions, so a plate that can print most high temp filaments glue-free is well worth the price.

The only other thing I can think of is PA don’t like layer times being too long so batch printing in by-layer sequences should be avoided but I don’t think that’s what’s going on here :thinking:

10

What vendor’s PA6-GF are you using?

I’ve printed 3DXtech’s in a P1S before, but not as tall as a benchy. I was able to get away with it by heat soaking the bed at 100C for a half hour and using Magigoo PA. And luckily not needing to do anything over 10grams. When I tried to run my normal calibration print it showed up like your benchy too :smiley:

I’ve heard everyone and their brother recommend the Vision Miner Nanopolymer adhesive, too, if you don’t want to go garolite bed

11

image
image1920Ă—1462 182 KB

Mine (from bambu, printed on glacier) did fine printing this desiccant container. Not that there’s need for PAs here but it’s a useful tool to benchmark (porous, minimal bed contact, tall, thin layers, weird geometries + concentric infill testing extrusion forgiveness) new filaments, while still being something useful for dry boxes.

Not the best looking result but at least perfectly flat bottoms.

12

I use that one: PA6-GF | Bambu Lab US Store
I got it for almost half the price in the last sale.
The BIQU Glacier build plate is the last thing i could try.
And if that does not work … maybe it will become a premium clothesline… but because the filament is so brittle, it probably isn’t even suitable for that :confused:

13

To be honest, my bag of PA6-GF is still in its bag and waiting to be tackled. I got it specifically because I like a challenge and read that it is quite tricky regarding warping and curling.

From my experience with warping/curling and similar defects in FDM and related processes:

120°C should be the best.

Using the 290° (upper Bambu recommendation) for the first layer and the profile default (265°C?) for subsequent layers should give the best anti-warping strategy.

A larger nozzle can, to some extend, introduce more material per time/layer and hence more heat, promoting uneven cooling which in turn causes warping and curling. Using larger nozzles/higher flow is detrimental regarding warping and curling.

Excellent. Some people prefer to have it sitting at that temp for a little while (15-30 Min IIRC) to ensure a good heat soak before starting a print.

Slowing down (usually) is a very effective anti-warping/-curling strategy. There’s still some room to the bottom, but you have already slowed down a lot.

Another, equally effective method to reduce the heat input per layer (and hence reduce the uneven cooling pulling the print up) is to halve the layer height. That will reduce the heat input by a factor of 8, reducing curling and warping.

The video appears to show an initial defect in second 6, with the print being pulled of the bed shortly after. This indicates to me, that curling is pulling the benchy into the path of the nozzle. Without these collisions, the print may have completed fine.

Two more items from the video:

  • A brim may help to improve adhesion.
  • Is that a “Grid” sparse infill? It should work OK in principle. But with curling, you may want to try Honeycomb. It is locally thicker, giving a slower and hence more even cool down.

So even though you have almost exhausted the key counter-warping/-curling strategies, you may have a chance yet with

  • a lower layer height,
  • adding brims,
  • going for a different infill and
  • going even slower (although that could really be a bummer if it was neccessary).

Please keep us posted. At the moment, I am just trying to transfer my experiences to a challenging material which I do want to print soon. And of course, please accept my apologies if none of this actually transfers to PA6-GF.

:crossed_fingers: & :four_leaf_clover:

1 Like
14

Thank you so much for this recommendations! I will try these as soon as possible.
But my printer is currently busy printing other things and there are still a few objects in the pipeline. I don’t know when I’ll be able to get back to this filament. But I’ll definitely try again!
In the meantime, I’ve also got myself the BIQU Glacier build plate. Maybe that will help things go a little better.
I’ll be in touch!

1 Like
15

So I gave it another try. This time with brim, gyroid infill, and a new engineering plate. Still the same problem. After a while, the material lifts up and detaches from the plate due to deformation/cooling. This has nothing to do with print speed. It’s just a matter of time.

I talked to a friend about it recently. He has an X1C himself and I gave him my roll. The amazing thing was that he was able to print a benchy without any problems on his second try. So I can rule out material defects.

So it’s clear. Either, and most likely, I’m doing something wrong, or it’s the H2D.

CB61F3F4-D9F0-4328-8596-8E27766BEE58
CB61F3F4-D9F0-4328-8596-8E27766BEE581920Ă—2560 191 KB

16

You said earlier that you “don’t think glue would help much”, which is wrong. Adhesive not just a recommendation, but pretty much a prerequisite to print nylon. Get the Bambu cylinder stick glue and don’t look back.

17

What filament and print profiles are you using? Is the printer in a cool room? Things you can do:

  1. Raise the chamber temp to 65c on the filament profile.
  2. Raise the bed temp to 110-120c.
  3. Slow down the print, use the “high quality” profile and lower layer height like 0.16mm.
  4. Manually pre-heat the printer for 15 min before printing.
  5. Rotate the object 90 degrees and place it on the back right part of the plate where the heat is most stable.
  6. Use a cover for the poop chute to avoid cold air being sucked in, lots of simple/quick ones on MW.
  7. Whatever you do DO NOT open the door during printing. When print is done let it cool naturally for 15-20 min before removing print.
  8. If all else fails, get some Nano Polymer bed adhesive and use it together with the Glacier plate.
18

I made two changes to the default Bambu PA6-GF profile

Bed temp and chamber temp both at 50c

I used the standard texture plate with PA glue

This combo has worked for me every time.

See my post