I finished the print yesterday with the PAHT-CF. It was fresh out of the bag, no drying.
The surfaces that were not overhangs looked beautiful, but the overhangs were blotchy and look bad.
I took the PAHT out of the AMS and put it in the S2 at 70°C overnight w/ a 4mm screw propping the lid open a smidge. It was reading 14% this morning and I popped it back in the AMS and am re-running the part.
I’m running on .08 High quality and the overhang angle (15°) is not generating any supports on the blades. If I crank it up to 45° it’ll generate supports in all the bad areas, but absolutely kills the print time.
When I used to print on an Ender w/ Cura there was a setting in there to print overhangs from the inside to the outside. Is there a setting like this in Bambu?
Or do you think it’s just wet material? I guess I’ll know for sure in 5hrs…
I have often heard of people using food dehydrators as the better dryers. They are similar price as something like a Sunlu S2, reach 80°C, are isolated, have convection, are typically quite precise and I think they let the moisture escape.
I’m considering the Graef DA-2042, but I don’t find the space for it.
No idea if that is available outside europe.
Nah, wet filament will have poor bubbly surfaces pretty much everywhere (not just the overhangs). The good surfaces suggest the filament is fine.
The issue here is purely the steep overhang. Don’t hide from supports. Sometimes they are just needed. That’s a very steep overhang. If you think the time penalty is too high, use the support blocker and only support the edges of the blades and let it bridge 5 - 15 mms (as shown below).
I ran it after drying again and got mostly the same results.
I’m running it again w/ support threshold angle at 50° and no special support material, that pretty well covers the whole problem area. Downside is going to be cleaning out the groove and holes.
I’m wondering if I can get my hands on the assembly design and tweak it to be more 3d printer friendly.
I trick I use for almost EVERY supported model is to increase the distance from the support to the model to .5mm - 1.0mm (setting called Support/object xy distance). What this does is, forces the supports to bridge away from the model and keep it from fusing to the side of the model. This allows for MUCH, MUCH easier support removal.
Support/object xy distance
The XY distance between the support and the object, as shown below.
Right click the object and click support blocker. Then choose what shape you want. Size it and place it where you do not want supports. This is much easier than painting if you’re trying to cover a larger area.
**edit, the camera is not really capturing the defects, but they are pretty bad right where the overhang is the largest.
Getting better, but still not perfect like it needs to be.
To recap, this is PAHT-CF w/ stock settings for the filament.
100% infil, 6 wall loops,
Supports are on. Support material is the PAHT-CF, Threshold angle is 60°, Top Z dist is .08mm as it XY distance. Top layers for support is 6 under the theory that I don’t want to be taking this stuff off in little pieces.
I tried 1mm as suggested for the XY and got a huge gap between the part and support in preview.
The support barely come off, and when it does it looks nice except where it leaves this really rough part. I don’t understand why it’s doing that, the part is fully supported where it’s f’ing up.
I’m re-running the part w/ PA support material in the interface layer just to see if this is even possible. Problem here is it more than doubles the print time and material cost and then the part is too expensive. Somehow I need to get it to run from a single material and then I’d be in the gravy.
I think you are getting there. The difference from start till now, looks significant. I would try a couple of things to improve.
First, I’d test in PLA. Find you support quality settings in PLA and then transfer them to the PAHT CF when you are fairly well honed in.
Second, it sounds like the main concern is the surface quality. Being a fan blade I can understand. That means using a large Z support distance is a bad idea. However, stick with 0.5 - 0.8 on the XY though. That’ will have nearly no affect on the quality. The Large Z gap is likely what is causing an issue for surface quality.
Too much Z gap will cause the filament to droop (causing surface issues). However it will be easier to remove because it isn’t solidly connected.
Too much XY is only an issue when you have model geometry directly above the gap that needs to be supported. This is pretty rare, because any support needed can usually be bridged by a 1mm or so. That said, I find that 0.5mm gap is all I normally need. The default 0.35 is a little too tight, and can sometimes merge the support and model when I don’t want it to.
The final thing I would suggest, is altering the model by a tick. If you can’t change the angles of the blade, consider making the blade thicker at the bottom so it can be printed without supports.
Like this, (Note, I think I modeled it backwards, but I think you can get the idea)
Probably the best idea is to do this in PLA. Can’t believe I didn’t think of that. Excellent advice.
I think you miss-read my Z gap. It’s .08mm, not .8mm, so less than you are suggesting. I also have XY set at .08mm too, because when I tried large XY gaps, it left a lot of space in the preview tab. I could see a huge gap. I think because it’s not a level surface and is rising more than it’s bridging.
As to changing the model, though I am the one who generated it, I’m not at liberty to do that.
I’m wondering if my high number of wall loops could be partially to blame for this poor surface.
Ah… I did miss understand that. The 0.08mm Z and XY gap is definitely too tight. Open up your XY to 0.5mm, and keep your Z gap to the default (I believe it 0.20mm).
As far as the wall loops, I guess it could be a factor, but I don’t think that is the issue. When I look at how that effects the supported area, only one wall touches the support at any time. I am finding some new ideas that could help though.
Right now, I’m running some simple tests to see what settings makes for the best surface (in one filament) for a filleted surface (similar to your fan blades). Its not a perfect replica, but I think it will give a good understanding of the best thing to do.
Ok, my best results were with the following settings.
Style - Snug
Threshold angle - 35 degrees
Top Z-Distance - 0.10mm (but the default 0.20mm was very close)
Interface pattern - Concentric
XY distance - 0.50mm
In my tests, they were all very similar, but this was the best. I don’t think it is an absolutely great finish, but I believe it’s about as good as it greats. I think if you use a 0.12 layer height the finish will be better as well, for reference this was 0.28. NOTE: I used ASA to represent PAHT CF. This only recreates the low fan usage and high bed temps. The speeds are different and likely will result in a slight better finish on the PAHT CF.
Well, my experiment using ‘Support for PA/PET’ failed miserably. It won’t come off at all. I had better luck on another item using ABS as the support interface than this stuff. Something is wrong with this combination.
Yes, I used zero for both Z and X/Y, but the material was stuck hard even on Z. Just wouldn’t come off.
I’ve messed with this enough for a few days. I’ve got 5 (4x at 6hrs and one at 13hrs) of these failed turbines and need to do some production w/ the printers for a bit until I can catch up. Then I’ll mess with the turbine again.
I’m probably not going to do another with support material. Even if I use ABS which worked much better on another part I did (reversed, part was ABS and support was PA6-CF, but it came right off.) It just kills the print time and I’m not going to be able to get them to bite on the price. I think either I figure out a way to print it w/ 1 material or I have to no-quote the job.
