Unable to print with Bambu PAHT-CF - it destroys the printing plates

I don’t know what is going on but the PAHT-CF is sticking so hard and shrinks that it essentially destroys the print plate (both Engineering and Hot). I even pre-heat the chamber to 50c before starting the print.

It prints fine for a few hours. I can see the chamber is a constant 50c (I tried plate temps of 100c, 110c and 120c with same results).

I also dried the filament for 14hrs using the drying function and printed from a Sunlu filament dryer.

After a few hours (chamber steady at 50c) I notice that the area in front of the camera starts to looks horrible and when I take a look I can see that the metal plate is coming of the magnet. When I look closer it clearly warped / shrank etc. as when I remove it - the plate goes back straight.

BUT the surface is essentially ruined, even the writing comes off. So this isn’t an adhesion issue but a problem with shrinking / warping:

1 Like

That bottom layer looks weird, I almost always get a perfect one with PAHT-CF and I’m printing on Engineering Plate almost exclusively with it. I use Magigoo PA as adhesive. I probably haven’t printed such large parts but I’ve never had it warp that badly. I also haven’t had the text come off like that but I try to avoid that region and sidies of bed with high-temp filaments anyway as the temperature is not really that uniform in there)
I would probably add some clips to the sides of the bed if it warps that badly, but you can also try adjusting the model to alleviate some of that warping, like printing the first few layers slower, using fewer perimeters and a different infill pattern (gyroid should be more springy I think, which is both good and bad) or just split it into multiple parts with smaller area. Or use a raft… No good options if you need a solid part with 100% infill or something like that…

Do you have the aux fan on? I think that will keep the heat uniform in your chamber, I could be wrong though.

That is quite odd! I have never experienced that myself. I have had parts come up off of the build plate, but I exclusively use the high-temp plate for basically every material I print (primarily PAHT-CF and Support-G).

I don’t really know why that would be. You might check the readme for info on how to attach the build surface sheets to the metal plate? X1 series Manuals | Bambu Lab Wiki

How large are the prints. I am ok with smaller, but if I have a print that goes across the build plate - it shrinks too much. I had a look at the manual and it essentially says to avoid shrinkage to use less infill and configure the infill to be circular.

I tried that to no avail … I must ask Support as I am getting the same with ASA and I never had problems with ASA.

So yea - I wonder if the heat distribution has something to do if you use the whole printbed with high temp matieral.

I am actually trying to flip it 90 degrees now so it is tall rather than wide (although it needs plenty of support).

1 Like

It is somewhat common to print at 45 degree angles, but a giant brim around the bottom may help hold it down a lot better.

Don’t know if you see the pictures. The problem isn’t to hold it down because with enough glue it won’t come off. But the shrink force is so strong that it pulls the metal plate off the magnet. If I don’t use glue the metal plate stays down but the item still shrinks and lifts.

apparently only me and you use this forum, official bambu forum must be elsewhere idk

anyway, you could try a “raft” instead of a brim, the raft raises it up off the bed and should make that impossible. the warpage idk you need to keep the chamber closed and cooling fan off, typically.

it might turn on a single fan if it gets too hot or something, not totally sure.

It looks like he was printing a spool winder. I printed one with PAHT-CF and it worked just fine.

I second that, PAHT-CF works wonderfully on my machine, the only part I printed that spanned the whole bed width tho, it did warp pretty bad. Pulled up from the high-temp plate, however. I think that’s something that will require printing it slower.

I would ask, is he using glue stick? I don’t use glue stick, just keep the high-temp plate absolutely oil-free (dish soap *cuts grease/oil!)

So with glue stick it bulls the bed up, without it pulls itself up a few hours in.

I am confused. Ok which orientation / profile / print plate / glue / infill setting etc. etc. because based on Reddit I am not the only one

1 Like

I am just drying it again to try again.

Would you be able to share the m3f so I can check the settings you used or tell me, especially around speed, temps and infill, settings you used ?

You can try using Magigoo PA.

It’s an adhesive specifically made for 3D printing. It will hold a part stuck to the bed when the bed is hot and release the part when it cools down again.

Magigoo acts as an interface layer here, so it protects the build plate = extends its life.

Thanks. As I say though. Adhesion isn’t the problem. The shrinking of the object is. If adhesion is too strong the plate comes up and gets moved off the magnet. If adhesion isn’t strong enough the object lifts.

1 Like

Have you tried heating up your bed to like 100c for a while until your chamber temp get up and then start the print? Can’t hurt.

See original post first paragraph :slight_smile:

Yes I have.

I think you are up against the limits of the machine with that size print with that material.

The shrinking/warping forces are too strong when the prints get bigger like that. If you could somehow get the ambient air inside the chamber hot enough it, would solve the problem because you wouldn’t have the temperature differential with the bed and the air inside the chamber. But, without actively heating the air in the chamber, I’m skeptical that these machines will print PAHT-CF prints that fill the bed like you are attempting to do.

Yea I wondered as much. I don’t think heating the enclosure to that temperature is wise with all those components anyway. Gonna do some more tests based on their knowledge base (I.e less infill and non linear and see what happens)

I would suggest not using 100% infill on the outer 10mm on the part. It still may pull up but that’s where I’d start. Bambu Studio will allow you to use basic shapes to apply special settings for infill, so you can easily play with it. Personally I’d do 4 walls at 20% infill for the outer section and go from there. As well, instead of going with 100% infill just try printing the whole thing with 6 walls at 15% infill and see if that is strong enough. Many people assume the 100% infill is needed for a strong part, but that’s not always needed. The walls and top and bottom layers are super important for strength. For example, on your print, I’d bet you’d be happy with 15% Grid infill with 6 walls and 4 or 5 top and bottom layers.

On the plate damage, make sure you have glue everywhere. Glue is good to help hold on when hot, but more importantly, helps to release when cold. Glue is effectively a disposable plate surface that will rip off glue instead of plate.

Note: The plate sticker can be replaced, but I think the Engineering side is not replaceable. But I believe you can stick a second High Temp or Cool sticker on it, maybe someone else can confirm.